Industrial Product-Service Systems Industrial Product

0 downloads 0 Views 20MB Size Report
Apr 2, 2009 - for their meticulous preparation for the conference. ...... using 3D solid and realistic product data that make ...... mapping tools as well as database solutions (i.e. SAP ...... open, flexible and agile environment, by extending the.
CIRP – IPS2 2009

The following topics are covered: • PSS design: methodologies and challenges

Most western manufacturing companies are shifting their focus of business strategy towards selling services or functionality instead of products. The product-service system (PSS) strategy has a great impact on customers, product life cycle and company strategy. The design of PSS is a complex problem, and must meet the challenges of the changing financial and resource models that align with PSS strategy.

• PSS requirements engineering and management

New developments are taking place in industrial markets in the form of Industrial Product-Service Systems (IPS2). IPS2 represents a change in the competitive strategy for manufacturing companies, enabling innovative function, availability and result oriented business models.

• PSS evaluation techniques

The proceedings present multidisciplinary research encompassing concepts, methodologies and infrastructure development for successful IPS2.

• Service and supporting network in the PSS environment

Professor Rajkumar Roy

Dr. Essam Shehab

Rajkumar Roy is Professor of Competitive Design and Head of the Decision Engineering Centre at Cranfield University. He is also the President of the Association of Cost Engineers. His research interests include design optimisation and cost engineering for products, services and industrial product-service systems.

Essam Shehab is a Senior Lecturer in Decision Engineering at Cranfield University. His research and industrial interests cover multi-disciplinary areas including design engineering, cost modelling and knowledge management for innovative products and industrial product-service systems.

• Product, service and PSS knowledge representation • Service knowledge capture and reuse in PSS design • PSS information and knowledge management • Digital product life cycle systems for PSS • Service engineering • PSS life cycle management • Review of PSS design approaches • Impact of in-service and disposal issues on design • Impact of informated product in use on design • PSS: business requirements and techniques • Organisational complexity in the PSS model • Life cycle cost modelling for PSS

• PSS: strategy and transition

Industrial Product-Service Systems

CIRP – IPS2 2009

Industrial ProductService Systems (IPS2) Proceedings of the

1st CIRP IPS2 Conference

Rajkumar Roy Front cover car manufacturing image courtesy of Toyota

Essam Shehab

Rajkumar Roy, Essam Shehab Editors

Industrial Product-Service Systems (IPS2) Proceedings of the

1st CIRP IPS2 Conference

Rajkumar Roy, Essam Shehab Editors

Editors Professor Rajkumar Roy, Dr. Essam Shehab Cranfield University Cranfield Bedford MK43 0AL UK

ISBN 978-0-9557436-5-8 Cranfield University Press © Cranfield University 2009 All rights reserved. No part of this publication may be reproduced without the written permission of the copyright owner.

1st CIRP IPS2 Conference 2009

Industrial Product-Service Systems (IPS2) 01-02 April 2009, Cranfield University, UK

Organised by Cranfield University, UK

Sponsored by

S4T S4T - Support Service Solutions: Strategy and Transition

Conference Chairman R. Roy, Cranfield University, UK International Scientific Committee CIRP J Aurich, Germany D Brissaud, France P Gu, USA M Z Hauschild, Denmark S Kara, Australia F L Krause, Germany H Meier, Germany L Monostori, Hungary G Moroni, Italy A Nee, Singapore M Rese, Germany G Schuh, Germany G Seliger, Germany W Sihn, Germany D Spath, Germany S Takata, Japan T Tomiyama, The Netherlands E Uhlmann, Germany Non CIRP T Baines, UK S Evans, UK M Henshaw, UK L Leifer, USA N Morelli, Denmark A Neely, UK A Tukker, The Netherlands Y S Kim, Korea

Local Organising Committee A Al-Ashaab P Baguley R Barrett D Baxter E Benkhelifa S Bolton P Datta I Ferris M Goatman M Grant H Hassan J Mehnen (Organisation Chair) D. Saxena E Shehab (Programme Chair) A Tiwari (Finance Chair) B. Tjahjono Y. Xu T Bandee, (Organisation Secretary) E Pennetta L Brady

Foreword Most western manufacturing companies are shifting the focus of their business strategy towards selling services or functionality instead of products. The product-service systems (PSS) strategy has a great impact on customers, product life cycle and company strategy. The design of PSS is a complex problem, and must meet the challenges of the changing financial and resource models that align with PSS strategy. A better understanding of knowledge resources is needed in order to meet these challenges. This business strategy also has significant impact on reducing material consumption and thus minimising environmental impact. New developments are taking place in industrial markets, in the form of Industrial Product-Service Systems (IPS2). An IPS2 is defined as “an integrated industrial product and service offering that delivers value in use”. This integrated understanding leads to new, user-centric solutions. IPS2 represents a change in the competitive strategy for manufacturing companies, enabling innovative function, availability and results-oriented business models. The CIRP IPS2 Conference is lead by the CIRP IPS2 Working Group. This is a multidisciplinary working group that is supported by STC A, STC Dn and STC O within the CIRP. This 1st CIRP IPS2 Conference has 51 technical papers in the proceedings from 17 countries. This shows the popularity of the topic within CIRP and the outside research community at large. The conference has 9 technical sessions, 2 keynotes and one invited presentation. In addition the second day of the conference includes an industry panel discussion and industry visits. Over 75 participants are expected to attend the conference. This IPS2 conference will focus on research into design, cost of IPS2, novel business models, informatics, service networks, industrial and academic experience with IPS2 and integration of the various business components that support the PSS: marketing, design, manufacturing, logistics and maintenance. I would like to take this opportunity to thank all the authors for their quality research, industry panel members for their contributions, the international scientific committee members for their support in reviewing the papers and the local organising committee for their meticulous preparation for the conference. I would like to specially thank Dr. Jorn Mehnen, Dr. Essam Shehab, Dr. Ashutosh Tiwari and Mrs Teresa Bandee for their significant contributions towards the success of the conference. I would also like to thank our sponsor Mori Seiki,- the machine tool company, BAE Systems, S4T project and the exhibitors for their support for the conference.

Professor Rajkumar Roy Chairman CIRP IPS2 Conference, 2009

Table of Contents

Keynote Paper

Informatics

Service Engineering as an Approach to Designing Industrial Product Service Systems G. Schuh, G. Gudergan………………………………1

Comprehensive Complexity-Based Failure Modelling for Maintainability and Serviceability K. T. Meselhy, H. A. ElMaraghy, W. H. ElMaraghy…………………………………….89

Design Multi Operator BTS Aesthetic Tower Design for Metropolitan City A. Windharto, A. Setiawan…………………………8 Continuous Improvement of Industrial ProductService Systems E. Schweitzer, C.Mannweiler, J.C.Aurich..............16 Service Development and Implementation - A Review of the State of the Art M. Torney, K. Kuntzky, C. Herrmann…………...…24 A Product-Service System Representation and its Application in a Concept Design Scenario Y.S. Kim, E. Wang, S. W. Lee, Y.C. Cho …………32 Empirical Study Concerning Industrial Services within the Austrian Machinery & Plant Engineering Industry K. Matyas, A. Rosteck, W. Sihn…………………....40 Strategies for Designing and Developing Services for Manufacturing Firms A. R. Tan, D. Matzen, T. McAloone, S. Evans……46

Informatics-Based Products-Service Systems for Point-of-care Devices O. Ajai, A. Tiwari, J.R.Alcock……………………....94 Service Information in the Provision of Support Service Solutions: A State of the Art Review S. Kundu, A. McKay, R. Cuthbert, D. McFarlane, D. Saxena, A. Tiwari, P. Johnson………………...…100 An Infodynamic Engine Approach to Improving the Efficiency of Information flow in a Product-Service System C. Durugbo, A. Tiwari, J.R. Alcock……...……….107 A Periodicity Metric for Assessing Maintenance Strategies K. T. Meselhy, W. ElMaraghy, H. A. ElMaraghy……………………………………113 Development of an Extended Product Lifecycle Management through Service Orientated Architecture J. Cassina, A. Cannata, M. Taisch......................118

Business Models

A Method of Supporting Conflict Resolution for Designing Services Y. Akiyama, Y. Shimomura, T. Arai……………......54

Multimodal User Support in IPS2 Business Model R. Gegusch, C. Geisert, B. Hoege, C. Stelzer, M. Roetting, G. Seliger, E. Uhlmann…………...……125

Product-Service Systems - From Customer Needs to Requirements in Early Development Phases A. Ericson, P. Muller, T. Larsson, R. Stark………..62

Framework for the Integration of Service and Technology Strategies E. Juhling, M. Torney, C. Herrmann, K. Droder...132

Evaluation of ‘Design Loops‘ to Support the Design of Product Service Systems: A Case Study of a Helium Liquefier N. Maussang, P. Zwolinski, D. Brissaud................68 Product/Service Systems Experiences - an International Survey of Swedish, Japanese, Italian & German Manufacturing Companies M. Lindahl, T. Sakao, E. Sundin, Y. Shimomura…………………...…………………...74 Service and Manufacturing Knowledge in ProductService Systems: a Case Study N. Doultsinou, D. Baxter, R. Roy, J. Gao, A. Mann……………...……………………………….82

Strategy Assessment and Decision based Implications for Integrated Product-ServiceSuppliers R. Schmitt, S. Hatfield……………………………..140 A Framework for Cross Disciplinary Efforts in Services Research P. J. Wild, P. J. Clarkson, D. C. McFarlane….....145 New Models for Sustainable Fashion Industry System: A Case Study about Fashion Net Factories P. Ranzo, M.A. Sbordone, R. Veneziano……….153 Product Service Value Analysis: Two Complimentary Points of View T. Alix, Y. Ducq, B. Vallespir………………...……157

Business Implications of Integrated Product & Service Offerings M. Lindahl, T. Sakao, A. Ohrwall Ronnback…….165

Sensitivity Cost-Benefit Analysis to Support Knowledge Capture of Industrial Interests A.M. Paci, M.S. Chiacchio………………..………261

Innovative Service-Based Business Concepts for the Machine Tool Building Industry S. Biege, G. Copani, G. Lay, S. Marvulli, M. Schroter………………………………………….173

Whole Life Cycle

A Method to Analyse PSS from the Viewpoints of Function, Service Activity & Product Behaviour T. Hara, T. Arai, Y. Shimomura…………………..180

Metadata Reference Model for IPS2 Lifecycle Management M. Abramovici, M. Neubach, M. Schulze, C. Spura………………………………………...…..268

Use-orientated Business Models and Flexibility in Industrial Product-Service Systems A. Richter, T. Sadek, M. Steven, E. G. Welp…....186

Remanufacturing on a Framework for Integrated Technology and Product-System Lifecycle Management (ITPSLM) A. G. Filho, D. A. Pigosso, A. R. Ometto, H. Rozenfeld………………………………………..273

Analysis of Integrated Product and Service Offerings from Current Perspectives of Providers and Customers T. Sakao, E. Sundin………………………………..193

Commercializing Sustainable Innovations in the Market through Entrepreneurship D. Keskin, H. Brezet, J. C. Diehl……………...….280

Cost Engineering Uncertainty Challenges in Service Cost Estimation for Product- Service Systems in the Aerospace and Defence Industries J. Erkoyuncu, R. Roy, E. Shehab, P. Wardle…...200 Identifying Risk and its Impact on Contracting through a Benefit Based-Model Framework in Business to Business Contracting: Case of the Defence Industry I. Ng, N. Yip……………………………………...….207 Cost Modelling Techniques for Availability Type Service Support Contracts: a Literature Review and Empirical Study P. P. Datta, R. Roy…………………………………216 Cost Evaluation Method for Service Design Based on Activity Based Costing K. Kimita, T. Hara, Y. Shimomura, T. Arai……….224 Affordability Assessment of Industrial ProductService System in the Aerospace Defence Industry O. Bankole, R. Roy, E. Shehab, P. Wardle……..230 An Aerospace Component Cost Modelling Study for Value Driven Design J.M.W. Cheung, J. P. Scanlan, S. S. Wiseall…...238 Profitability of Industrial Product Service Systems (IPS2) – Estimating Price Floor and Price Ceiling of Innovative Problem Solutions M. Steven, M. Rese, T. Soth, W. Strotmann, M. Karger……………………………………..…….243 Life Cycle Cost-Orientated Service Models for Tool and Die Companies G. Schuh, W. Boos, S. Kozielski ...………………249 Obsolescence Challenges for Product-Service Systems in Aerospace and Defence Industry F. Romero Rojo, R. Roy, E. Shehab, P. Wardle..255

Environmental Impacts of Rental Service with Reconditioning - A Case Study R. Khumboon, S. Kara, S. Manmek, B. Kayis…..288 The Practical Challenges of Servitized Manufacture T. Baines, H. Lightfoot……………...……………..294 Challenges for Industrial Product/Service Systems: Experiences from a learning network of large companies E. Sundin, G. O. Sandstrom, M. Lindahl, A. Ohrwall Ronnback, T. Sakao, T. C. Larsson……...……...298

Service Network Dynamic IPS2-Networks and Operations Based on Software Agents H. Meier, E. Uhlmann, C. M. Krug, O. Volker, C. Geisert, C. Stelzer………...……………………….305 Standardization of Service Delivery in Industrial Product-Service Systems H. Meier, C. M. Krug…………………………...….311 Engineering Network Configuration: Transition from Products to Services Y. Zhang, J. Srai, M. Gregory, A. Iakovaki……...315 Roadmap to Self-Serving Assets in Civil Aerospace A. Brintrup, D. C. Ranasinghe, S. Kwan, A. Parlikad, K. Owens..............................................323 The Chinese Service Industry as a Challenge for European SME: A Systematic Approach for Market Entry R. Schmitt, S. Schumacher, C. Scharrenberg.....331 Industrial Services Reference Model M. Gerosa, M. Taisch……………………...……...336

Service Engineering as an Approach to Designing Industrial Product Service Systems G. Schuh1, G. Gudergan2 Director, FIR at Aachen University, Aachen, Germany, [email protected], 2 Department Service Management, FIR at Aachen University, Germany, [email protected] 1

Abstract Unique customer solutions which integrate products and services into a high value offering have the potential to successfully differentiate from competition even prices are dictating product markets. However, companies face tremendous challenges to develop customer solutions. Service engineering is considered to be the scientific discipline which supports the design task of intangible offerings and thus a foundation for solution design. We enhance the existing body of research in service engineering by proposing to apply the systematic approach of service engineering for solution design. An architecture for services design is introduced as an initial starting point to designing service based solutions. Keywords: Industrial Product Service Systems, Solution Systems, Customer Value, Design Framework, Service Engineering, Service Engineering Architecture

1 INTRODUCTION Providing business related services more and more means to solve a customer problem and deliver an individualized solution that is able to substitute a customer internal process or function rather then just to deliver a single service in a single transaction. For example, the automotive industry requires pre-production services (such as design services and research and development), production-related services (such as maintenance and IT services), after-production services (transport and distribution services) and financial services and finally other business services such as accounting or legal services. In business to business settings of producing companies, these services are usually bundled into an integrated offering which is configured by different tangibles such as capital goods, spare parts and intangibles such as repair services, remote services, joint project management and others [15]. It has been well realized that this integration of high quality services, business related services in particular, is crucial for the competitiveness of existing and future economies. Thus, producing companies increasingly link products, parts, after sales services and valued added services such as training, business consulting and engineering services into a integrated solution system to successfully differentiate from worldwide competition [12]. The underlying strategy in industrial markets is to substitute the subsequent and single offerings by integrated value adding solutions which lead to lasting relationships to closely link providers and customers. These often are characterized by collaborative engineering efforts and even link providers and customers on an emotional level. Belz has first introduced the term solution system to describe the integrative character of the solution delivered [2]. Companies in the future have to develop and establish solution systems to generate superior value to the customer [14]. The corresponding concept is illustrated in the following picture.

CIRP IPS2 Conference 2009

Joint Market ing, Customers partner visits, ... Operat ions and productivity service s, ... Joint D evelopment, Individualizatio n .. . Se rvice, Consu lting, Training, ... Pe rip hery, Parts, . ..

Custom er needs and Va lue

C ustomer B enefits

Modular Ext ension EDM 2000

Product Productsystem Parts, assortment Services Integr ation of solution Integr ated project manag em ent

Emotional relationship, Trust, Image

Figure 1: Solution system to deliver value to the customer (Source: [2], [14]). The transformation towards a solution provider however has tremendous impact on the whole company. It is not only important to formulate the appropriate strategy including for successful differentiation, it is the integration of all relevant company activities which has to be achieved: strategy, product definition, marketing concept and the solution design process itself have to aligned and inherently linked. In addition, all organisational structures and the company culture and employee behaviour have to be changed towards a more customer and solution orientated characteristics. E.g. there is a need for decentralised structures which concentrate the relevant competencies where they are needed near the customer. Figure 2 illustrates the integration needs and direction as mentioned for four important company activities: differentiation strategy, solution concept and configuration, solution marketing and communication and finally the solution design activity. The integration as illustrated means that all of these activities have to be changed simultaneously towards a solution, customer needs supporting and value driven orientation. This simultaneous shift is the prerequisite to successfully implement a solution orientation within a producing company. An unbalanced change in organisational transformation processes will cause tension and finally the fail of the initiative [3].

1

Value oriented

Differentiation Strategy

“Customer value is a customer’s perceived preference for and evaluation of those product attributes, attribute performances, and consequences arising from use that facilitate (or block) achieving the customer’s goals and purposes in use situation” [18]. This definition emphasizes the customer perspective of value. It incorporates both desired and perceived value and emphasizes that value originates from customers’ perceptions, preferences, and evaluation. It also links together products or services with use situations and related consequences. Customer value can be classified in several ways [18]. One possible classification suggests to specify types of value regarding to their contexts within a customer’s evaluation process and distinguishes product value, value in use, possession value, and overall value [18]. Value in use, for example, reflects the use of a product or service in order to achieve a certain goal or set of goals. Hassle free supplier relationships or a proactive services are examples for value in use. Possession value reflects the inherent meaning of the product or service to the customer. For example, value to an industrial customer may be resulting from the rate of return or cost reduction earned on the purchase of a new piece of equipment or on the use of an industrial service. If the cost reduction or revenue enhancements generated by the product or service purchase justify the price, value has been created. This purchase process can be objectively valued. In the case of value in use, this process is subjective, but benefits and costs are still compared so that in industrial settings value for the customer often means the difference between the benefits customers realize from using a product and the costs they incur in finding, acquiring and using it. If the benefits exceed the costs, then a customer will at least consider purchasing a product or service. To increase the understanding of the term customer value the model of three hierarchical levels of value as illustrated in Figure 3 serves as a useful explanation [see also 18].

Solution Configuration Customer needs supporting

Productoriented Product, Price, Place, Promotion

ProductSupporting Inside-Out

Producer Solution Provider „4P“ + Process Solution Communication and People

Solution Design

Customer Value driven

Figure 2: Required integration and orientation of company activities for successful transformation. We here concentrate on the methodological foundation of the solution design process. Designing solution systems as illustrated in Figure 1 is a challenging task. There are challenges to facilitate the rich interactions and cohesion between the different services or solutions and the customers. There are challenges to ensure flexibility and reconfigurability of services and solutions in processes and structures [8]. Unfortunately, managers of service organizations are facing tremendous difficulties in meeting these challenges. The high degree of integration and synchronization needed in services and solutions causes complexity which is not understood. Neither within the structure of service based solutions nor in implementing new service processes [9]. There is first a need for a comprehensive understanding of the nature of solution systems and second a need for systematic design processes. Otherwise, it will not be possible to properly handle the complexity in today’s and future service based solutions and relationships. 2 CUSTOMER VALUE PERSPECTIVE ON INTEGRATIVE SOLUTIONS Value is an important concept in the management literature. The term value is used in several very different contexts. From the perspective of managing an organization, creating and delivering superior customer value to high-value customers is considered to be an important means in order to increase the value of an organization [18]. From a different point of view, the term customer value takes the perspective of an organization’s customers. This perspective considers what customers want and what they believe that they get from buying and using an organizations product or service offering. This perspective is central to the resource based view of strategic management, which considers value to the customer to be the dominate prerequisite to produce a sustainable competitive advantage based on the companies resources and competencies. Only if resources and competencies are used to deliver a solution which is valuable to the customer, these resources and competences can be considered to be of significant relevance for a companies competitive position. The customer value perspective is coherent with the perspective applied in this paper and existing definitions integrative offerings of products and services such as the IPS² concept (IPS² Industrial Product Service Systems). In this context, an IPS² are understood as integrated product and service offering that delivers values in different a use and application contexts [see also 13]. Value is the underlying concept of solution systems as illustrated in Figure 1. There exists a brought variety of divers definitions of the term customer value. Customer value can be defined as follows:

Customer‘s Values System Higher order goals and believes

Consequences

Attributes

Figure 3: Customer values hierarchy (based on [18]). The customer value hierarchy as depicted suggests that customers experience value at different levels when they expect a desired value and when they perceive value as well. This hierarchical structuring is important to systematically designing solution systems as illustrated in Figure 1: The structuring i.e. allows to specify requirements for the different elements (tangible or intangible elements) of the solution system in a hierarchical manner and thus allows the application of systematic design approaches as illustrated in this paper. 3 SERVICE ENGINEERING ARCHITECTURE AS A FRAMEWORK FOR SYSTEMATIC SOLUTION DESIGN Handling complexity in solution design requires for frameworks and methods which help to systematize and structure complex tasks into pieces which can be overseen and handled properly. In the following, service engineering is considered to be the scientific discipline

2

order to achieve the development result represented in the SDRDM. To keep the complexity of a development project as low as possible, it is not useful to construct the service in detail from the start. Instead, the development can be stated in such a way, that first the requirements for the service system are implemented in a general concept. Afterwards, the general concept can be divided into components. The determined characteristics of the general concept result in requirements for those components. Each component can then be considered independently. This procedure of specifying concepts into partial concepts and their subsequent configuration can be continued at all levels of detail in the same way. An appropriate method to detail a service system is the Function Tree Analysis under consideration of Suh’s axiomatic design. Suh states that one can only detail a function tree with the embodying concept in mind [1], [13]. Based on the essential characteristics of professional services the architecture itself is divided into three partial models with regard to the characteristic elements of services: results, processes and resources. The partial models are intimately connected in the sense of meansend relationships. Since results are generated by a set of processes, which still has to be specified, a determined service result implies requirements for the service processes. Hence, service processes are means, which generate predetermined results. The processes in turn necessitate resources for their implementation. For this reason processes and resources represent a means end relationship. Therefore, a complete service concept always contains a result concept, a process concept and a resources concept.

and a foundation to solution design. An architecture for services design is introduced here as an initial starting point to designing service based solutions. This architecture as illustrated comprises steps for successful design and development of Services and has been introduced by Gill in 2003 [7]. The term Service Engineering becomes more and more prominent in the scientific literature as the discipline covering the development and design of new services Service Engineering can be further defined as the engineering discipline which covers the systematic design of services. Service Engineering covers the following perspectives [10]. The architecture of service engineering as illustrated in the following picture structures the overall service engineering task while linking tasks with the methods and tools required performing the tasks [7]. Servi ce Development Process Model (S DPM)

Service Development Result Description Model ( SDRDM)

Servic e Developm ent Met hods (S DMe) Service Development Tools (SDTo)

Development Step

Methods Tools

Development Step Result

Servi ce Developm ent Management Model (SDMM)

Service Results

Service Processes

S ervice Resources/Skills

To tal System Sub systems … Components

Figure 4: Architecture for service engineering: essential components (Source: [7]). The architecture as shown in Figure 4 consists of five essential components for designing and developing business related services: The Service Development Process Model (SDPM) comprises development steps that are necessary to determine requirements and to form the functions and processes that fulfil these requirements. This model also contains steps to identify the skills and resources that are essential to perform these processes professionally. The steps included in the SDPM will be described in detail in the following sections. The architecture component Service Development Methods (SDMe) comprises methods that enable a systematic approach to the development targets. Which methods are suited to support the design and development will also be shown in depth in the subsequent sections. The architecture component Service Development Tools (SDTo) contains only tools that directly support distinct methods. In the understanding of this architecture, the tools of the SDTo operationalize the methods of the SDMe. The Service Development Result Description Model (SDRDM) documents the specific outcome of design and development steps as well as of the service work itself. Thereby, this model builds a common understanding among the design and development team members at the same time. The SDRDM combines functional and graphical aspects of the representation of development results. The Service Development Management Model (SDMM) integrates the four other components. The SDMM connects the development steps of the SDPM with the methods and tools of the SDMe and SDTo respectively in

3.1 The result branch of the service engineering architecture This partial model of the architecture comprises activities to incorporate the external requirements of customers as well as the internal requirements; to check their plausibility, to prioritize and to detail them. One example for steps undertaken in the result branch and the corresponding methods are illustrated in the following picture. Advanced Sequential Inc ident Tec hnique (SIT)

Adv. SIT-Structure / Information Sources List

Customer Requirements

Gathering Internal Requirements Plausibility Check of Requirements

Advanced Sequential Inc ident Tec hnique (SIT)

Adv. SIT-Structure / Information Sources List

Internal Requirements

Qualitat ive Interdependence Analy sis

L-Matrix

Service Requirements

Pr o i ritization of the Requirements

Pair wise Compar ison

Comparisons Scheme

Prioritized Service Requirements

Concretion of Requirements

Pr ogressive Abst raction

Abstraction Sc heme

Precise Ser vice Requirements

Plausibility Check of Pr ecise Requirements

Qualitat ive Interdependence Analy sis

V-Matrix

Consistent Service Requirements

Benc hmarking of Requirements

Advanc ed Competitive Pr oduct Analysis

Assessment Scheme / Information Sources List

Assessed Service Requirements

Att ributes:

Working Customer rece ptio n

Customer decisi on

Easy to find Permanent open ing ho urs

Rel axi ng Pr oduct informatio n

Cleanliness Data security

Pa yment

“Easy cash ”

Custom er disbandi ng

A BC E xit

ABC Entran ce

Gathering Customer Requirements

Directi on sign Baggage claim

Figure 5: Result branch of the service engineering architecture. The first step on this level for example is the investigation into the customer and company requirements, it is recommended to employ the Advanced Sequential Incident Method [11]. In this method, individual process steps are identified along the chronological course of the service creation on a level, at which customers and suppliers have direct contact. In the following development step “plausibility analysis of the service requirements”, requirements from the perspective of customers and the company are brought together and

3

covered by existing task” respectively. In he following development step “transfer of service tasks into service delivery processes”, those primary service tasks, which are necessary to fulfil the customer requirements, are further detailed by a Process Modelling Method. As a supporting tool for this, the Service Blueprinting of Shostack introduced in 1984 [6] has been identified. The Service Blueprinting is a flow chart particularly for the service delivery process, which distinguishes several ways of customer interaction and visually separates them by so called lines-of-visibility. The customer section contains only processes the customer is directly involved in. The onstage processes are visible to the customers, but they do not take an active part in it. The third section of the process flow chart comprises the backstage activities that are entirely performed by the employees without any contact to the customer. With this differentiation the service delivery processes can be adjusted with respect to performance, robustness and reproducibility. For a detailed analysis of potential risks associated with service delivery processes, the application of the ServiceFMEA (Failure Mode and Effects Analysis) is implemented into the architecture. Using the ServiceFMEA, first, potential failures linked to the process steps are determined and rated on a 1 - 10 scale with respect to their severity (S) and their detectability (d) [6]. For processes with direct customer interaction as ascertained in the Service Blueprinting, the detectability is irrelevant since there is no chance to prevent the customer from experiencing the failure. Afterwards, the causes of each potential failure need to be discovered and evaluated with regard to their probability of occurrence on a 1 - 10 scale as well. Subsequently, these three values of severity, occurrence and detectability, if applicable, are multiplied. The result is the so called Risk Priority Number (RPN), which identifies the greatest areas of concern and indicates what kind of corrective actions should be taken. Particularly, preventive measures can be taken, which helps to avoid cost intensive failures before they might occur. Once the development steps for all identified primary service tasks have been undertaken, the development of the service delivery concept is complete.

analyzed with respect to their plausibility. The Qualitative Interdependence Analysis is employed to show the mutual dependence between requirements, which are regarded as coequal by analyzing the reactions of the elements to changes in one element [4]. For this purpose, the requirements for the service from the perspective of the customer are confronted and compared with those from the perspective of the company in a matrix. Criteria for the Qualitative Interdependence Analysis are “targetneutrality”, “target-harmony” and “target-conflict”. The results of this development step are consistent service requirements from the perspective of customers and the company. As a next step, the service requirements are prioritized from the customer perspective with respect to their impact on the success of the service. The Pair wise Comparison has been identified as a suitable method for this prioritization [6]. In the development step “concretion of the service requirements” the method of Progressive Abstraction is used in the architecture. With the Progressive Abstraction the requirements in terms of their benefit of use are edited, and the levels of measures are revealed which contribute to a large extent to the achieved objectives of the development. 3.2 The service process branch of the architecture Starting from the service requirements, the respective tasks are identified and defined. The leading question for this task can be formulated as follows: “How can the individual service requirements be implemented?” After having found implementation methods for each requirement, the requirements are summarized hierarchically with the help of Transfer Graphs as a tool of the Affinity Method [11]. The results of using this method are hierarchically structured service tasks, which are deduced from the requirements. One example for methods in the process branch of the architecture is illustrated in the following picture. In the next step, the service tasks have to be analyzed with respect to their type. By allocating the service tasks to the types “overall task, “primary task” and “secondary task” distinctions can be made. Mapping of Service Functions and Requirements

Affinity Method

Transfer Graph

Service Functions

Anal ysi s of Func tion Types

Affinity Method

Service Types

Typed Service F unctions

Target / Actual Comparis on of t he F unction Portfolio

Qualitative Interdependence Analysis

Function Target / Actual Comparison Matrix

Bala nc e of Functions

Transfer of Functions into Serv ice Processes

Process Modeling Method

Symb ol List

Process Fl ow Chart of th e Service Del vi ery

Qualit y Anal ysis of Selected Proc ess es

Ser vice-F MEA

Service-FMEA Scheme

Measures to be Taken

Over all servi ce task

Pri mary service task

Provi si on of worki ng space work ing

Tr anslation servi ce

3.3 The service skills and resources branch of the architecture This partial model of the architecture helps to develop a concept for the essential service resources. The skills, which are necessary to perform the identified service tasks and service processes, are identified first with the help of’ the Affinity Method and hierarchically structured by means of a Transfer Graph. The result of this development step is a target skills profile, which should be understood as the sum of skills necessary for delivering the service. A part of the the skills and resource branch of the architecture is illustrated in the following picture. Afterwards, the individual skills are analyzed regarding to their type: professional competence, social competence, personality competence and method competence can be distinguished. Besides the allocation of the identified skills to these types, a qualitative evaluation with regard to the marks “no competence necessary”, “basic understanding necessary”, “first practical experience and advanced understanding necessary” as well as “management, practical experience and distinct understanding necessary”. In order to benefit from synergies a target/actual comparison should be conducted with the skills, which are already available throughout the

Secondar y s ervic e task

Provi sion of • Telephone • Fax • Internet • E-Mail Access • PC • ...

IT-Support

Figure 6: Process branch of the service engineering architecture. The overall task shall be defined as to meet a maximum amount of service requirements the customer is willing to pay for. The primary task fulfils at least one service requirement and can also be priced. Although a secondary task must also fulfil at least one requirement functionally, the customer is often not willing to pay for that. In order to benefit from synergies, an alignment of service tasks, which are already implemented, and the service to be developed need to be conducted. For this purpose, the Interdependence Analysis is again a suitable method. Therefore, all primary service tasks should be evaluated by an ordinal rating scale, which distinguishes target is covered by existing task” and “target is not

4

In the following, we illustrate how the framework and architecture as presented can be used to systematically develop solution systems. We here take an example from the capital goods industries which is a company delivering assembling system as turn key solutions and offering the operation as well. The company designs and produces complex assembling systems, i. e. for the automotive industry. The unique capability of the company is to design the assembling systems based on a physical model or digital model of the part to assemble. The company then fully integrates the assembling systems into the customer’s production processes. The company offers leasing arrangements for their solutions and different service contracts including the operation of the assembling system at the customer’s site. Challenges the company has to overcome are illustrated in picture 8.

company and the determined skill profile. A suitable method for this is again the Interdependence Analysis.

Sk ills Tar get / Actual Comparison Mat rix

Balance of Sk ills

Identification of Required R esour ces

Affinity Method

Transfer Graph

Profile of Requir ed Res ources

Target / Actual Comparison of t he Resourc e Profiles

Qualitative Interdependence Analysis

Resources T arget / Ac tual Comparis on Matrix

Balance of Resourc es

s kil s l

S o c ial

s k ill s

P er s on a l

s ki lls

s kil ls

M e th o dic

P ro f e ss io n al

C om petence

No competence required

Pro duct knowledge

Basi c understandin g required

Pro duct portfoli o

First practical experience and advanced understa ndi ng req uired

IT-Support Resol ving probl ems ...

Com petence rating:

...

... ... ...

Management, practical expe ri ence and distinct un der standing requir ed

Figure 7: Resources branch of the service engineering architecture. Subsequently, the key resources related to the skills for the Service delivery have to be identified with the help of the Affinity Method. It is important to find as many resources as possible, which embody he required skills. A Transfer Graph is again an adequate tool for the structuring. Afterwards, a target/actual comparison is conducted between those resources that are necessary for the Service delivery, and those that are already available throughout the company. Again, an adequate method is the Interdependency Analysis with an ordinal rating scale of “target is covered by existing resources” and “target is not covered by existing resources” respectively. In case of resource coverage or a resource excess, the service which should be developed can be generated with the already available resources of the company. In case of a resource deficit, the corresponding resources have to be obtained. When the development steps for the identified competencies and resources are finished, the development of a potential service provision concept is completed. 4 INTEGRATIVE FRAMEWORK FOR SOLUTION ENGINEERING Integrative industrial solutions such as delivering a comprehensive assembling line are more complex in their nature then single services and thus require an even more structured and systematic approach for their development or engineering. Combining the hierarchical perspective of customer values, the illustrated concept of the solution system and the architecture of service engineering into an integrative framework reveals into the framework illustrated in the following picture. The framework as illustrated follows the basic design principles of structuring and systematizing. The framework is based on the basic assumption, that goals can be broken down into consequences and consequences can be broken down into attributes within the customer’s value system. Customer defined goals, consequences and attributes define how elements in the solution system have to be specified. Based on this specification, each element of the solution system as illustrated in Figure 1 is considered to be designed within a capsulated engineering process: The spare delivery service for example is based on specific resources and processes which guaranty for a specific service level. The repair service is based on specific skills and process which are implemented to fulfil the repair task. However, this does not mean, that elements are not interrelated. The functional interdependencies between these single services are determined by their contribution to the specific customer value at the three hierarchical levels.

Design of solution portfoli o and system

Design of engineering process for hybrid solutions

Design of organizational structure for solution delivery

Design of change management towards solution orientated behavior

Figure 8: Potential to apply service engineering framework and architecture to introduce company transformation. The solution delivered by the company can be best described by the term “assembling capability”. Following the hierarchical concept of customer values as introduced, the consequences for the customer in the specific use situation are that parts of the production system are controlled by the provider and thus efforts and costs can be reduced. Efforts for designing and integrating the assembly system are fully outsourced. Control authority over the assembly system is outsourced as well. Cost Reduction of internal engineering efforts and subsequent costs are anticipated goals at the customer site. The provider guaranties for reliability and hassle free operation and thus satisfies the customer requirements at the attribute level. I de nti fication of R eq uired Skil ls

Affini ty Method

Tr ans fer Graph

Prof ile of Required Skil ls

Analysis of Skil l Types

Affini ty Method

Skill Types

T yped Skills

T arget / Actual Compari son o f t he Skil l Profiles

Quali tati ve Interdepende nce Analysis

Skills Target / Actual Compari son Matrix

Balance of Ski lls

I de nti fication of R eq uired Resou rces

Affini ty Method

Tr ans fer Graph

Prof ile of Required R esources

T arget / Actual Compari son o f t he Res ource Profil es

Quali tati ve Interdepende nce Analysis

Resou rces Target / Balance of Resources Actual Compar ison Matri x

Target of c om pet enc es

Type of S kil l

Ac tua l competences

Prof . skills

Product kno wle dge

Prof . skills

Prod uct por tfolio

Mixed skills

Mul itling ual Prese nt at ion …



P rof. ski lls

P rof. skills

M ixed skill s

Mi xed skil ls

Prof. skills

.. .

Type of Sk ill

.. .

Qualitative Interdependence Analysis

W ork-f low Mgt .

Typed Skills

Target / Actual Comparison of t he Skill Profiles

Resol vi ng p rob lems

Profile of Requir ed Skills

Sk ill Types

Un iligual P re sen tatio n

Transfer Graph

Affinity Method

Prod uct p ort fo lio

Affinity Method

Analysis of Skill Types

Product kno wledge

Identification of Required S kills

.. .

Ca ption : ... ta rget is co vered by existing re source s ... ... ta rget is no t cove red by existing re source s ...

! .. .

. ..

. ..

!!

...

.. .

.. .

Figure 9: Architecture of service engineering to develop a resource profile for solution delivery. One major challenge when designing a solution as described is to exactly specify the behavioural skills needed to successfully implement the solution concept into practice. Employees need very specific skills, in

5

solution design. Implementing this into practice including work and task coordination and information processing within task execution will be the next challenge. Industrial engineering will then complement the existing body of service engineering and will provide - with engineering science, marketing, and organisation design, a future set of disciplines for the multidisciplinary structure of service engineering. 6 ACKNOWLEDGEMENTS This research has been made possible with funding provided by the German Bundesministerium fuer Bildung und Forschung (BMBF), grant number 01FD0674. 7 REFERENCES [1] Akiyama, K. (1991). Function Analysis: Systematic Improvement of Quality Performance. Productivity Press, Cambridge. [2] Belz, Ch. (1997). Leistungssysteme, in: Thexis, Leistungs- und Kundensysteme. Kompetenz für Marketing-Innovationen, Schrift 2, 1997, S. 12 – 39. [3] Bleicher, K. (2004). Das Konzept integriertes Management. Campus Verlag, Frankfurt am Main. [4] Clausing, D (1994), Total Quality Development A Step - By - Step Guide to World-class Concurrent Engineering, New York: ASME. [5] DIN (1990), Ausfalleffektanalyse (FehlerMöglichkeits und Einfluss-Analyse), (Failure Mode and Effects Analysis). Beuth. Berlin. [6] Eversheirn, W. , Kuster, J., and Liestmann, V. (2003), Anwendungspotenziale ingenieurwissenschafflicher Methoden für das Service Engineering. In: H-J Bullinger and A-W, Scheer (Eds), Service Engineering Entwicklung und Gestaltung innovativer Dienstleistungen, Berlin, Heidelberg, New York: Springer, 417-442. [7] Gill, C. (2003). Architektur für das Service Engineering zur Entwicklung von technischen Dienstleistungen. Shaker, Aachen. [8] Gudergan, G. (2008), Erfolg und Wirkungsmodell von Koordinationsinstrumenten für industrielle Dienstleistungen. In: Schriftenreihe Rationalisierung und Humanisierung. 239. Shaker, Aachen. [9] Gudergan, G., H. Luczak (2003), Coordination mechanisms in industrial service organizations. In: Human Factors. in Organizational Design and Management – VII (H. Luczak and K. J. Zink. (Eds.)). [10] Luczak, H., H. Keith and Ch. Gill (2002). Kompetenzentwicklung für das Service Engineering. Proceedings HAB-Jahrestagung 2002. Aachen. [11] Luczak, H., Ch. Gill (2003), Service Engineering Industrieller Dienstleistungen. In: Proceedings of the 7th Southeast Asian Ergonomics and 4th Malaysian Ergonomics Conference (SEAMEC) (Khalid, H.M.; Yong, L.T.; Kion, L.N. (Eds.)), S. 346-353. University Malaysia Sarawak, Kuching, Sarawak, Malaysia. [12] Ramaswamy, R. (1996). Design and Management of Service Processes - Keeping Customers for Life Reading: Addison-Wesley. [13] Baines et al. (2007). State-of-the-art in productservice systems. JEM. Proc. IMechE Vol. 221 Part B: J. Engineering Manufacture [14] Schuh, G., T. Friedli and H. Gebauer (2004), Fit for Service: Industrie als Dienstleister. Carl Hanser Verlag, München, Wien. [15] Schuh, G. and G. Gudergan (2008), Fakten und Trends im Service 2008. Verlag Klinkenberg, Aachen.

particular when the solution as described here is i.e. operated at the customer’s site. Employees then need specific communication or language skills in order to provide a beneficial problem solution to the customer. As illustrated in the following picture, the architecture supports to systematically identify the adequate method to identify a skill profile for the service technician. Another challenge when designing solutions such as the assembling system as described is to design the required flows of activities and communication. Solutions as illustrated often require remote service concepts which require complex interaction and communication flows between the customer’s site and provider’s site. At the provider’s site, processes have to be handled with customer interaction or by the back-office employees. Designing the process and communication structure requires methods and tools which allow structuring and systematic drawing. The following picture illustrates, how the architecture supports to identify the right methods and tools to designing process and communication flows. Map ping of Service Affinity Method Functions and Requir ements

Transfer Gr aph

Service Funct ions

Analys is of Function Types

Affinity Method

Se rvice Types

T yped Serv ice F unctions

Target / Actual C ompar iso n of t he Fun ction Portfolio

Qu alitative Interdepende nce Analysis

Funct ion Target / Actual Balanc e of Functions Compar ison Matr ix

Trans fer of Functions in to Ser vice Process es

Proce ss Modeling Method

Symbol List

Pr ocess Flow Chart of the Service Delivery

Quality Analys is of Select ed Proces se s

Service- FMEA

Se rvice-FMEA Scheme

Measur es to be Take n

Customer Custo mer Entry ( AB C) Line of inte rac tion

Customer order C ustome r asks for translatio n servi ce at the information

Fron t O ffice (Onstage) Order con firmation

Line of visibility

Yes

ABC-employee proves a vailability

Back Office (Backstage)

No

Trans lation service avai lab le?

F

Figure 7: Application of service engineering architecture to identify methods and tools to designing process and communication flows. Both examples for application of the service engineering framework and architecture demonstrate that both, the framework and architecture, can support the engineering of complex solution systems. The main contribution is to reduce the complexity in engineering complex solutions as the architecture supports structuring the associated planning and design steps for the single components which are put together into the overall solution after their design. In addition, the architecture contributes with the suitable methods and tools to design the single services of the overall solution system. The architecture of service engineering as described provides a rich and comprehensive set of methods and tools to develop new solutions. The architecture is based on findings and research in the area of business related services which are provided to solve an often complex and comprehensive customer problem with an adequate service based solution. Planning and conception of new services is supported by the architecture in a structured and systematic fashion. 5 SUMMARY The aim of this article is to enhance the existing scope of the discipline of service engineering and science. Service Engineering is considered to originate from engineering and design theory and the discipline of Service Engineering provides processes and an architecture for the systematic planning of new Customer solutions in an business to business context. The architecture for service engineering introduced provides the methodological framework for successful

6

[16] Schütze, A.,(2001). Ansatz zur prozessorientierten Planung Industrieller Dienstleistungen, Dissertation Dortmund. [17] Suh, N. P. (1990), The Principles of Design. Oxford University Press, New York.

[18] Woodruff, R.B. (1997). Customer Value, The Next Source for Competitive Advantage, in: Journal of the Academy of Marketing Science, 25(1997)2, S.139153.

7

Design

Multi Operator BTS Aesthetic Tower Design for Metropolitan City

A. Windharto, A. Setiawan ITSDC (Institute Technology Sepuluh nopember Design Center) DESIGNCENTER building 2nd floor, Jl. Teknik Sipil no, 1 Kampus ITS, Surabaya 60111, Indonesia, South East Asia [email protected], [email protected]

Abstract The growth of wireless and mobile technology stimulate development of telecommunication infrastructure, Placement of BTS (Base Transceivers Station) tower in the city cause environmental problems by inadequate space while operators install their own Tower lead to “tower forest” that caused aesthetic disruption and other social problems. The method is Concurrent engineering-Integrated Digital Design for good quality, cost, delivery and so all important process can be conducted simultaneously. The result is 3D geometry and photorealistic image for decision makers. The Implementation are Aesthetic BTS Tower as lighting tower, church tower and adaptable for any city needs of icon and landmark. Keywords: Aesthetic BTS City Tower – Design – Multi operator

1 INTRODUCTION BTS as telecommunication infrastructure product growth along with increasing of wireless and mobile technology demand. In the implementation, placement of the BTS (Base Transceivers Station) tower in the city often caused many environmental problems in the residential areas. Inadequate space of the city limits telecommunication infrastructure building progress. Every BTS Tower has own operator from telecommunication company. Each operator can install their own BTS tower, we can imagine if this condition continuous, there will be “tower forest” in the city that will be followed with other problems such as, disruption for city aesthetic, maintenance, and other social problems. Other considerable aspects are legal aspect and manufacturing aspect. for example Indonesian authorities have prepared to launch new regulation to provide multi user telecommunication operator in the same BTS. In other hand, local manufacture does not ready to accommodate it yet. They do not have any design solution of BTS Tower that meet requirement of multi operator need, aesthetic and cost effectiveness. to import total product would be very expensive and not considered at all by local operators. The aim of this research is to answers the problems trough cooperation with manufacturer that hopefully will produce aesthetic multi operator tower (e-tower). This e- tower is designed into city aesthetic element and regulation tools in city planning. This product is meant to be alternative product for local manufacture in order to compete in global market for telecommunication infrastructure. This research will also support to local manufacturer and authority. The design research method is Concurrent engineering Integrated Digital Design using digital process design (CAD-CAM-CAE) in order to gain shorter lead-time, better product quality and competitiveness that meet good QCD

CIRP IPS2 Conference 2009

8

(Quality, Cost and Delivery). With this method, design activities, engineering analysis, marketing activities and cost estimation can be conducted simultaneously. Resumed digital geometry data from this research will be useful for CNC machine for prototyping activity. The result in form 3D solid model and photorealistic image can be used as comprehensive presentation for business decision makers. The Implementation of the design solutions are Aesthetic BTS Tower (E-Tower) as City lighting tower, mosque/ church tower, city signage, landmark, city clock tower, and adaptable for any city need and characteristic, icon and city landmark. 2

BTS INFRASTRUCTURE IMPLEMENTATION AND DESIGN 2.1 Problems in Indonesian cities caused by BTS placement Indonesia has ten telecommunication operators, and each of them has their own BTS tower and can place it anywhere they want as they got the permit action by regulator. This uncoordinated action cause sometime conflict both for the operators and authority. Single tower for single operator cause business confrontation for BTS coverage area and local permit conflict. The tower itself has bad shape, structure and construction and becoming disruption to city plan. Worse case happened in settlement area which causes other social environmental problem. Discomfort, accident, not well maintained tower often happened and become more complicated problem. 2.2 Existing BTS tower in Indonesia BTS Tower design in Indonesia is dividing into two kinds which are conventional and camouflage tower. The conventional BTS tower is tower with commonly metal structural that only meet technical requirement to holds the

radio transceivers that define a cell and coordinates of the radio link to protocols with the mobile device. The BTS just become networking component of a mobile communications system from which all signals are sent and received. The conventional BTS Design is would not be discuss here because of it design and technically aspect would not meet the requirement of being Multi Operator BTS Aesthetic Tower (e-tower). E-tower design would refer to camouflage tower design in concealment but give better aesthetic and availability for local manufacturing ability and multi operator usage. The existing camouflage tower can be derivate as lighting tower monopole, city clock tower & monument, tree tower, street light & flag pole. 2.3 Camouflage tower design: monopole lighting tower and city clock monument tower This monopole design is concept idea by Duta Cipta Konsultama Indonesia emphasize in aesthetic and meant to be blend to local environmental. This Tower meant to be designed in order to reduce environmental and social hazard that begin to appear in Indonesia. Descriptions of the structure are: • Qualified as Safety Requirements Thru Compliance of TIA / EIA -222-E Standard Design Criteria & Loading (Dead & Wind load) • Height maximum 36 m, accommodate Mounting Floodlights Single / Multiple Levels • Positioning & Adjustment of Lights for Optimum Lighting • Mounting of Hidden Radio Antennas • Mounting Option for Billboard • Mounting Option for Security Surveillance Cameras • Ease access for maintenance This Monopole towers designed for: Public Walkways, Public Transportation Meeting Areas, Shopping and Parking Areas, Recreation Areas, Residential Areas, Business Premises, Roadway

Figure 2: existing city clock and monument 2.5 Camouflage tower design : tree tower These are existing product of Alan Dick Co. That camouflage BTS Tower into natural object as local common tree i.e.: Palm Tree, Pine Tree, and Conifer Lightning Tree

(a) (b) (c) Figure 3: Camouflage tower design as tree 2.6 Camouflage tower design : mosque tower Other Alan Dick design in Malaysia and Indonesia is mosque tower that camouflage Tower BTS into mosque tower in some cities.

(a) (a) (b) Figure 1 : existing monopole design

(b)

(c)

2.4 Camouflage tower design : city clock and monument There are existing BTS with idea of combining its function as City clock and city monument although some product was not designed very well. This idea seems become preferable for city authority because of its flexibility and idea of using existing building or city monument offer easier implementation and lower cost production.

(c) Figure 4: Camouflage mosque tower

9

2.7 Camouflage tower design : street light and flag pole.

(a) (b) Figure 5: Camouflage pole towers Based on these existing design of camouflage tower, the design of e-tower are developed. 3 CITY REGULATION AND DESIGN SOLUTION 3.1 Public regulation research Indonesia as an archipelago country with varies geological, ecological and social condition and its cities planning are very complicated. The focus of this research would be the two biggest cities that are Jakarta and Surabaya.

Figure 6: distribution BTS placement in Indonesia caused environmental problems, business and regulation conflict Jakarta's and Surabaya’s spatial planning authority will allow base transceiver stations to be attached to the tops of buildings to reduce the number of BTS towers in the capital. Jakarta authority had determined several areas, including Jl. Sudirman in Central Jakarta and South Jakarta, Jl. Thamrin in Central Jakarta and Jl. Rasuna Said in South Jakarta, where numerous high-rise buildings are located. Areas with large concentration of high-rises are called "white areas", There will be BTS devices attached to the tops of buildings, approximately in 250 different locations situated in the white areas,"

10

3.2 Public regulation and Aesthetic design that accommodate multi operator (1-10 operators) Indonesian government agency had made the option available for cellular operators, which would be required to share monopole BTS towers. "A monopole BTS tower needs to be built around 40-50 meters above the ground to work correctly. Otherwise, the signal will be blocked by high-rise buildings”. The Jakarta administration plans to replace between 2,500 and 3,000 existing BTS towers with 850 newer monopole towers. The newer towers will be occupied by at least two cellular operators, as stipulated in a 2001 gubernatorial decree on tower-sharing, which is backed by a 2008 regulation from the Communications and Information Ministry. Jakarta's and Surabaya’s property management and control agency stopped issuing licenses for new towers in 2006 and has taken down 75 illegally built BTS towers. In 2006, the licenses for 1,508 towers expired but were temporarily extended earlier this year as the authorities wait for the completion of tower-sharing studies, which will be available by the end of this year. There are at least 2,750 towers in need of temporary licenses, which will expire when the administration starts implementing the new monopole towers. Indonesian Telecommunications Regulatory has warned that the administration's plan to dismantle BTS towers could cause all cellular connections to be disrupted as operators may be forced to move from one BTS tower to another. They also said the plan was a breach of the 2008 ministry of communications regulation that says "old cellular operators should strengthen their towers by 2010 to share them with new players in anticipation of additional load". Several cellular operators have declared that they will support the city administration's multi operator BTS plan, although they claim to have received very little information on the switch. This situation solution would be good supported by E-Tower design and local manufacturing because the demand and regulation implementation are urgent. 4 METHODOLOGY 4.1 Design Methodology The design method is Concurrent engineering Integrated Digital Design (IDD) using digital process design (CAD-CAM-CAE). This method expected to be able to help gaining shorter lead-time, better product quality and competitiveness that meet good QCD (Quality, Cost and Delivery). With this method, design activities, engineering analysis, marketing activities and cost estimation can be conducted simultaneously. Resumed digital geometry data from this research will be useful for manufacturer machine for prototyping activity. The result is in form of 3D solid model and photorealistic images.  The expectation of this method are : creating product that capable in integrating 5-10 telecommunication operators in one Tower  Solving environmental problems in landscape, safety, health, in by making city icon in form of aesthetic tower along context in city need.  Superiority In engineering process with integrated digital design with concurrent engineering, Concurrent engineering is the simultaneous consideration of product and process downstream requirements by

multidisciplinary teams." (NASA Systems Engineering Handbook SP6105)  Produce design that lead to QCD (Quality Cost & Delivery) standard, shortening lead-time (time consumption for design engineering process).  Accomplish comprehensive design: result in form of data that ease any level of decision maker to decide next important decision 4.2 Design engineering process

DESIGN AND ENGINEERING Stage 1 PROCUREMENT

Stage 2 MANUFACTURING

CONSTRUCTION

Stage 3 MARKETING

by PT. BJA engineering department, public regulation formula by local authority of Surabaya and Jakarta city. 4.5 Scope of research work Implementation of IDD method and process formulation, the Scope of research work (SOW) work was be defined. The scope of work are: PRODUCT CONCEPT

New Product Design and Engineering Analysis Engineering Virtual Testing and Simulation Physical Prototyping and Testing Certification

STUDI & DRAFT DESIGN in 3D PRELIMINARY DESIGN / ALTERNATIVE DESIGN

Material Equipments

Selected DESIGN – modeled surface and rendered

Inhouse Manufacturing / Outsource Manufacturing

3D construction SOLID GEOMETRI MODEL in CAD

On the Spot Construction

Presentation and Consultation by ITSDC Design Selection By local authority and PT. BJA City light and signage camouflage Monument Engineering Analysis Geometry Testing by manufacturer Final Design Evaluation

Marketing & Promotion Finished Scaled 2D and 3D Drawings with DR&O

Figure 7: Integrated process diagram The development divides in three stages. Stage 1: Technology acquiring for digital Designprototyping and city landscape study Stage 2: Implementation in prototyping, preproduction, publication and patent Stage 3: To do manufacturing and commercialization Produce publication and patent, formulating economic added value, application in form of regulation and public rule advices. All design manufacture and Marketing stages can be conducted in the same time because the process using 3D solid and realistic product data that make impossible to use in every stag.

First year Stage 1 DETAIL DESIGN

MARKETING TOOLS

4.3 Design Requirement & Objectives (DR&O) MARKETING COMMUNICATION



BTS tower with Hi-Tech design functionally as city landmark, signage , city light giving added value as aesthetic City information facilities.  Camouflage with visual concept that differ from Conventional BTS tower.  Approach to modular construction & Completely Knock Down system  Cheaper production cost.  Consider both society and operator security and safety.  Ease of installation and maintenance. Main Concept: To be solution for environmental impact that caused by BTS structural installation in the city, Meet demand of visual aesthetic environment in the city 4.4 Partnership Partnership would be conducted between researchers (ITS design centre), local authorities and manufacturer PT Bukit Jaya Abadi (PT.BJA). Main job description areas are basic concept by ITS Design Center, manufacturing and testing

Preliminary by ITSDC Data Support from PT. BJA

PRODUCT LAUNCHING Tooling and Production Preparation

Product Design Images and specification for Marketing Tools Structure & Construction Material Specification Pre-Production / Engineering Drawing and Engineering Analysis by ITS Design centre PROTOTYPING Digital and Phisical Similar design object can be derivated Component & Material Procurement Engineering Mock Up and Assembly

PROTOTYPE TESTING

MANUFACTURING Evaluation CONSTRUCTION & INSTALLATION

Second year Stage 2 and 3

Figure 8: implementation of IDD - Concurrent engineering methods 

11

This SOW is collaboration between research team and local manufacture in order to gain optimum result suitable to production capability and raw materials readiness. . The design process involve local partner PT BJA as engineering and prototyping manufacturer industry for telecommunication Product concept analysis and usage function, aesthetic concept analysis, Preliminary Design, Detail Design & Engineering aspect:, Prototyping, Testing & Evaluation. Mass Production, Marketing & Promotion plan.  5 BTS TOWER DESIGN 5.1 Study for E-tower Design Basic consideration for e tower design is the study of required component and dimensions as technical engineering limitation.

City monument tower

RF unit (a) (b) (c) Figure 10: city monument design idea City lighting, city clock and information signage 3000 cm OPERATOR 1,2,3,4...10

1750 cm Figure 9: BTS component The main configuration of equipments inside the BTS are : Main Processor Unit initialization and self-testing, configuration, O&M signaling, software download, collection and management of external and internal data Clock Source Unit : Deliver a stable clocking pulse to all digital equipment inside BTS. Interface Unit : Interface unit have function to translate between Source data which has specific Electrical Standard (E1, T1 or IP) to digital data and this data will deliver to other digital unit to be next processed Base Band Unit : In the base band unit, the digital data will be processed and following the GSM standard, this unit creates a data which ready to be feed to RF Unit. Power Supply Unit : produce a power for whole equipments in the BTS. With input the AC voltage unit produce DC voltage as a power. RF Unit : RF Unit converts the digital signal to Radio Frequency --RF-Signal (air interface signal) following the GSM Standard. This signal type is still as an electrical signal. Antenna Unit :Antenna as a traditional unit, have a function to convert electrical signal to electromagnetic signal. This unit is very important unit for creating cell dimension. Combination of horizontal - vertical polarization, antenna height and antenna tilting influence the radiation pattern of cell. 5.2 Preliminary idea For preliminary idea, e-tower design are City monument tower, City lighting & information signage, City clock & information signage, Big Ben / Building clock tower. Next figures are 3D solid model E- tower designed for Jakarta and Surabaya for multi operator purpose.

12

(a) (b) (c) Figure 11: city BTS signage design Visual Concept : Main structure consist of Polygonal monopole structure somehow camouflage added with support by structure function as aesthetic element and based of beautiful lighting pole to give artistic touch along with city landmark style. Additional structure made of concrete, galvanized steel, and composite panel. Mosque/church tower

(a) (b) Figure 12: mosque/church BTS tower design

Visual Concept : BTS Tower main structure adjusting to building structure and camouflage with other building component as well become aesthetic element function.

Alloy Monopole Structure structure

Mullion structure

5.3 Design refinement After comprehensive discussion among researcher, local authorities and manufacturer, the design refinement can be defined.

Trans Arm ACP Detail Panel (a) (b) Figure 14: 3D Geometry for detail design As IDD method the solid 3D geometry available to be used for other such as marketing tool and CAD document (a)

(b) (c) (d) Figure 12: type design of based e-tower As discussion with metropolitan local authority and manufacturer, the design of E-tower for independent product focused on tree legs and four legs support.

Antenna

Monopole Structure

Signage and Traffic light

ACP Panel

Figure 13: one of chosen e-tower design 5.4 Detail Design integrated with manufacturing Example of integrated digital design result that simultaneously generate design both for marketing tools and detail manufacturing shows on next figure.

(a) (b) Figure 15: example of detail CAD drawing derivate from IDD method of 3D solid BTS geometry. 6 FINAL DESIGN & DISCUSSION 6.1 Design solution Final designs that have been proofed by local authority, able to be manufactured by PT.BJA are independent etower attached to building that suitable for Indonesia metropolitan. Independent design to be made the prototype is City–street light-banner-signage E-tower. The other proofed e-tower is big clock city building. Both would need more exploration in aesthetic and design development as it is still prototype design. 6.2 Design Implementation. Other factors that influence the implementation for multi operator E-tower is the good will of authority and local operators. Yet the E tower design couldn’t give any solution alone. Help from local authorities to promote and socialize program for sharing and built trust between operators is absolute. The special committee are needed chosen among operators in order to guarantee good operation and maintenance for E-tower. Environmental oriented regulation that support E-tower considered to be most important component inn implementation since the product design of E-tower has already prepared.

13

Big Clock City building E-tower

Microwave Antenna Existing Monopole Structure

ACP Panel

Analogue City Clock

Maintenance area Backlight Area

.

(a) (b)

.

(c) Figure 16: E-tower design as Big Clock City

14

Monopole structure

ACP Panel 3

ACP Panel 2

ACP Panel 1

(b)

(a)

Figure 17: E-tower design as City light, signage and monument. 7 SUMMARY This research conducted as Indonesian characteristic and need for design solution to environmental problem as BTS implementation, By Using research method Concurrent engineering - Integrated Digital Design using digital process design (CAD-CAM-CAE) in order to gain shorter lead-time, better product quality and competitiveness that meet good QCD (Quality, Cost and Delivery). With this method, design activities, engineering analysis, marketing activities and cost estimation can be conducted simultaneously. Resumed digital geometry data from this research will be useful for CNC machine for prototyping activity. The result in form 3D solid model and photorealistic image can be used as comprehensive presentation for business decision makers. The Implementation of the design solutions are Aesthetic BTS Tower (E-Tower) as City lighting tower, mosque/ church tower, city signage, landmark, city clock tower, and adaptable for any city need and characteristic, icon and city landmark. This design of E-tower is expected to solve every possible problem such as “tower forest” in the city that will be followed with other problems such as, disruption for city aesthetic, maintenance, and other social problems. 8 ACKNOWLEDGMENTS We extend our gratitude to Jakarta’s and Surabaya authority, PT. Bukit Jaya Abadi and to all who contribute to this research. Many thanks to CIRP IPS2 ‘09 International Conference that give such a honour for Indonesian researcher to participate in this design conference. 9 REFERENCES [1] Americans Planning Association, November 1995, “Cellular Tower Survey.” APA, Washington D.C

[2] [3]

[4] [5]

[6]

[7]

Bell Atlantic NYNEX, April1996, “Cellular Fact Book”.. Cassity, Pratt, November / December 1996, “When Historic Meets High Tech.”, Historic Preservation Forum News Volume 3. No.1 Gregory, Michelle, June 1995, “Local Planning Issues in Sitting Cellular Towers.” Zoning News. May 1996, “Wireless: A Planning Information Report on Mobile Communication Facilities“, New Jersey Planning Officials Report. Lai, H. -H., Lin, Y. C., Yeh, C. H., & Wei, C. H. (2006), User-oriented design for the optimal combination on product Design. International Journal of Production Economics, 100(2), 253-267. Llinares, C., & Page, A. (2007). Application of product Differential semantics to quantify purchaser perceptions in housing assessment. Building and Environment, 42(7), 2488-2497.

[8]

Moore, W. L., & Pessemier, E. A. (1993). Product planning and management: Designing and delivering value. New York: McGraw-Hill.

[9]

Nagamachi, M. (2002). Kansei engineering as a powerful consumer-oriented technology for product development. Applied Ergonomics, 33(3), 289-294.

[10]

Norman, D. A. (1990). The design of everyday things. New York: Doubleday.

[11] Norman, D. A. (2004). Emotional design. New York: Basic Books. [12] Osgood, C. E., Suci, G. J., & Tannenbaum, P. H. (1957). The measurement of meaning. Urbana: University of Illinois Press

15

Continuous Improvement of Industrial Product-Service Systems E. Schweitzer, C. Mannweiler, J.C. Aurich Institute for Manufacturing Technology and Production Systems, University of Kaiserslautern, Germany [email protected]

Abstract Industrial Product-Service Systems (PSS) are realized within the value creation network of the PSS-provider in close cooperation with customers. Thereby, the organizational and operational structure of the value creation network as well as the customer interaction itself must be designed in order to guarantee the PSSprovider continuous product and customer feedback. This feedback provides the basis for a continuous PSS-improvement process, comprising customer specific and customer spanning improvement measures. This paper analyzes demands on the value creation network structure in order to enable a PSS-provider to implement a continuous PSS-improvement process. Based thereon, a continuous PSS-improvement process is provided. Keywords: Product-Service Systems, Design Process, Development Process, Improvement Process

1 INTRODUCTION Industrial customers are increasingly expecting to be provided with services such as maintenance, upgrading, operator trainings or process improvement. These services do not only contribute to keeping up existing product functionalities [1] but also provide additional ones along the whole life cycle [2]. Since in the past capital goods manufacturers have largely focused on design, realization and distribution of high quality products, a gradual change of traditional manufacturing companies to producing service providers [3] that focus on customer solutions in terms of benefit oriented industrial Product-Service Systems (PSS) becomes necessary. To support this change, processes for product and service planning, design and realization need to be integrated [4]. Because of focus in the phase of PSS-realization lies both on providing the customers with a desired benefit through a specific configuration of products and services as well as on the establishment of measures for continuous customer specific and customer spanning PSSimprovement, both organizational and operational structures of the value creation network have to be designed in order to enable the manufacturer to operate the processes connected therewith. This paper analyzes the demands on the organizational and operational structure of the extended value creation network of a PSS-provider in a first step, complemented by a description of the demands on a systematic performance measurement of PSS as well as the processes of information exchange connected therewith. Based on these two points, an approach for the implementation of a continuous improvement process for PSS involving all members of the extended value creation network is provided by this article. 2 INDUSTRIAL PRODUCT-SERVICE SYSTEMS Industrial PSS are defined as customer life cycle oriented combinations of products and services, realized in an extended value creation network, comprising a manufacturer as well as suppliers and service partners [5], [6]. Industrial PSS, as provided in the capital goods

CIRP IPS2 Conference 2009

16

industry, are made up of a complex physical product core dynamically enhanced and individualized along its life cycle by mainly non-physical services. Thus, a PSS represents an integrated product and service offering that delivers value in use [7]. A PSS comprises customer and manufacturer related sub-systems with multiple interrelations (Figure 1) [4], [8]. Manufacturer System

Service Sub-System (Infrastructure Dimension) • Branch Offices • Staff • Tools • … Life-Cycle (Process Dimension) Purchasing Disposal Usage Product-User Sub-System (Result Dimension) • Machine • Machine User • Production Process •…

Customer System

Figure 1: Product-Service Systems. The aim of the product-user sub-system, comprising the physical product core as well as the staff of the customer responsible for its operation, is to provide an expected set of functions during a production process. The second sub-system is represented by the service network with its elements: branches, service partners, personnel, technical equipment etc. By means of the delivery of services its main functions are on the one hand to keep up and enhance the above mentioned functionalities in a customer individual way and on the other hand to continuously provide the manufacturer with customer feedback [9]. For example, maintenance services contribute to the preservation of the functional level of a product, while trainings provide the user with the

a key role in designing and realizing of PSS, there is not any separate information dimension described. This is caused by the interdisciplinary character of information. Thus, the necessary specification of information exchange processes can be covered by the three dimensions presented above [11]. A systematization of information retrieval processes builds the basis for the implementation of a continuous improvement process. Therefore, a detailed organization of the extended value creation network, comprising both organizational and operational structures, is required. While the structure of the extended value creation network in practice mostly varies between the different PSS-providers, there are general requirements of the organizational and operational structures that result from already existing PSS-Life Cycle Management concepts.

competencies, necessary for conducting different applications. Besides, due to their delivery implying direct product and customer contact, information on e.g. product reliability and usability can be obtained. Taking these sub-systems into consideration, a distinction between two life cycle perspectives needs to be drawn [2], [4]. From the point of view of the product manufacturer, the product life cycle starts with product design, followed by product manufacturing, servicing and remanufacturing. From the point of view of the customers it consists of product purchasing, usage and disposal. Since the PSS aims at enhancing the performance of industrial products by corresponding services, the non-physical service components must be provided with respect to the customer’s point of view. Considering these perspectives, the manufacturer has to design both physical products, optimized for manufacturing, servicing and remanufacturing as well as non-physical services that support his customers during product purchasing, usage and disposal. Following PSS design and the production of the physical product core at a limited number of production locations, the services are delivered at the place of product usage by the service partners of the manufacturer [4]. With respect to designing and realizing PSS, three dimensions need to be distinguished (Figure 2) [4]:

3 PSS LIFE CYCLE MANAGEMENT In order to promote the implementation of PSS in practice, physical product focused Life Cycle Management (LCM) can be taken as a promising starting point [11]. LCM aims at organizing the interactions between the different partners within the value creation network along the life cycle by a set of methods and processes for the design and realization of physical products [12]. Based on current physical product focused LCM concepts, the following fields of action are addressed by already existing PSS-LCM concepts [4]:

• The physical and non-physical product and service components together provide the customer with a certain set of expected functionalities that represent the product or result dimension of the PSS.

• Methods and processes for customer-oriented PSSplanning that support the proactive specification of Life Time Management measures in terms of services.

• PSS realization is based on different processes such as product maintenance and training that continuously change the state of the product-user sub-system and the service sub-system along the life cycle, e.g. in terms of improved machine or service components. They represent the process dimension of the PSS.

• Methods and processes for integrated PSS-design that allow the specification of the different models required for describing the three dimensions of a PSS. • Methods and processes to exploit the service potentials for providing the manufacturer with product, customer and market feedback.

• Finally, the service network provides the resources for executing the state changes as well as for providing the manufacturer with continuous product, customer and market feedback. It thus represents the infrastructure dimension of the PSS. The realization of information retrieval processes represents a basic function of the service components of the PSS and aims at providing the manufacturer with widespread product, customer and market feedback. It thus represents the basis for a continuous optimization of the product and service offerings as well as the corresponding processes and resources by complementing the already existing knowledge of the manufacturer [10]. Although the information aspect plays

Result Dimension

• PSS Performance Measurement Systems for customer individual Life Cycle Evaluation and definition of consequential PSS-improvements. • Life-cycle oriented process management based on standardized process descriptions to get all partners in the extended value creation network having a common understanding of the necessary design, production and servicing processes. Based on the manufacturer point of view concerning the PSS life cycle, the framework for PSS-LCM underlying this paper [11] consists of four phases (Figure 3).

Process Dimension

Infrastructure Dimension

i

nein Kunde entscheidet Angebot

ja Kunde kontaktiert Wirtgen-Service

Pauschalangebot Wirtgen-Service an alle Kunden

Kunde erteilt Auftrag

Serviceabteilung bearbeitet Kundenanfrage

Bewertung der Anfrage

i i

i

durchführbar

nicht durchführbar

• System Elements • Product • Service • Relations • Service-Product • Service-User • Service-Usage • … • Type of Relation • Maintain Benefit • Enhance Benefit • Control Benefit

• Process Type • Order Processing • Service Delivery • Procurement of Information • … • Process Specification • Type of Process Model • Degree of Detail • Addressee • …

Figure 2: PSS Dimensions.

17

• Extended value creation network • Network Management (PSS-Managem., Controlling) • Involved Parties (Customer, Manufacturer, Service Technician etc.) • Communication Channels • Information Specification • Quality (What?) • Quantity (How Detailed?) • Time (Until When?) • Context (In Which Situation?)

Organizational Implementation

These functions only partially can be fulfilled by the manufacturer. Furthermore, all partners of the extended value creation network have to be involved in performing the tasks connected therewith. This leads to different requirements on the expertise of the network partners. The extended value creation network of a PSS-provider comprises partners necessary for both the production of the physical product core of the PSS as well as for the delivery of the service components (Figure 4).

PSSRealization

operational [customer control loop spanning] [customer specific] PSS-Planning

strategic control loop

Production Network Supplier 1

PSS-Design [project proposal rejected]

[project proposal accepted]

Manufacturer Supplier 2

Figure 3: Control Loop Model for PSS LCM. In the organizational implementation phase, the basis for PSS-LCM is laid in terms of building up the necessary design and realization processes. Besides this, the implementation of the required organizational and operational prerequisites within the extended value creation network of the PSS-provider takes place. The PSS-planning phase pertains to the identification and definition of physical and non-physical PSS-components contributing to the aims of both manufacturer and customer [8]. For their model based description, the PSSdesign phase covers the subsequent planning and execution of a PSS development project. Thereby, focus is on the integration of product and service design [13]. After the customer individual life cycle oriented configuration of the PSS [14], the phase of PSSrealization aims at providing the customer with a certain benefit. Thereby, the gathering and the analysis of feedback information that supports continuous PSSimprovement is addressed by the manufacturer. Thus, the operational and strategic control loops as shown in Figure 3 can be established in the value creation network. 4

DESIGN OF THE VALUE CREATION NETWORK

4.1 Functions of the Value Creation Network To fulfil the customer demands and to achieve a high customer satisfaction, the manufacturer (PSS-provider) has a higher degree of responsibility for the product’s full life cycle [6], [15]. Taking a look at the PSS life cycle from both points of view, it becomes apparent that an integrated product and service offering results in a long term cooperation between the PSS-provider and the customers. Consequently, this also leads to high requirements for the organizational and operational structure of the belonging value creation network [15]. Compared to networks of manufacturers only providing physical products, the service components of a PSS pose a challenge for structuring of the value creation network. Thereby, these service functions [16] that can be distinguished as follows have to be fulfilled by the network in addition to the functions of a single product: • The support function refers to ensuring the expected benefit of the product by means of technical services. • Requirements fulfilment refers to enhancing this use by means of complementary offers, e.g. upgrading, user training or application consulting. • The procurement of information represents an internal function since it aims at providing the manufacturer with an expected customer, product and market feedback.

18

Supplier 3

Branch PSS-Provider

Service Partner

Service Network Dealer Figure 4: Extended Value Creation Network. 4.2 Production network The production network includes both the provider of the PSS as well as suppliers of parts, components, modules or systems [17]. It is responsible for the production of the physical product core of the PSS at a limited number of production locations of the manufacturer. Following elements of the production network can be identified: Manufacturer / PSS-provider The manufacturer of the physical product core represents the centre of the value creation network. Above all, the manufacturer is responsible for planning and design of the physical products provided all over the world as well as their production realized in closed cooperation with the supplier network. Suppliers The suppliers are responsible for the delivery of parts, components, modules or systems. Thus, they support the manufacturer within the production of the physical product core of the PSS. 4.3 Service Network The service network comprises both branches and service locations of the manufacturer as well as independent distribution and service partners [18]. This service network is responsible for the successively delivery of the services throughout the product life cycle and comprises the following elements: Manufacturer / PSS-provider Additionally to the production of the physical PSScomponents, the planning and design of the complementary service offerings is realized centrally by the manufacturer. Within these strategic processes, the local branches or service partners being responsible for the service delivery are not directly involved in. Dealers and branches At important markets, the PSS-provider relies on own dealers and branches. They depend on the PSS-provider in terms of economical and legal aspects. Additionally to the dealers, branches usually engage product specialists that rely on knowledge of a specific product type.

customers or that can inform about customer preferences have to be involved in. PSS-Design The design of a new type of PSS is realized in two steps. Firstly, the main characteristics of the new PSS type is planned and designed centrally by the manufacturer. Thereafter, the corresponding product and service components of the PSS are adapted according to the specific market requirements. While the responsibility for the detailed design of the physical product still remains with the manufacturer, the service offerings are customized by the partners of the service network that are present at the local markets. PSS-Configuration Based on the resulting market specific product and service components, the customer order processes represent the next step in PSS-LCM. Supported by the staff members of the service partners, customers firstly configure the physical product core of the PSS in terms of selecting desired components and functionalities, followed by the definition of the corresponding service components. If necessary, these service components can be customized by the service partners. PSS-Realization After the delivery of the product to the customers, the services are delivered by the service partners in close cooperation with the customers. Additionally, the service partners are responsible for the customer benefit oriented continuous improvement of the PSS as well as the gathering of feedback information. The resulting improvements are realized either centrally by the manufacturer or locally by the service partners. The assignment of tasks has to consider that, within the scope of organizational implementation, the different members of the value creation network have to be qualified in order to fulfil their tasks in a right way. This particularly regards the integrated qualification of the network partners in view of a systematic planning, design configuration and realization of PSS that comprises both product and service components. Hence, it is crucial that all network partners rely on further expertises in designing and adapting processes as well as the corresponding infrastructure. 5 PERFORMANCE MEASUREMENT OF PSS Focus in the PSS-realization phase lays both on providing the customers with a desired benefit through a specific configuration of products and services as well as on the establishment of the described operational and strategic

Service partners At markets that do not have any branches owned by the PSS-provider, independent partners are responsible for the distribution of the products and services as well as the delivery of the services. Thereby, they act on behalf of the PSS-provider. Customers While not directly being a member of the value creation network, the customer itself plays an important role during the delivery of services. Due to the characteristics of services, he is involved in the delivery of services in terms of an external factor. It becomes apparent that in the extended value creation network the branches, service partners or dealers can be regarded as being in the “middle”, linked forward to the customers and backward to the manufacturer [19]. Thereby, the so called account managers usually employed at a specific dealer or branch are responsible for the individual customer contact in terms of a one-faceto-the-customer policy. 4.4 Assignment of Tasks during PSS-LCM All along the long term cooperation of the network partners, the assignment of tasks and responsibilities often changes because of the different challenges in PSSLCM (Figure 5). Thus, the following phases need to be distinguished. Organizational Implementation Based on the decision to change from a traditional manufacturing company to a producing service provider that focuses on customer solutions, the manufacturer analyses the existing organizational and operational structures in product and service management. Thereby, main focus lays on the analysis of the characteristics of both existing design processes as well as already implemented information exchange processes. Based thereon, processes of PSS-LCM are defined by the manufacturer in order to enable the branches and service partners to support and to participate in planning, design, configuration and realization of PSS. PSS-Planning PSS-planning aims at identifying, selecting and specifying of PSS ideas to be developed within the following PSSdesign phase. Thereby, the planning of both product and service components of the PSS take place centrally by the manufacturer. In these processes, an essential input is represented by information gathered while realizations of similar PSS that already exist. For this reason, the organizational units that are directly in contact with the

Realization

Development Design

Manufacturing

Servicing

Manufacturer Recycling point of view

Configuration

Planning

Customer point of view manufacturer

Decision

Purchasing

Usage

Investment branch / serivce partner / dealer

customer

Figure 5: Assignment of Tasks during PSS-Life Cycle.

19

Disposal leading part

control loops according to Figure 3. To measure the performance of a PSS systematically, two important prerequisites have to be fulfilled. On the one hand, a continuous information flow between the customer respectively the service technician being in contact with the customer and the branch, service partner or dealer responsible for the support of a customer individual PSS has to be established. On the other hand, key figures to scrutinize the desired benefit need to be defined. 5.1 Key Figures Predefined reference values underlying the evaluation of the performance of PSS require a standard of comparison that should be quantifiable [20]. For example, a standard of comparison can be requirements that may be derived from customer or manufacturer targets. Thereby, the characteristics of this standard as well as their possible values must be known (e.g. the life cycle costs of a PSS and their range). In general, this standard of comparison is characterized as a key figure. Consequently, key figures are defined as information that allows a qualitative and quantitative description of circumstances and facts in a concentrated way [21]. Thereby, key figures can have both a monetary and a non-monetary character and aim at controlling and analysing environmental conditions as well as business processes. They provide the basis for a systematic planning and evaluation of alternative solutions or are used for controlling processes [22]. Thus, key figures provide a promising device for performance measurement of industrial PSS. 5.2 Key Figures in Context of PSS The definition of key figures for each customer individual PSS is carried out within the PSS-configuration in order to provide a measurement of the PSS performance. This happens in collaboration between the customer and the respective branch, dealer or service partner and depends both on manufacturer and customer targets to be fulfilled along the PSS life cycle. Thus, at the beginning of the PSS-realization phase key figures are defined that allow a continuous examination of the benefit by gathering and analysing field data. Thereby, the key figures can refer to the different dimensions of a PSS as described above: • Product or result oriented key figures describe the benefit provided by the PSS for both customers and manufacturer. These key figures are often related to certain functionalities of the PSS that are realized by both product and service components of the PSS (e.g. product oriented quality figures, such as availability). • Process oriented key figures are related to the processes necessary for providing the expected benefits. This kind of key figure is for example used for an assessment of the efficiency and effectiveness of maintenance and repair processes. • Infrastructure oriented key figures describe the resources (e.g. energy) necessary for processes that aim at guaranteeing the realization of expected benefits. • Information oriented key figures allow a description of the information exchange processes between the customer and the PSS-provider during the product usage phase. One of these key figures can e.g. be the availability of a service hotline. The key figures can thereby be related to both product and service components of a PSS as well as the interrelations of the product and service components. In connection with the definition of the key figures, processes for gathering and analysing the corresponding information have to be determined. This also includes the definition of the organization units responsible for gathering the information as well as the definition of the

20

target values of each of the key figures. The definition of critical values of the key figures determines, in which situation the PSS-provider respectively his network partner has to think about or to initiate appropriate product or service related improvement measures that can be either customer individual or customer spanning. 5.3 Processes of Information Gathering To implement a systematic evaluation of the PSS performance, a few network partners has to be involved in performance measurement mechanisms. Primarily, this concerns the network partners communicating with the customer or servicing a customer individual PSS. These network partners have to support the processes of information gathering and analysis with the objective of an immediate forwarding of the information to the appropriate organizational units. Thereby, following sub-processes of information gathering can be distinguished [9]: • Identification of information: Information gathering processes can be initialized both internally by an organizational unit (e.g. information about a PSS required for the PSS-planning are specified by the manufacturer) as well as externally by a staff member communicating with a customer. In the latter case, a staff member being in closed contact with the customer independently gathers information that maybe is of interest to the manufacturer. • Gathering information: Gathering information describes the admission of field data. Thereby, service reports are generated by the service technician in form of paper based standard forms or by means of electronic resources. Afterwards, these service reports are transferred to the manufacturer or his network partner responsible for servicing the customer individual PSS. • Analysis of information: This sub-process comprises the examination of the information for completeness and plausibility and an assessment of their relevancy for the manufacturer. Thereby, it can e.g. be distinguished whether information can be used for further purposes or if information only has statistical characteristics. • Allocate information: The last step of the information gathering is the appropriation of the information to an organizational unit responsible for the further processing of the information. Simultaneously, the person who gathered the information gets feedback about the relevancy of this information. 6 CONTINUOUS IMPROVEMENT PROCESSES A Continuous Improvement Process (CIP) demonstrates a mindset adopted by all persons involved in. CIP aims at continuously changing for the better while the developed solutions should have sustainable effect. Thus, it is related to product, process or service quality. Thereby, CIP comprises all activities as well as the whole company network of a product, service or PSS-provider. 6.1 Fundamentals The idea of establishing Continuous Improvement Processes traces back to the Japanese Kaizen concept that aims at consistently changing things for the better. Thereby, the basic idea of Kaizen is the increase of productivity by a stepwise continuous improvement. CIP as a synonym for Kaizen concentrate on the processes necessary for either the production of a product or the delivery of a service. Thereby, deviations, differences, anomalies or failure in these processes are systematically determined. Based thereon, solutions have to be developed in formalized problem solving processes in order to eliminate the identified weaknesses. These

solutions afterwards are realized by the affected organizational units. Following phases of CIP can be distinguished (Figure 6): • Understanding existing problems, • Analysis of causes, • Development of product or process improvements and planning of remedial action, • Taking selective measures, • Testing the measures for target achievement, • Improvement of the taken measures and definition of the measure as a new standard.

Understanding the problem

Analysis of causes

Definition of a new standard

Testing for achievement

Development of improvements Taking selective measures

Figure 6 : Continuous Improvement Processes. 6.2 Towards Continuous Improvement of PSS To implement CIP in context of PSS, different requirements to be fulfilled can be identified as follows: • The PSS to be improved has to be described sufficiently with regard to all three dimensions (result, processes and infrastructure). In the same way, a border between the PSS and its environment has to be established. • The model based description of the PSS has to be standardized within the whole value creation network to enable the different network partners to work on the improvements of the PSS together. • The operation conditions of the PSS resulting from the specific customer life cycle must be describable in order to model them systematically. • The PSS-improvement processes have to be standardized within the value creation network as well in order to guarantee a high process quality. Furthermore, the resulting requirements on the expertises of the network partners have to be defined clearly in order to succeed the processes. • All staff members within the extended value creation network have to internalize the interrelations of the business processes within the network to guarantee the success of the network partner spanning

improvement processes. • Since improvement measures predominantly aim at improving a customer individual PSS, an appropriate Performance Measurement System has to allow an assessment of the actual performance of each PSS. • Information and communication networks not only have to ensure the forwarding of information generated within the improvement process. Furthermore, the networks must enable a judgement of information by all network partners in order to attach importance to information. These prerequisites have to be taken into account while implementing a PSS-CIP. 7

CONTINUOUS IMPROVEMENT OF PSS

7.1 Overview Due to the characteristics of PSS described above, a PSS-improvement process can be seen as a process taking place within the whole value creation network, especially the service network. Based on a systematic problem description, a problem analysis and assessment takes place. Thereafter, possible solutions are planned and assessed. Thereby, it has to be determined, whether customer individual or spanning measures will be taken. The implementation of the selected solution closes the PSS-CIP (Figure 7). 7.2 Organizational Prerequisites As an initial step of improvement processes, the system to be improved has to be defined. In case of PSS, this description comprises both the precise definition of the result, process and infrastructure dimension of a customer individual PSS as well as the establishment of a border between the PSS and its environment. This can be done by means of the model based description of the PSS developed during the PSS-design phase and standardized all over the value creation network. Thus, the system to be improved subsequently becomes clear. To provide support to the service technicians at problem solving, a solution database has to be introduced. This database contains product and servicing problems appeared in the past and described in connection with appropriate solutions. Since this database is available all over the world, all partners of the extended value creation network can access this database as well as update both problems and solutions described therein. During configuration, PSS specific key figures are determined by account managers operating within one of

Problem report

Solution database

Solution

Service technician is not able to solve the problem 1 2 Research

Solution

Solution

Update of the solution database

Continuous update

Product improvement by redesign

Identification of a significant problem Standardized problem report Product specialists at the branch

Problem assessed by the manufacturer Figure 7: Continuous Improvement Processes for PSS.

21

the service network partners. These key figures allow for the measurement of the actual performance of each customer individual PSS. 7.3 Problem Description and Analysis The description of problems and the analysis of its causes represent the first step of a PSS-improvement process. The measurement of the actual PSS performance can be seen as an initial step. The measurement on the one hand can be initialized as a reaction of customer complaints. In this case, service technicians are sent to the customer in order to solve an existing problem detected and reported by the customer. Thereby, the individual key figures of the PSS can be determined. On the other hand, these key figures can be determined pre-emptively. This means e.g. that measurements can be taken in planned intervals or in the context of planned on-site inspections. By means of the measurement results, the actual performance of the PSS can be described. Thereafter, the information is analyzed by account managers in order to compare this state with a target state individually defined for each PSS. In case of deviations from the target state, the actual condition of the PSS as well as the problems addressed have to be described in a detailed way by a service technician. Furthermore, the problems are assessed with regard to the requirements resulting form customer production processes to be fulfilled by means of the PSS. This allows conclusions to malfunctioning processes that indeed cause the problem but that cannot be solved by improving the product or service components of a PSS. A further specification of the problem follows the problem description described above. Thus, possible causes for the problem, interfaces to operation conditions as well as side effects can be described. This description usually is done by a service technician who is in face of the customer. Based thereon, the targets underlying the improvement process can be specified by either the service technician or the account manager. This includes the limitation of the field of activity. Thus, it becomes clear, if the measures to be developed address the result, process or infrastructure dimension of the PSS. While improvement measures predominantly address customized PSS, some specified measures can not only be significant for a unique PSS, but also concern a whole PSS-type. Therefore, the problem occurred has to be assessed by product specialists within the responsible branch. This comprises on the one hand the case that the problem can directly be solved by the service technician because of knowing the solution or finding a similar problem yet described in the solution database. If so, the problem is only documented in form of a service report that enables the product specialists at the branch to analyse and to assess the specific case. On the other hand, if the service technician is not able to solve a problem occurred, he can directly apply to the branch. After a description of the problem, the product specialists can support the service technician and help him to select an appropriate immediate measure. Concurrently, the solution database has to be updated by the product specialists. In both cases, the assessment of the problem done by the product specialists allows them for drawing conclusions on the character of the problem. Thus, problems that require customer spanning improvements of product and service components of PSS can be identified and distinguished from individual problems. Thereafter, the product specialists assess, whether a customer individual PSS-improvement is sufficient in the further course of the problem solution process or if customer spanning improvement measures have to be taken. Customer specific improvement measures are related to a customer individual PSS and hence are

22

implemented on-site by either a branch or a local service partner. Thereby, it is primarily about the adjustment of product, user or process oriented services, e.g. minor modifications of a machine or machine specific user trainings. In contrast, customer spanning improvement measures are relevant for a multitude of PSS of the same type. For example, these measures are related to often occurring machine faults or often occurring customer requirements that are not yet fulfilled. In case of a thinkable development of customer spanning improvement measure, information about the problem as well as rooms for improvements identified by the product specialists are forwarded to the product manager of the manufacturer responsible for the concerned PSS-type. This can e.g. be done by means of standardized problem reports. Based on these problem reports, the product managers of the manufacturer can define customer spanning improvement measures and take them in the worldwide value creation network. 7.4 Planning of Solutions and Measures The development, assessment and selection of solutions to be suggested represent the next step of the PSS-CIP. Based thereon, a definition of the improvement measures as well as a planning of their implementation takes place. Gathering of ideas for the solutions of the specified deficits can take place on different levels. At local level, the solution can be generated by the service technician either autonomously by means of the solution database or in cooperation with the customer. If the problem case is forwarded to the branch by the service technician the solution has to be created by the branch (in case of customer specific solutions) or by the product managers of the manufacturer (customer spanning solutions). If there is more than one solution for one specific problem, the solution ideas have to be assessed by the responsible organizational unit. Based thereon, a selection of the most promising solution that builds the basis for the derivation of corresponding improvement measures takes place. Thereby, revenue and expense of the special improvement measure have to be considered. After the specification of the improvement measures, the responsibilities for taking the measures as well as the required resources are defined. This can either be done by the branch in case of customer specific improvement measures or be done by the manufacturer in case of customer spanning improvements. 7.5 Implementation of the Improvement Measures In the next step, the implementation of selected and preplaned improvement measures follows. PSS-improvement measures are mostly implemented by the local branch. Above all, this concerns customer specific adaptations of the service components of a PSS (e.g. additional user trainings or upgrading). Therefore, it is important that the branches are able both to develop and implement such product and service related measures as well as documenting their implementation autonomously. Customer spanning measures mostly are extensive since they usually are connected with a modification of the physical product. Therefore, these measures are coordinated by the manufacturer himself and realized in close cooperation of all network partners. After the implementation of the solution, the success of the taken measures has to be measured by means of the established Performance Measurement mechanisms. Thereby, in case of extensive PSS-improvements it could be necessary to adjust the key figures existing for each customer individual PSS. This can also be connected with an adjustment of the corresponding performance measurement and information gathering processes.

The documentation of the taken measures concludes the improvement process. Thereby, the standardized modelbased PSS-description enables each of the network partners to document the improvements comprehensibly. 8 CONCLUSIONS Processes that aim at continuously improving PSS need to match the requirements resulting from the PSScharacteristics, e.g. the realization of PSS in a worldwide value creation network. Special focus in this paper has therefore been laid on the establishment of organizational and operational network structures as well as appropriate performance measurement mechanisms. Based thereon, an approach to customer individual and customer spanning PSS-improvement has been introduced. Thereby, it became apparent that, in contrast to physical products, the PSS to be improved has to be considered as a system that can be described with regard to result, process and infrastructure dimensions. Consequentially, these dimensions have to born in mind while both establishing performance measurement mechanisms within the whole value creation network as well as selecting appropriate measures for PSS-improvement. Due to the fact that PSS are realized within the whole value creation network, the comprehensive definition of information exchange processes represents a crucial factor of success on implementation of a continuous improvement process for PSS. Further research has to be done to refine the information exchange processes within the value creation network, especially the service network. This also includes the detailed specification of information with regard to the specific characteristics of PSS. Additionally, the system characteristics of PSS have to be analyzed in order to enable the organizational units responsible for the PSSimprovement to take improvement measures concerning both product and service components of a PSS while considering their interactions. 9 REFERENCES [1] Takata, S., Kimura, F., van Houten, F.J.A.M., Westkämper, E., Shpitalni, M., Ceglarek, D., Lee, J., 2004, Maintenance: Changing Role in Life Cycle Management, Annals of the CIRP, 53/2: 1-13. [2] Aurich, J.C., Fuchs, C., 2004, An Approach to Life Cycle Oriented Technical Service Design, Annals of the CIRP, 53/1: 151-154. [3] Schuh, G., Friedli, T.; Gebauer, H., 2004, Fit for Service: Industrie als Dienstleister, Carl Hanser Verlag, München. [4] Aurich, J. C., Schweitzer, E., Fuchs, C., 2007, Life Cycle Management of Industrial Product-Service Systems, Advances in Life Cycle Engineering for Sustainable Manufacturing Businesses, Springer, London: 171-176. [5] Aurich, J.C., Fuchs, C., Wagenknecht, C., 2006, Life Cycle Oriented Design of Technical Product-Service Systems, Journal of Cleaner Production, 14/7: 14801494. [6] Mont, O.K., 2002, Clarifying the Concept of ProductService System, Journal of Cleaner Production, 10/3: 237-245.

[7]

[8]

[9]

[10] [11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21] [22]

23

Baines, T.S., Lightfood, H.W., Evans, S., Neely, A., Greenough, R., Peppard, J., Roy, R., Shehab, E., Braganza, A., Tiwari, A., Alcock, J.R., Angus, J.P., Bastl, M., Cousens, A., Irving, P., Johnson, M., Kingston, J., Lockett, H., Martinez, V., Michele, P., Tranfield, D., Walton, I.M., Wilson, H., 2007, Stateof-the-art in product-service systems, Journal of Engineering Manufacture, 221 (B), 1543-1552. Aurich, J.C., Schweitzer, E., Fuchs, C., 2007, Life Cycle Oriented Planning of Industrial ProductService Systems, Proceedings of the 5th International Conference on Manufacturing Research, Leicester: 270-274. Warnecke, G., Schülke, P., 2002, Design of Preventive Customer Service Processes, Production Engineering, 7/2: 75-78. Brissaud, D., Tichkiewitch, S., 2001, Product Models for Life-Cycle, Annals of the CIRP, 50/1: 105-108. Fuchs, C., 2007, Life Cycle Management investiver Produkt-Service Systeme Konzept zur lebenszyklusorientierten Gestaltung und Realisierung, Dissertation, TU Kaiserslautern. Westkämper, E., Alting, L., Arndt, G., 2000, Life Cycle Management and Assessment: Approaches and Visions towards Sustainable Manufacturing, Annals of the CIRP, 49/2: 501-522. Aurich, J.C., Schweitzer, E., Mannweiler, C., 2008, Integrated Design of Industrial Product-Service Systems, The 41st CIRP Conference on Manufacturing Systems, Tokyo, Japan, 26-28 May: 543-546. Aurich, J.C., Wolf, N., Mannweiler, C., Siener, M., Schweitzer, E., 2008, Lebenszyklusorientierte Konfiguration investiver PSS, wt Werkstattstechnik online, 98/7-8: 593-600. Meier, H., Völker, O., 2008, Industrial ProductService-Systems - Typology of Service Supply Chain for IPS² Providing, The 41st CIRP Conference on Manufacturing Systems, Tokyo, Japan, 26-28 May: 485-488. Aurich, J. C., Fuchs, C., Wagenknecht, C., 2006, Modular Design of Technical Product-Service Systems, Life Cycle Engineering and Sustainable Development, Springer, Berlin: 303-320. Wildemann, H., 1996, Beschaffungslogistik, Produktion und Management Teil 2, Springer, Berlin: 15-11 – 15-52. Aurich, J.C., Fuchs, C., Jenne, F., 2005, Entwicklung und Erbringung investiver ProduktService Systeme, wt Werkstattstechnik online, 95/78: 538-545. Hope, C., Mühlemann, A., 1997, Service Operations Management - Strategy, design and delivery, Prentice Hall, London. Luczak, H., P. Drews (Hrsg.), 2005, Praxishandbuch Service-Benchmarking, Service Verlag Fischer, Landsberg am Lech. Reichmann, T, 2001, Controlling mit Kennzahlen und Managementberichten, Vahlen, München. Gladen, W., 2001, Kennzahlenund Berichtssysteme, Gabler, Wiesbaden.

Service Development and Implementation A Review of the State of the Art M. Torney, K. Kuntzky, C. Herrmann Institute of Machine Tools and Production Technology (IWF), Technische Universität Braunschweig, Langer Kamp 19B, 38106 Braunschweig, Germany

Abstract As service development is a complex and interdisciplinary issue, service research has been conducted in numerous disciplines and sectors. As a result, different approaches containing manifold models, methods, and tools with different foci have been developed and discussed in recent years. Due to the diverse nature of the approaches, it is a challenge for organizations to distinguish between them. Against this background, the aim of this paper is to propose a framework for classification and, based on the framework, to present an overview of the state of the art in the research field of service development and implementation. Keywords: Service Development and Implementation, Service Engineering, State of the Art

1

INTRODUCTION

The relevance of services, which belong to the tertiary sector, has increased significantly in industrial countries during the last century. Today, in many industries, services are the most important business sector, amounting to more than 70% of the national economy [1] [2] (cf. Figure 1). This increasing impact of the service sector is not only based on the growth of self-contained services. Changes in the producing industries involve modification of product ranges. Product-relating and product-supporting services have become integral part of the tertiary sector [3]. Percentage of employees per sector

primary sector secondary sector

tertiary sector

70 60 50

So far, no distinct definition of the term “service” is established in scientific literature [7] [8]. Generally, three distinctions are made in the literature [9] [10]: the definition by enumeration (q.v. [11] [12]), the definition by negative demarcation (q.v. [13]), and the definition by the constitutive characteristic of service. The latter definition is the most prevalent [14] and will be used as basis here. The constitutive characteristics of services are intangibility and the integration of external factors (i.e. customers) and therefore the uno-actu-principle (q.v. [9] [15]). Compared to conventional product development processes, this service’s related characteristics imply particular challenges for the development process of new services and the required methods and tools. Differences in the degree of customer integration and in the variety of service variants lead to a heterogeneous range of service types (Figure 2). Four different service types can be distinguished [16]: high

40

Customer-focused services

Knowledge-focused services

e.g. call center, fast food restaurant

e.g. consulting, medical practice

Process-focused services

Flexibility-focused services

e.g. car wash, online banking

e.g. life insurances, IT outsourcing services

30 20 10 0 1850 1870 1890 1910 1930 1950 1970 1990 2010

contact intensity

Year

Figure 1: The economic significance of services in Germany [1] [2]. For self-contained services as well as for product-relating and product-supporting services, quality is essential for the success of offered services. However, as a study of Cooper/Edgett [4] showed, more than 40% of services introduced failed to survive in the market. Reasons for this high failure rate are often caused by two (main) factors: the spontaneous and unsystematical procedure for the service development [5] and a lack of service-specific methods and tools [6].

CIRP IPS2 Conference 2009

24

low low

high

variety

Figure 2: Types of services [16] [17]. •

Process-focused services: The integration of an external factor as well as the variety is low. Therefore, a high-standardised service process is applicable.



Customer-focused services: Because of its low variety and high contact intensity, clearly defined standardised services are proposed which can be influenced by customers only to a certain degree.



Flexibility-focused: Regarding the high variety, the emphasis is on a systematic and efficient way for creation of variants (i.e. modularisation).



Knowledge-focused: Services are determined by high contact intensity and high variety. Due to this fact, standardisation is difficult, as a considerable amount of customisation is necessary. Due to these, a lot of rather problem-specific, singular solutions exist. In consequence, a deduction of a uniform standardised development process is very difficult. Simultaneously, the services to be developed are significantly influenced by different factors [18]. Thereby, the most import factors are the strategic alignment and the general conditions. Along this background and its complexity the main question is how to develop service in an efficient way in order to ensure high-quality service processes. As services are not just “black boxes” but “a designable part of the business activities” [17], an adequate development process is required. Consequently, the development process for services needs to be systemised or standardised in the same way as it is done for physical products. Hence, the differentiation between “physical product” and “service product” will be made in the remainder of this article. A development process that is standardised tested and adapted to the special requirements of services benefits from the quality and professionalism of the development process itself [19]. By introducing standardised processes, the existing high failure rates in the implementation of services can be reduced [20]. Furthermore, positive influences on the time-to-market [21] can be achieved as well, due to a more convenient operation method and subsequently reduced development costs [22] [23]. 2 FIELDS OF SERVICE RESEARCH Service research considers a socio-technical system. Therefore it is a complex and interdisciplinary issue, which is investigated by various scientific fields. Research in services is not a self-contained scientific discipline up to today. As a result, research has been conducted in numerous disciplines and sectors like engineering technology, information technology, economics or psychology (see Figure 3) [15] [24]. Engineering technology - Product development - Industrial engineering - System engineering - etc.

Information technology - Software engineering - Information engineering - Software development - etc.

Psychology - customer orientation

Service Research

- HR development - HR management - Interaction organisation - etc.

Business administration - Marketing - Controlling - Quality management - Innovation management - etc.

Figure 3: Important factors of influence for the organization of services.

Accordingly, different approaches containing manifold models, concepts, and tools with different foci have been developed and were discussed in recent years. Until now, none of theses approaches could be established as a general accepted one [24]. Service researches can be divided in service development and service management, respectively service operation (see Figure 4). While on the one hand the research in the development of services deals with the process from the first idea for a new service up to the post processing of its introduction or with an improvement of an existing service, service operation research on the other hand focuses mainly on management and the continual improvement of already implemented services. This research paper concentrates on the existing approaches for service development. Here, the field of research can be divided into two different mainstreams:

Displacement Idea Management Evaluation

Requirements Definition Production

Implementation

phase consistently in focus of definitions

Development of service design

phase consistently in focus of definitions

Figure 4 : Research in service life cycle. New Service Development and Service Design First researches were based on the area of business administration and on particular marketing. They focus on customer satisfaction and service quality. Using the terms of “new service development” they were conducted with the aim in mind of devising a systematic and analytic method for the development of new services, which can reproduce an expected result in a satisfactory quality with reasonable costs [8]. The first approaches referring to the topic as “New Service Development” in the 1970s in the Anglo-American sphere were held on a simple level and did not offer practical methods [16]. This field of research mainly started in the USA with the “Service Design and Service Management” Model of RAMASWAMY (1996) [25]. This model can be described as the basis of many subsequent researches and models for service development. With respect to the definition given by LUCZAK, “Service Design” focuses on the elaboration of perceivable elements of a service (e.g. colour, sound) [26]. Whereas in the definition of Service Design given by RAMASWAMY all steps of the design of a new service are included [25]. In recent years the “New Service Development Model” was developed by EDVARDSSON/OLSSON (1996) [27]

25

Service Engineering The second mainstream in the development of services is the new research discipline “Service Engineering”, which was created especially for the systematic development of services. The notion of “Service Engineering” was chosen on the assumption that services can be developed in the same way as physical products. Hence the term “service engineering” clarifies the aim of finding a process for systematic development and design of new services by using appropriate methods and tools [28] [29] [30]. Service Engineering mainly aims at the improvement of service planning and service developing procedures, in order to create more professional services. This approach is mainly prevalent in Germany. Today, research regarding the development of services is mainly done in the USA, the UK, Canada, Scandinavia and Germany [23]. 3

FRAMEWORK OF REFERENCE FOR SERVICE DEVELOPMENT Due to the complexity and the diverse natures of existing service development approaches, it is a real challenge for organizations to distinguish between them. Furthermore, it is a task for them to identify and evaluate the different approaches, finding out which is capable of fulfilling the given requirements. Against this background, a framework (Figure 5) is proposed in this paper to classify the manifold kinds of concepts, methods and tools of service development and implementation with their different foci. The aim thereby is to render a concise overview and to provide transparency for concerned organisations. The constructivist point of view is based on the assumption that a service can be designed just as a physical product (cf. chapter 1). Thus, the framework is based on the definition of the term service engineering. In this way, the disciplines service design and new service engineering are neither considered separately nor will there be made a distinction between them. The presented framework distinguishes between three aspects: activity dimension, service dimension and aggregation layer. 3.1 Activity dimension According to the definition of service engineering, the framework deals with the fields of activity within the service

development process. The individual stages of the service development process are visualised in Figure 5. The process is divided into three main stages: “service planning”, “service conception” and “service implementation”, which are again divided into three substages (cf. [26] [29] [31]). The first phase, “service planning”, includes the situation analysis of the company and the environment analysis, in order to identify the requirements of the stakeholders. However, depending on the stakeholder group, the derived requirements can have different and partly opposing foci. Based on this, a target system will be created for the later design and assessment of the service concept. The result of the service planning phase is a detailed project description of the service development project. It includes fixation of the project’s scale and its main objectives and ultimately the particular instructions for the development team. In the first step of the “service conception”, the specified requirements of the service format, as described by the project specifications, are supplemented with the object definition and the fixation of the object system. Subsequently, firstly the design of (several) detailed possible solutions with respect to the different service dimensions is mapped out and secondly the adjacent evaluation (which is done i.e. by prototyping or service simulation) of the service concept is formulated. The result of the “service conception” stage is an evaluated service conception. It describes the planned objectives (i.e. target group, service quality and range of offer) and gives detailed illustrations regarding the different aspects of the service, both of which are to be implemented in the following steps. Finally, the “service implementation” phase deals with the preparation of the implementation, the testing of the service concept (this is done i.e. by running a pilot project), and the confirmation of improvements, which will be adapted in a feedback loop. Furthermore, the objective of the implementation phase is to control the service concept, which deals with the specified objectives and requirements. If these are not attained, the process step in question has to be repeated until the requirements and goals are fulfilled completely. The result of the service implementation phase is a service format which describes the objectives as well as the tested and fixed operation plan which fulfils the customer requirements according to the service concept. After this step, the service format is

di

se r m vice en si on

market outcome process structure activity dimension

service development process ideas research

situation analysis

requirements definition

objectives definition

development

valuation/ decision

implementation

controlling

planning

conception

implementation

project description

service concept

service format

en pr sc ag tre oj ie gr ec pr n eg tif la en l la t-re i a c ye eu ye ati ye la la r r t r r on ye ia ed l r

[28]. This model puts its focus on the service life cycle phases from idea management to implementation [27]

procedure model, methods, tools

Figure 5: Framework for the classification of service development approaches

26

redirected to the service management. Procedure model, ,methods and tools The general aim of a procedure model for developing services is the description how to structure and manage the complex, multi-disciplinary service development process more efficiently. By combining appropriate methods and tools into defined, single process steps, the procedure model assists service development from the initial idea to the final implementation. This leads to transparency, lucidity, and defined steps. Therefore it ensures high service quality [24]. Methods include definite instructions, which determine activities, in order to reach a certain objective [17] [32]. Tools support the operationalisation of a method [33]. The procedure model defines the single stages and activities of the service development process in a first step. Subsequently, suitable methods and tools have to be chosen for every stage in a second and third step. Methods and tools have to be selected considering the type of service (cf. Figure 2), the service’s complexity and the situation of the company. The latter concerns the range of offered methods and tools as well as the experiences employees have with them [15]. 3.2 Service dimension The service dimension is derived from the constitutive service characteristics. This is done analogous to the single phases of the service. Hence a distinction can be made by analysing the service’s dimensions, i.e. its structure, its process, and its outcome [10] [16] [32]: •

The structure dimension focuses on the provision of services. It describes the basic ability and readiness to deliver a service by combining internal (i.e. employee, technology, information) and external (i.e. customer) factors. • The process dimension focuses on the process of service by integrating external factors. • The outcome dimension describes the (tangible as well as intangible) outcome and its respective impact on the customer. • In addition to this three dimensions mentioned above, which are based on the constitutive service characteristics, Meiren/Barth [34] takes the market dimension into account. This underlines the necessity of considering market requirements to avoid undesirable developments for the customer. 3.3 Aggregation layer Finally, as the third aspect, a distinction can be made by analysing the focus and therewith the aggregation level with regard to a scientific, a company- or a project-related viewpoint (see Figure 6) [15]. The scientific layer deals with the provision of approaches to procedure models, of methods, of tools and of classifications of services as well as of human resource concepts. The entrepreneurial layer reconciles these scientific approaches to the company’s requirements by selecting either a range of suitable methods and tools for the company’s service development system process or by determining the reference models and employee qualification standards. The aim is to configure a holistic service development system. Finally, the project-related layer refers to separate, concrete service development projects with an applicable project plan, with methods, tools etc. Consequently, by linking these three introduced aspects together, a grid emerges, making the systematisation of approaches with respect to different aspects possible. An example is on the one hand the classification of methods regarding the procedure model respectively the single

stages and on the other hand regarding the service dimension. Science field Scientific layer

service typology

procedure model

methods and tools

Service development system Entrepreneurial layer

service program

reference models

methods and tools selection

Single service development project Projectrelated layer

service objects

procedure model

methods and tools utilisation

Figure 6: Aggregation layer [15] [23] 4 REVIEW METHODOLOGY AND RESULTS Approaches for the service development research field with its different foci and structures can be distinguished in approaches for the service development procedure model, methods and tools regarding different criteria. Therefore, starting from the activity dimension, classification is made based on the service development process (Figure 7). Hereby, all reviewed approaches can be differentiated depending on whether the approach considers the whole service life cycle (service life cycle spanning) or only some aspects respectively only one stage (service life cycle phase). Furthermore, the approaches are classified with regard to the service dimensions and the aggregation layer. 4.1 Classification by activity dimension Service development procedure model Many approaches of the service development research field focus on the development of a general, efficient and life cycle spanning procedure model for the service development process [23]. The most important are [25] [27] [29] [35] [36] [37]. Beside this, further important research foci and life cycle spanning approaches have been identified. Thereby, some of this approaches also integrated methods as well as approaches for tool support: •

Approaches which focus on the modularization of the service development process to offer a customised service with a high degree of standardisation with regard to the base of modules. Important authors are [19] [38]. Moreover, standards are defined by DIN [29].



Further approaches deal with innovation-management [39] [40] and knowledge-management [41] for the service development.



Finally, some approaches from the different research fields consider special issues: Marketing and psychology (integration of the internal and external factors with emphasis on the customers integration [42] [43]) and information and computer science (with focus on the tool-supported service development [44]). Furthermore, procedure models can be differentiated according to the following aspects: •

27

Degree of physical product integration: Approaches can be differentiated according to the degree of physical product integration. While the main part of the research in service, as well as the basic models in service development are concerned only with the development of pure services, further research approaches (q.v. [45] [46]) focus on the integration of

of service-specification [17] [31]:

services and products (as in product-related and product-supporting service developments) as well. The technical service development is concerned with the development of product-related services based on existing products [46]. In contrast, an integrated and reciprocal approach which considers planning, developing, producing and using of services and products is known as a hybrid performance bundle. This integrated approach tends to bear utilisationorientated results [47].

Existing methods i.e. from the research field engineering or economy are used in the service development process. Examples are brainstorming, expert interview as well as the different kinds of analysis methods such as the competition analysis, the cost-benefit analysis and the value-benefit analysis.



Modified methods (mainly form the research field engineering), which are based on proved methods and were adapted to the service characteristics (i.e. intangible, uno-acto-principle, integration of an external factor). Examples are Service-FMEA (Failure Mode and Effects Analysis), an analytic method to identify prospective failures in a complex system and thereby remedy failure sources betimes or ServiceQFD (Quality Function Deployment),. a systematic method for quality control in the planning process to interpret customer requirements into objectives in a multistage process [31].

Kind of procedure model: Two kinds of procedure models for the service development process exist: sequential phase models and iterative structured models [14]. Most approaches, which are presented in the service development literature, integrate a model with sequential phases [48], q.v. [17] [25] [27] [29] [35]. Only a few models contain an iterative structure (q.v. [32] [36] [37]) with feedback loops.



Level of detail: The procedure model’s level of detail as it is composed of the number of aggregation levels in the model as well as of the number of stages respectively the number of activities [23]. While on the one hand the model of Fähnrich [17] includes only four phases on one single aggregation level, the model of Meiren [33] on the other hand considers 74 phases on three aggregation levels. These procedure models are supported by suitable life cycle spanning and life cycle phase related methods and tools, which are classified in the following sections.



Service-specific methods, which were specifically developed for service development like Service Blueprinting [49], a service modeling method, which offers transparency with regard to the customer contact intensively in the service process [31], or ServQual [50], an approach for multiattributive measuring of the attitudes and contentment orientation in services by uncovering different kind of gaps in the service offered. Hereby, a study [17] showed that the best known and most used methods in the practitioner’s field are analytical methods from the economic research domain, such as the competition analysis, the cost-benefit analysis, the valuebenefit analysis or the profitability analysis. Less used are engineering methods, such as process modeling, prototyping, FMEA and QFD. Service-related methods, such as gap-analysis [51] and Service-Blueprinting are little known and less utilised. Furthermore the selection and application of methods depends on the service type (cf. Figure 2). Hereby the

Methods In order to pass the steps of the service development process in a goal-oriented way and in order to achieve high-quality services, suitable methods are used in the individual steps. Because of its interdisciplinary background, various methods from different research disciplines are available. Most of them are methods from the fields of economy and engineering. At first, methods for service development can be divided into the three following categories, according to the degree Service life cycle spanning approaches Service life cycle phase related approaches

service development process ideas research

situation analysis

requirements definition

planning service development procedure model

Distinction regarding: • kind of procedure model • level of detail • physical product integration

Distinction regarding: • service type • degree of servicespecification

development

valuation/ decision

conception

implementation

controlling

implementation

general procedure model Innovation management, knowledge management, Customer Relation Management Modularisation creativ methods (i.e. brainstorming)

methods

objectives definition

methods for requirement analysis modelling, simulation and assessment methods (i.e. Service-FMEA, Service-Blueprinting) (i.e. Service-QFD)

analytical methods (i.e. market analysis, cost-benefit analysis, value-benefit analysis, gap analysis, monitoring) Project management methods, quality management methods (i.e. ServQual)

tools

Workflow management, project-management, knowledge management, Distinction regarding: documentation management, communication management (i.e. MS Project, Lotus Notes, DocuWare) • content- or functional supported tool Modelling and simulation tools Planning-supporting tools • service type (i.e. ARIS, Visio) (i.e. mind mapping tool, database solution) • degree of service-specification Holistic, service-specific service engineering tools (i.e. CASET, ServCASE)

Figure 7: Classification by activity dimension

28

for all service dimensions and aggregation layers





mapping tools as well as database solutions (i.e. SAP R3, Data Warehouse). The conception phase is supported by modeling tools such as ARIS and Visio.

contact intensity has a significant impact [16]: Low contact intensity: Service methods, stemming from the conventional physical product development process, are used. A possible explanation for their application might be seen in the similar character of this type of services and of a physical product because of the latter’s low contact intensity.



High contact intensity: On the other hand, engineering methods are less relevant for this type of services. Due to the fact of high integration of the external factor (mainly the customer), economic and service-specific methods, but also social and behavioural aspects accentuating the customer interaction, are brought into focus. Furthermore, the methods can be differentiated with regard to the life cycle phase and their objectives as well as to the different service dimensions. Hereby, the methods have to be select regarding the specific character of the service respectively the service classification (cf. Figure 2) [14] •

Service planning: For example creative methods like brainstorming or morphological box are used for the idea generation. Methods for requirements analysis, such as customer survey, studies and monitoring, and analytical methods, such as the competition analysis, the cost-benefit analysis, or the value benefit analysis are applied for the situational and requirements analysis.



Service conception: Modeling methods (i.e. Service Blueprinting, Event-driven process chain), simulative approaches and assessment methods are used in the conception phase.



Service implementation: In this stage methods for the testing (i.e. pilot projects) and controlling are necessary. Due to this, in this phase also analytical methods such as customer survey, gap-analysis, monitoring and again competition analysis, the costbenefit analysis are used as well as simulation methods for the testing of services.



Tools, which give a cross-sectional support, such as project management tools (i.e. MS Project), communication and groupware tools (i.e. Lotus Notes, MS Office), and knowledge- and documentation tools (i.e. DocuWare). Existing service engineering tools have the problem that they can be considered as isolated applications rather than solutions, offering an integrated support. Until now, there is a lack (in practise) of service life cycle spanning tools, which support the service development process through the whole service life cycle in every step [23]. The essential task of a holistic service engineering tool can be identified in the area of knowledge management, the methodological integration of existing tools, as well as the monitoring of service development projects [52]. Therefore, current researches focus on the development of a holistic, service-specific, engineering tool for the support of the service development process. First theoretical approaches are shown in [45] [52]. Furthermore, concepts and approaches for engineering tools are developed in the research projects CASET (Computer Aided Engineering Tool) and ServCASE (Computer Aided Engineering für IT-basierte Dienstleistungen) [15]. 4.2 Classification by Service Dimension According to chapter 3.2, the four service dimensions (structure, process, outcome and market) have to be considered for a systematic design of services in every step of the development process [31]. An example for a service development process in the automotive industry is given in Figure 8. Activity dimension (procedure model, methods, tools) service development process ideas research

Tools which give a content-oriented support for the core functions of the service development process. Examples for the planning phase are planningsupporting tools like MS Office tools and mind-

situation analysis

requirements definition

planning

objectives developdefinition ment conception

valuation/ impledecision mentation

controlling

implementation

resource

Finally, the whole service development process is supported by life cycle spanning methods such as general project management methods (i.e. time and work scheduling or team coordination) and quality management methods.

Tools Tools support the operationalisation of a method [33]. Therefore, tools can, as well as methods, be applied for single phases or for the whole service development process. Furthermore, the selection of tools also depends on the type of service, according to the supported method. In the context of service development, tools are defined as information and communication tools which provide functions that support the steps of the service development process [31]. Thus, most of the tools stem from the field of information science or computer science. Examples are tools for business process management, analysis and simulation as well as modeling and project management [32] When comparing the usage of modern development tools for products or software to the domain of services, discrepancies become visible. If any, software tools in the field of service engineering offer a more functional than content-oriented support. Therefore, Corsten [9] distinguished between •



process

outcome

market

Figure 8 : Consideration of the different service dimensions in the service development process (q.v. [34]) Due to that, also appropriate methods and tools have not only assigned to the single steps but also to the associate service dimensions [16] [32]. The methods can be assigned to one or more of the four service dimensions. While on the one hand i.e. methods for the capacity planning, information system planning and employee qualification serve for the resource dimension, on the other hand i.e. methods for process modeling are used for the process dimension. Beyond, many methods consider several service dimensions or even all dimensions. Most of the important service development procedure models include the differentiation of the described service dimensions and include all of them [32]. An integrated framework for the comprehensive illustration of a service concept based on resource model, process model and outcome model is shown in [23].

29

for all aggregation layers



4.3 Classification by aggregation layer This paper focuses on approaches belonging to the scientific layer. Therefore, a detailed classification, regarding the three layers was not made in this paper, but the following distinction: The procedural models can be distinguished depending on whether they are more generic (q.v. [25] [29]) or specialised for certain kinds of services, such as financing (q.v. [53]) or health services (q.v. [54]) [23]. Due to that, these mentioned approaches can be linked to the scientific layer respectively to the project-related layer. 5 SUMMARY A classification of approaches (procedure models, methods and tools) for the service development process was done on the basis of a proposed framework. This framework distinguishes between three aspects: activity dimension, service dimension and aggregation layer. The aim thereby was to render a concise overview and to provide transparency for concerned organisations. Further research opportunities can be found in the mapping of the individual approaches to the proposed framework. Based on these results gaps in the research fields can be identified and discussed. 6 REFERENCES [1] Statistisches Bundesamt (ed.), 1995, Statistisches Jahrbuch 1995 für die Bundesrepublik Deutschland, Wiesbaden. [2] Statistisches Bundesamt (ed.), 2006, Statistisches Jahrbuch 2006 für die Bundesrepublik Deutschland, Wiesbaden. [3] Statistisches Bundesamt (ed.) 2007, Statistisches Jahrbuch 2007 für die Bundesrepublik Deutschland, Wiesbaden [4] Cooper, R.G., Edgett, S.J., 1999, Product development for the service sector: lessons from market leaders, Perseus, Cambridge, MA. [5] Scheer, A.-W., Service Engineering bringt die Prozesse auf Vordermann, VDI-Nachrichten, 56 (2002d), No. 35: 2. [6] Bullinger, H.-J., Entwicklung innovativer Dienstleistungen, Bullinger, H.J. (ed.), Dienstleistungen – Innovation für Wachstum und Beschäftigung, Herausforderungen des internationalen Wettbewerbs, Wiesbaden, Gabler, 1999: 49-65. [7] Kleinaltenkamp,C., 2001, Begriffsabgrenzungen und Erscheinungsformen von Dienstleistungen, Bruhn, M., Meffert, H. (ed.), Handbuch Dienstleistungsmanagement. Von der strategischen Konzeption zur praktischen Umsetzung, 2nd ed., Wiesbaden: 27-50. [8] Haller, S., 2005, Dienstleistungsmanagement – Grundlagen – Konzepte – Instrumente, 3rd edition, Wiesbaden. [9] Corsten, H., Gössinger, R., 2007, Dienstleistungsmanagement, 5th edition, München. [10] Meffert, H., Bruhn, M., 2005, Dienstleistungsmarketing: Grundlagen, Konzepte, Methoden, 5th edition, Meffert-Marketing-Edition, Gabler, Wiesbaden. [11] Ascher, B., Whichard, O. G., Improving services trade data,: Giarini, O. (ed.): The emerging service economy, Pergamon, Oxford: 255–281. [12] Langeard, E. et al.: Services marketing: new insights from consumers and managers, Marketing Science Institute, Cambridge,MA, Report No.81-104.

30

[13] Altenburger, O. A., Ansätze zu einer Produktionsund Kostentheorie der Dienstleistungen, Duncker & Humblot, Berlin, Betriebswirtschaftliche Schriften, Heft 101. [14] Bullinger, H.-J., Meiren, T., 2001, Service Engineering – Entwicklung und Gestaltung von Dienstleistungen, Bruhn, M., Meffert, H. (ed.), Handbuch Dienstleistungsmanagement. Von der strategischen Konzeption zur praktischen Umsetzung, 2nd edition, Wiesbaden: 149-175. [15] Fähnrich, K-.P., Opitz, M., 2006, Service Engineering – Entwicklungspfad und Bild einer jungen Disziplin, Bullinger, H.-J., Scheer, A.-W., Schneider, K. (Ed.) 2006, Service Engineering – Entwicklung und Gestaltung innovativer Dienstleistungen, Berlin: 85-112. [16] Fähnrich, K-.P., Meiren, T. 2006, Service Engineering: State of the Art and Future Trends, Spath, D., Fähnrich, K.-P., Advances in Service Innovations, Berlin. [17] Fähnrich, K.-P., Meiren, T., Hertweck A., Baumeister, M., Demuß, L., Gaiser, B., Zerr, K., 1999, Service Engineering, Ergebnisse einer empirischen Studie zum Stand der Dienstleistungsentwicklung in Deutschland, IRB, Stuttgart. [18] Husen, C. van, 2007, Anforderungsanalyse für produktbegleitende Dienstleistungen, Jost-Jetter Verlag, Heimsheim. [19] Burr, W, 2002, Service Engineering bei technischen Dienstleistungen – Eine ökonomische Analyse der Modularisierung, Leistungstiefengestaltung und Systembündelung, Wiesbaden. [20] Zeithaml, V. M., Bitner, M. J., 2006, Services marketing - integrating customer focus across the firm, 4th edition, McGraw-Hill, Boston, MA. [21] Demuß, L., 2003, Ein Reifemodell für die Bewertung und Entwicklung von Dienstleistungs-organisationen - das Service Management Maturity Model (SMMM), Karlsruhe, WBK – Forschungs-berichte aus dem Institut für Werkzeugmaschinen und Betriebstechnik der Universität Karlsruhe, volume 117. [22] Hofmann, H. R., Klein, L., Meiren, T., 1998, Vorgehensmodelle für das Service Engineering, IM – Die Fachzeitschrift für Information Manage-ment & Consulting, 13, special edition ‘Service Engineering’: 20-25. [23] Klein, R., 2007, Modellgestütztes Service System Engineering – Theorie und Technik einer systematischen Entwicklung von Dienstleistungen, Wiesbaden. [24] Schneider, K., Daun, C., Behrens, H., Wagner, D., 2006, Vorgehensmodelle und Standards zur systematischen Entwicklung von Dienstleistungen, Bullinger, H.-J., Scheer, A.-W., Schneider, K. (ed.) 2006, Service Engineering – Entwicklung und Gestaltung innovativer Dienstleistungen, Berlin: 113-139. [25] Ramaswamy, R., 1996, Design and Management of Service Processes – Keeping Customers for Life, Reading. [26] Luczak, H., Sontow, K. et al., 2000, Service Engineering, Der systematische Weg von der Idee zum Leistungsangebot, Tcw-report, München. [27] Edvardsson, B., Olsson, J., 1996, Key Concepts for New Service Development, Service Industries Journal, 16, 2: 140-164. [28] Bullinger, H.-J., Scheer, A.-W., 2006, Service Engineering – Entwicklung und Gestaltung innovativer Dienstleistungen, Bullinger, H.-J.,

[29]

[30] [31]

[32]

[33]

[34]

[35]

[36] [37]

[38]

[39]

[40]

[41]

Scheer, A.-W., Schneider, K. (Ed.) 2006, Service Engineering – Entwicklung und Gestaltung innovativer Dienstleistungen, Berlin: 3-19. DIN, Deutsches Institut für Normung e.V. (ed.), 1998, Service Engineering: entwicklungsbegleitende Normung (EBN) für Dienstleistungen, Berlin. Shostak, L., 1984, Designing Services that deliver, Harvard Business Review 62: 73-78. Eversheim, W. et al, 2006, Anwendungspotential ingenieurwissenschaftlicher Methoden für das Service Engineering, Bullinger, H.-J., Scheer, A.-W., Schneider, K. (Ed.) 2006, Service Engineering – Entwicklung und Gestaltung innovativer Dienstleistungen, Berlin: 423-442. Bullinger, H.-J., Schreiner, P., 2006, Service Engineering, Ein Rahmenkonzept für die systematische Entwicklung von Dienstleistungen, Bullinger, H.-J., Scheer, A.-W., Schneider, K. (ed.) 2006, Service Engineering – Entwicklung und Gestaltung innovativer Dienstleistungen, Berlin: 53-84. Meiren, T., 2000, Management der Dienstleistungsentwicklung, Spath, D., Bullinger, H.-J., Demuß, L. (ed.), Service Engineering 2000 : Entwicklung und Gestaltung innovativer Dienstleistungen, WBK/IAO/VDLI, Karlsruhe, Stuttgart : 79-106. Meiren, T., Barth, T., 2002, Service Engineering in Unternehmen umsetzen Leitfaden für die Entwicklung von Dienstleistungen, Stuttgart. Scheuing, E.E., Johnson, E.M., 1989, A proposed model for new service development, Journal of Marketing, 3, no. 2: 25-34. Jaschinski, C., 1998, Qualitätsorientiertes Redesign von Dienstleistungen, Aachen. Shostack, G.L., Kingman-Brundage, J., 1991, Congram, C.A., Friedman, M.L. (ed.), The AMA handbook of marketing for the service industries. New York (NY), American Marketing Association: 243-261. Böhmann, T., 2004, Modularisierung on ITDienstleistungen: Eine Methode für das Service Engineering, Wiesbaden. Meyer, A., Dienstleistungs-Innovation, 1998, A. Meyer (ed.): Handbuch Dienstleistungs-Marketing, Stuttgart 1998. Reichenwald, R., Schaller, C., 2006, Innovationsmanagement von Dienstleistungen – Herausforderungen und Erfolgsfaktoren in der Praxis, Bullinger, H.-J., Scheer, A.-W., Schneider, K. (Ed.) 2006, Service Engineering – Entwicklung und Gestaltung innovativer Dienstleistungen, Berlin: 167 – 194. Kleinaltenkamp, M. ., Schaller, C., 2006, Innovationsmanagement von Dienstleistungen – Herausforderungen und Erfolgsfaktoren in der Praxis, Bullin-

[42]

[43] [44]

[45]

[46]

[47]

[48]

[49]

[50]

[51]

[52]

[53]

[54]

31

ger, H.-J., Scheer, A.-W., Schneider, K. (Ed.) 2006, Service Engineering – Entwicklung und Gestaltung innovativer Dienstleistungen, Berlin: 3-19. Reckfelderbäumer, M., Busse, D., 2006, Kundenmitwirkung bei der Entwicklung von industriellen Dienstleistungen – eine phasenbezogene Analyse, Bullinger, H.-J., Scheer, A.-W., Schneider, K. (Ed.) 2006, Service Engineering – Entwicklung und Gestaltung innovativer Dienstleistungen, Berlin: p. 141 – 166 . Scheer, A.-W., Spath, D., Computer Aided Service Engineering, Berlin Kallenberg, R., 2002, Ein Referenzmodell für den Service in Unternehmen des Maschinenbaus, Aachen Meier, H., Sadek, K., 2008, Industrial ProductService-Systems – Integrated Development of Products and Services, 15th Cirp International Conference on Life Cycle Engineering. Spath, D., Demuß, L., Entwicklung hybrider Produkte – Gestaltung materieller und immaterieller Leistungsbündel, Bullinger, H.-J., Scheer, A.-W., Schneider, K. (ed.) 2006, Service Engineering – Entwicklung und Gestaltung innovativer Dienstleistungen, Berlin: 463-502. Meier, H., Kortmann, D., Golembiewski, M., Hybride Leistungsbündel in kooperativen Anbieter-Netzwerken, Industriemanagement, 4/2006, Berlin: 2528. Daun, C., Klein, R., 2004, Vorgehensweise zur systematischen Entwicklung von Dienstleistungen im Überblick, Scheer, A.-W., Spath, D. (ed.) : Computer Aided Service Engineering : Informationssysteme in der Dienstleistungsentwicklung, Berlin: 43-67. Shostak, L., 1995, Service Blueprinting, Heilbrunn, J. (Ed.), Marketing encyclopedia ; issues and trends shaping the future, American Marketing Association, Chicago, IL : 107-113. Parasuraman, A., Zeithaml, V.M.; Berry, L.L.: SERVQUAL: a multiple-item scale for measuring consumer perceptions of service quality. In : Journal of Retailing, 64 (1988), Nr. 1: 5-6. Zeithaml, V., Parasuraman, A., Berry, L. L. ,1992, Qualitätsservice – Was Ihre Kunden erwarten – was Sie leisten müssen, Frankfurt/ Main 1992. Gill, C., 2004, Architektur für das Service Engineering zur Entwicklung von technischen Dienstleistungen, Aachen. Donnelly, J. H., Berry, L. L., Thompson, T. W. 1985, Marketing financial services: a strategic vision, Homewood (IL), Dow Jones-Irwin. Bowers, M. R., 1987, Developing new services for hospitals: a suggested model, Journal of Health Care Marketing, 7, Nr. 2: 5-44.

A Product-Service System Representation and Its Application in a Concept Design Scenario Y. S. Kim, E. Wang, S. W. Lee, Y. C. Cho Creative Design Institute, Sungkyunkwan University, 300 Chunchun, Jangan, Suwon, 440-746, Korea [email protected]

Abstract Product-service systems (PSS) can give diverse value provision to consumers reflecting their individual needs, while also addressing multiple issues from manufacturers’ viewpoints. We propose graph and ontological representations of PSS, consisting of values, product and service elements, and their relations. PSSs may also be included as subsystems in a larger PSS. We illustrate a case scenario of PSS concept design starting from an existing product. Diverse requirements of stakeholders of the product life-cycle are transferred to persona generation. From state parameters of these personas, we identify operations to improve the personas’ values, and implement each operation as a sub-PSS. Keywords: PSS Design, PSS Representation, Case Scenario, Visualization and Manipulation of PSS Concept Design

1 INTRODUCTION In the past decade, product-service systems (PSS) have received much research effort as a means of innovative value proposition through the integration of products and services. The European Union (EU) has sponsored various projects on PSS for sustainable consumption and production [1]. Considerable research has also been done in developing the tools and methodologies for effective development of PSS, as reviewed by Baines et al. [2]. Neely has studied the potential advantages of PSS [3], and noted that industrialized countries have shown higher numbers of combined manufacturing and service firms than other countries. The concept of PSS was first introduced in 1999 by Goedkoop et al. to address the challenges of environmental and economical issues [4]. They defined PSS as a marketable set of products and services, jointly capable of fulfilling a client's need. In addition, they addressed several advantages of PSS such as: creation of value for clients through quality and comfort; customization of offers; delivery of offers to clients; decrease in the cost of initial investment through sharing, leasing and hiring; decrease in environmental load; etc. Mont also defined PSS as a system of products, services, supporting networks, and infrastructure that is designed to be competitive, to satisfy customer needs, and to have a lower environmental impact than traditional business models [5][6]. He proposes a theoretical framework for PSS reflecting societal infrastructure, human structures and organizational layouts to enhance environmental values. Manzini and Vezolli present a strategic design approach for sustainability by describing potential benefits of PSS with some examples of eco-efficient PSS [7]. In the area of PSS design, Morelli has studied a methodological framework that considers designers’ views [8]. He presents a concrete case study of an urban telecenter, in which the major functions and requirements are first extracted, and then linked to the elements of products and services for the development of a PSS. Aurich et al. researched the life-cycle oriented design processes of products and services [9]. They proposed a systematic

CIRP IPS2 Conference 2009

32

design process for technical services associated with products, which is later integrated with the product design process. They also introduced the concept of process modularization for integration of product and service design processes by selecting, combining and adapting appropriate process modules [10]. Matzen and McAloone introduced the activity modeling cycle (AMC) model [11] as a tool for conceptualizing the development of PSS, and investigated the effectiveness of the AMC model by conducting a case study on service delivery in the container ship industry. More recently, they describe a structured modeling scheme to differentiate and categorize development tasks during a migration toward service orientation, with a case study of the maritime equipment industry [12]. Shimomura et al. have contributed significant research on service engineering [13][14][15][16][17]. They propose the use of receiver state parameters (RSPs) as the underlying representation of values and costs to be managed throughout the service design process. They introduce the service model, which includes the sub-models of flow model, scope model, view model and scenario model. They have also implemented a prototype computer-aided design tool for service design called Service Explorer. Maussang et al. have developed a PSS design process by incorporating Shimomura’s service model into an engineering product design process, considering functional analysis and agent-based value design [18]. Maussang et al. have also proposed the evaluation of PSS during the early design phase using an optimization methodology, considering economical and environmental factors [19]. Meier and Völker identified challenges and opportunities in adapting existing product supply chain techniques to support service supply chains, and illustrated a scheme to maintain supply chain networks autonomously through the application of multi-agent systems [20]. Although considerable research for the effective design of PSS have been conducted, there has not yet been an equivalent effort in developing a comprehensive, machineunderstandable representation of PSS itself and its elements, including values, product elements, service

elements, and their relations. To properly support the development of new computer-aided tools and frameworks for PSS design, we see a need for a formal ontological representation of PSS. This paper is part of an ongoing research effort toward the implementation of an intelligent PSS design framework, which will be an integrated development environment for the PSS design process itself. In this paper, we propose an ontological representation of PSS, which will form a core part of the data model for the PSS design framework. Section 2 describes our approach for conceptual PSS design, using the case scenario of a PSS for a meal assembly kitchen. It also introduces a graph representation of PSS. Section 3 presents an ontological representation of PSS, including product and service elements, their associated values, and relations between them. In Section 4, we illustrate how this PSS ontology could be used to model one sub-PSS of the meal assembly kitchen scenario. 2

GENERATION OF PRODUCT-SERVICE SYSTEMS

2.1 Generation of PSS from a Product A PSS includes several product elements and service elements that are closely related to other. A schematic diagram of the generation of a PSS from a single product is shown in Figure 1. A product P is sub-divided into its constituent product elements Pi. To these, we may add new product elements Pi, and identify new values Vk. Then, new service elements Sj are added and combined with these product elements so that all values are achieved. The resulting system is a PSS. P12 P10

s

V1

P2

P

P5

P12

P11

V3

Product

P4 P1 P3

V2

P5

P8 P13

P

P1

P P4

P2 P6

P7

P3

s

P

s

s

Product-Service Systems

Figure 1. Generation of PSS 2.2 Case Scenario for Conceptual Design of PSS In this section, we illustrate our approach to the conceptual design of a PSS by presenting a case scenario. Starting from an existing product, which is a typical kitchen oven, we will generate a PSS for a meal assembly kitchen. A meal assembly kitchen is an innovative concept in meal preparation that moves the meal assembly process out of people's kitchens and into specially equipped stores. It offers the benefits of simplified menu planning, support and instruction in cooking, and elimination of shopping, food preparation, and cleaning. Other social benefits include increased opportunities to engage in mutual family activities and make new friends. We develop the conceptual design case scenario by the following steps. First, we generate the user requirements of an oven from the oven’s life cycle. We identify stakeholders throughout the oven’s life cycle, and by considering their requirements and needs, we extract stakeholders’ values. Second, from these values, we generate several personas, where each persona represents a target group of users that share one set of values. Third, we identify operations that could improve customers’ values, and model each operation as a subPSS of product and service elements that achieve the value improvement. Finally, these sub-PSSs are merged

and expanded to obtain the complete PSS for the meal assembly kitchen. These steps are further explained in the following subsections. Proposition of Values and Requirements We adopt Matzen and McAloone’s activity modelling cycle (AMC) model [11] to represent stakeholders and their activities and values. In accordance with the AMC approach, we divide the entire life cycle of the oven into three phases: pre-usage, during usage, and post-usage. In each phase, we list activities and associated stakeholders, values and requirements. The pre-usage phase includes the production, selling, buying, delivering and installation of the oven itself. The during-usage phase includes normal cooking and cleaning activities. The post-usage phase includes removal of the oven, its disposal (including recycling), and elimination of traces that the oven may leave behind in the environment, such as odors. For each activity, we obtain the values and requirements associated with the stakeholders through usage observation, interviews and surveys. Persona Generation A persona is an abstraction of a target user who should be satisfied with the designed PSS; hence it represents a group of users, all of whom are characterized by the same set of values. The set of personas is chosen to cover a significant portion of the intended customers of a PSS. Each persona has diverse and different values, which may be positive or negative. PSS design then proceeds by devising ways to improve the negative values of all personas. In this case scenario, we have developed three personas: an elderly woman, a housewife who is recently married, and a businessman who lives alone. For example, the businessman persona has one positive value: he knows how to cook with the oven. However, he has four negative values. First, he possesses a second-hand oven that needs frequent attention and maintenance. Second, he does not have proper oven maintenance skills. Third, he has insufficient kitchen space for the oven. Fourth, shopping for ingredients is a burden in distance and time. These four negative values become the targets for value improvement, described next. Value Improvement and PSS Generation By considering the negative values defined for all personas, we devise a set of operations that collectively address and overcome all of the negative values. For the case scenario, we devise three operations: provision of chef’s advice, provision of wide space and several ovens, and management of ovens and ingredients. Each operation generally improves a different value, and demands different product and service elements. We implement these three operations as three sub-PSSs, shown in Figure 2, using a graph representation of PSS. Each PSS graph is composed of product elements, service elements, value nodes, and edges. The edges denote the relations among the nodes. This PSS graph representation can quickly show how many product elements, service elements and values exist, and whether any two elements have some relation. For example, the operation of Provide wide space and several ovens is shown as a sub-PSS in Figure 2, lower left. Two values, V21 wide space and V22 variety of ovens, are identified. These are related to three product elements: P1 wide space, P22 various ovens, and P23 utensils, and two service elements: S21 provide space for cook and S22 provide tools for cook. These element nodes are joined by edges to make this sub-PSS. These three sub-PSSs can be expanded by linking them with other sub-PSSs to make the complete PSS, as shown

33

in Figure 3, using the graph representation. The expansion of the sub-PSS can be made to link the values, product and service elements in the sub-PSS and those in other subPSSs. For example, consider the elements P22: various ovens and S22 provide tools for cook. Since we need

many tools for cook such as a burner, micro-oven, mixer, and so on, P22 and S22 can be associated with the subPSSs including these tools as elements. In this way, the sub-PSSs are linked together and expanded to make the complete PSS for the meal assembly kitchen.

 V11 : proper cooking methods  V12 : correct usage of Oven  V13 : nutrition information Provide advice of chef  V14 : using the right utensils  P11 :  P12 :  P13 :  P14 :  S11 :  S12 :  S13 :  S14 :

chef cook book cooking instruction media various utensils

V13

S 13 P12 V11

V12

P11 P13

Provide cooking methods Provide usage of Oven Provide nutrition information Provide utensils

S11

P14

 V31 : various ingredients  V32 : fresh ingredients  V33 : tools management

S12 S14

V14

V31

S32 P32 S31

P31

 V21 :  V22 :  P21 :  P22 :  P23 :

wide space for cook various Ovens for cook wide space various Ovens S22 utensils

 S21 : Provide space for cook  S22 : Provide tools for cook

V22

P23

V32

P21

P35  S31 : buying ingredients

S33

S21 P33

P22

P34

V33

V21

 P31 : ingredients  P32 : display stand for ingredients  P33 : tools for cleaning  P34 : tools for maintenance  P35 : tools that need repair  P36 : engineer for fixing/checking

P36

 S32 : keeping ingredients  S33 : managing tools

Provide manager for Ovens/ingredients

Provide wide space, several Ovens

Figure 2. Sub-PSS Graphs

Figure 3. Expanded PSS Graphs

34

3 ONTOLOGICAL REPRESENTATION OF PSS The graph representation of PSS, as shown in the preceding sections, provides a simple, visual way to relate entities to each other. This representation could be useful for some applications. However, it omits considerable detailed data. Other non-graph perspectives of PSS data could also be useful for different applications, e.g. tabular representations may be more suitable for editing and sorting. To represent all data of a PSS in a format that could support future automated reasoning applications, we present an ontological representation of a PSS. Our PSS model describes values, product elements, and service elements, and their relations, as shown in Figure 4. We present the ontology in UML format, while we have also modeled it in OWL using Protégé, with conversion to Jess. 3.1 Representation of Value What is Value? A key idea in our approach for PSS representation, in both graph and ontological representations, is to model values explicitly as first-class elements, i.e. at the same level of abstraction as for product elements and service elements. “Value” has been variously defined or used in previous research; hence we first clarify exactly what we mean to be a value. We take the definition of value as used in

economics: value is the market worth or estimated worth of products or services [21][22]. We also take Maussang et al.’s approach that values within a PSS are deduced from the needs of stakeholders [18]. Examples of Value in PSS Human stakeholders are very flexible in perceiving different kinds of value within a PSS; hence, we need a flexible representation. Some examples of values in PSS include availability of X, condition of X, and usage rights for X, where X = bicycle [18]. Similarly, in our meal assembly kitchen scenario, we have identified values such as (availability of) V21: wide space for cook, as shown in Figure 2. X could be any resource, which may be a tangible object, but it is composed with a discriminator such as availability of to realize the value. That is, X is not, by itself, a value. Ontological Representation of Value We model a Value class, and its component classes, as shown in Figure 4, left side. A Value object has one ValueNature discriminator, such as AvailabilityOf, which indicates that this value is related to the availability of a resource. “Available” implies numerous conditions: that the resource exists, is within easy reach, is in working condition, the receiver has adequate permissions, the receiver must actively get the object himself, etc.

Figure 4. Ontological Representation of Value and PSS

35

The ValueNature object is composed with one IValueCategory, which could describe a tangible resource (such as ingredients), or intangible resource (such as cooking methods). It also has 0 or more Constraints. Multiple Realizations of Value A Value could have an absolute (non-subjective) measurement, such as a time in seconds. However, different stakeholders could perceive or appreciate the value differently. To support many different subjective interpretations of a Value by different actor groups, we introduce a ValueRealization class to represent one specific instance of an actor group and its valuation of the Value. A Value then has any number of ValueRealizations.

Each ValueRealization consists of an IActorGroup, a StateParameter, and an IValuation that specifies how this group appraises the value. Each state parameter could be quantitative, qualitative, or both. We define one specific kind of valuation as a Distribution subclass, which presumes that its state parameter is quantitative, and defines a mathematical function of state value to the actors’ degree of satisfaction. While this has some similarity to the approach of Shimomura et al. [15][16], we explicitly model the notions that a Value can have multiple subjective interpretations, and could be nonquantitative, or could have valuations that aren’t expressed as distributions.

Figure 5. Example of Value with Multiple Conditions and Realizations An example of Value modelling, including multiple constraints and realizations, is shown in Figure 5. We choose the value of (availability of) V11: cooking methods from the meal assembly kitchen scenario in Figure 2. It defines two constraints that refine the availability: constraint01 denotes that, during certain hours only, a chef is present to provide cooking methods in a highly desirable way; while constraint02 indicates that, for other the times, only videos are provided. We model two realizations of this value, using the businessman and housewife personas of Section 2.2. They both share the same RSP, but define different distributions, reflecting different personal valuations of the cooking methods. For this example, we could say that the businessmen do not know how to cook, so they greatly appreciate every cooking method; but housewives are more familiar with

36

cooking, so they already know many of the cooking methods, and they react with boredom. 3.2 Representation of PSS Our ontological model of PSS is shown in Figure 4, upper right. A PSS is represented as a class which aggregates three constituent classes, representing a set of Values, a set of PSElements, and a set of Relations of this PSS. Furthermore, PSS uses a notion of an abstract base class that includes function, behavior, structure, context, and environment attributes, which is named FBSCE. PSS also has a subPSSs relation to 0 or more other PSSs, which allows a PSS to be composed from subPSSs in a recursive manner.

Element subclasses The PSElement class is an abstract base class, whose subclasses represent the elements of a PSS. It conveniently extracts common operations from its subclasses, which simplifies numerous algorithms. PSElement also inherits function, behavior, structure, context, and environment attributes from the FBSCE base class. •

PElement (product element) describes a product design.



SElement (service element) describes a service, including its provider and receiver roles.

Values in PSS Values can be associated with product elements, service elements, and PSSs themselves. To ensure a consistent interface for accessing the Values associated with any object, we define an abstract interface IHasValues, which carries a hasValues property to 0 or more Value objects. We then add IHasValues as an additional base class of both PSS and PSElement classes, using multiple inheritance. This provides the standard benefits of interface inheritance: (a) it ensures that both PSS and PSElement inherit the same property, and (b) it allows algorithms to access all PSS, PElement, and SElement objects in a uniform manner. Relation subclasses Relation represents a relation between two elements, or between a value and an element. It corresponds to an edge between two nodes in the simpler graph representation of PSS. We define three specific subclasses, in order to exploit exact type information in modeling these subclasses. •

PVRelation represents a relation between a product element and a value.



SVRelation represents a relation between a service element and a value.



PSRelation represents a relation between a product element and a service element.

Types of Relations Borrowing UML terminology for relationships, we use supplier to denote the value element of a P-V or S-V relation, or the product element of a P-S relation, and consumer to denote the other element. We characterize the following types of relations, based on Shimomura et al.’s terminology for service categories [13]. •

Enable. An “enabling service” is one that “makes the receiver easily achieve its aim”. We generalize it to indicate that the supplier element satisfies a strong requirement, necessary dependency, or prerequisite of the consumer element.



Enhance. An “enhancement service” is one that “helps, supports, or enhances the achievement” of the receiver’s aim. We generalize it to indicate that the supplier element improves or otherwise contributes to the consumer element, but is not a requirement for it.



Proxy. A “proxy service” is one where an agent performs an activity on behalf of the receiver. We take the meaning that the supplier element performs an activity for the consumer element.

4 CASE SCENARIO: MEAL ASSEMBLY KITCHEN We apply our PSS ontology to the meal assembly kitchen scenario, as described in Section 2.2. For this example, we focus on the second of three sub-PSSs in Figure 2, which is labeled “Provide wide space, several ovens”. This sub-PSS is shown again in Figure 6, with the dashed circles elided for clarity.

V22

P23 P21

S22

S21

P22 V21

Figure 6. Sub-PSS of Meal Assembly Kitchen (Graph Representation) In the simpler graph representation of this sub-PSS, the edges highlight the relations between the elements, but do not provide any attribute or other information, such as the types of the relations. Using our PSS ontology, we model this same sub-PSS as shown in Figure 7. Each Value node is represented in more detail, with a greater degree of self-documentation. Edges are modeled as PVRelation, SVRelation, and PSRelation objects, as appropriate. The types of the relations are explicitly represented as enable or enhance types. The sub-PSS itself is explicitly represented as a PSS object named PSS-B, shown at the top of Figure 7. PSS-B has five hasPSElements relations and two hasValue relations, as shown. It also has eleven hasRelations relations, but these are omitted for clarity. From this example, we obtain evidence that our PSS ontology is sufficiently complete to represent the subPSSs of this case scenario, and by extension, any PSSs of similar complexity. By construction, it is no less informative than the graph representation. Through additional model instantiation steps, we can model the complete PSS for the meal assembly kitchen example, which combines all three sub-PSSs. Currently, we perform the model instantiation manually, using Protégé. While this task is straightforward, it is also low-level and thus tedious, since standard tools such as Protégé don’t provide high-level support tailored to our PSS ontology.

37

hasPSElements PSS-B

V22 P23-V22

hasRealization

type = enhance

availabilityOf

P23 Utensils

hasPart oven

S22-V22 type = enhance

S21-V22 type = enhance

P23-V21 type = enhance

S22 Provide tools for cook

S21-P21 type = enable

S22-P23 type = enable P22-V22 S22-P22

type = enable

S21 Provide space for cook

P21 Wide space

type = enable hasValues

S21-V21

P21-V21

type = enable

type = enable

P22 Various ovens V21 P22-V21

availabilityOf hasRealization hasPart

type = enhance

Figure 7. Sub-PSS of Meal Assembly Kitchen (Ontology Model)

5 SUMMARY AND FUTURE WORK We have described an approach for conceptual design of PSS, which includes the following steps: proposition of values and requirements, persona generation, value improvement, and PSS generation. We illustrate this approach with a case scenario, in which we develop a PSS for a meal assembly kitchen. We also present graph and ontological representations of PSS. A key idea in our representation is that values are first-class elements, and that they, and their relations with other elements, are explicitly represented in both the graph and ontological representations. We have shown that our ontological representation is sufficient to model a sub-PSS obtained from the case scenario. In future work, various portions of our ontological representation can be further expanded. In particular, the ValueNature, IValueCategory, and IValuation class hierarchies will be greatly enriched with additional subclasses. A major future direction is to implement automated reasoning services using this ontology, e.g. by converting the ontology and a PSS model to Jess. This can support numerous kinds of validation and checking computations to assist designers and other users in various tasks within a framework for PSS design.

38

space

REFERENCES [1]

[2]

Tukker, A., and Tischner, U., 2006, Product-Services as a Research Field: Past, Present and Future. Reflections from A Decade of Research, Journal of Cleaner Production, 14: 1552–1556. Baines, T. S., Lightfoot, H. W., Evans, S., Neely, A., Greenough, R., Peppard, J., Roy, R., Shehab, E., Braganza, A., Tiwari, A., Alcock, J. R., Angus, J. P., Bastl, M., Cousens, A., Irving, P., Johnson, M., Kingston, J., Lockett, H., Martinez, V., Michele, P., Tranfield, D., Walton, I. M., and Wilson, H., 2007, State-of-the-Art in Product-Service Systems, Proc. IMechE, Journal of Engineering Manufacture, 221: 1543–1552.

[3]

Neely, A., 2007, The Servitization of Manufacturing: an Analysis of Global Trends, Proc. 14th European Operations Management Association Conference, Ankara.

[4]

Goedkoop, M. J., van Halen, C. J. G., te Riele, H. R. M., and Rommens, P. J. M., 1999, Product Service Systems: Ecological and Economic Basics, Report for Dutch Ministries of Environment (VROM) and Economic Affairs (EZ).

[5]

Mont, O., 2002, Clarifying the Concept of ProductService System, Journal of Cleaner Production, 10, 237–245.

[6]

Mont, O., 2004, Product-Service Systems: Panacea or Myth?, Ph.D. Dissertation, Lund University.

[7]

Manzini, E. and Vezolli, C., 2003, A Strategic Design Approach to Develop Sustainable Product Service Systems: Examples Taken from the ‘Environmentally Friendly Innovation’ Italian Prize, Journal of Cleaner Production, 11, 851–857.

[8]

Morelli, N., 2003, Product-Service Systems, a Perspective Shift for Designers: A Case Study: the Design of a Telecentre, Design Studies, 24(1): 73– 99.

[9]

Aurich, J. C., Fuchs, C., and Wagenknecht, C., 2006, Life Cycle Oriented Design of Technical Product-Service Systems, Journal of Cleaner Production, 14: 1480–1494.

[10] Aurich, J. C., Schweitzer, E., and Mannweiler, C., 2008, Integrated Design of Industrial Productst Service Systems, Proc. 41 CIRP Conf. on Manufacturing Systems, Tokyo. [11] Matzen, D. and McAloone, T. C., 2006, A Tool for Conceptualising in PSS Development, Design for X, Beiträge zum 17. Symposium. Lehrstuhl für

Konstruktionstechnik, Erlangen, 131–140.

Technische

Universität

[12] Matzen, D. and McAloone, T. C., 2008, From Product to Service Orientation in the Maritime st Equipment Industry – a case study, Proc. 41 CIRP Conf. on Manufacturing Systems, Tokyo. [13] Tomiyama, T., Shimomura, Y., and Watanabe, K., 2004, A Note on Service Design Methodology, Proc. ASME Int’l. Conf. of Design Theory and Methodology, Salt Lake City. [14] Lindahl, M., Sundin, E., Sakao, T. and Shimomura, Y., 2005, An Application of a Service Design Tool at a Global Warehouse Provider, Proc. Int’l. Conf. on Engineering Design, Melbourne. [15] Sakao, T., Shimomura, Y., Comstock, and M., Sundin, E., 2005, Service Engineering for Value Customization, Proc. 3rd Int’l. World Congress on Mass Customization and Personalization (MCPC), Hong Kong. [16] Sakao, T., Shimomura, Y., Comstock, M., and Sundin, E., 2006, A Method of Value Customization, Proc. Int’l. Design Conference, Dubrovnik. [17] Sakao, T., and Shimomura, Y., 2007, Service Engineering: a Novel Engineering Discipline for Producers to Increase Value Combining Service and Product, Journal of Cleaner Production, 15: 590– 604. [18] Maussang, N., Sakao, T., Zwolinski, P. and Brissaud, D., 2007, A Model For Designing ProductService Systems Using Functional Analysis and Agent Based Model, Proc. Int’l. Conf. on Engineering Design, Paris. [19] Maussang, N., Zwolinski, P. and Brissaud D., 2008, Evaluation of Product-Service Systems during Early st Design Phase, Proc. 41 CIRP Conf. on Manufacturing Systems, Tokyo. [20] Meier, H. and Völker, O., 2008, Industrial ProductService Systems – Typology of Service Supply 2 st Chain for IPS Providing, Proc. 41 CIRP Conf. on Manufacturing Systems, Tokyo. [21] Ohtomi, K., 2005, Importance of Upstream Design in Product Development and Its Methodology, Keynote Presentation, Proc. 6th IEEE EuroSimE Conference, Berlin. [22] ENGINE, Design for Service for Both Service and Manufacturing Businesses, http://www.enginegroup.co.uk.

39

Empirical Study Concerning Industrial Services within the Austrian Machinery and Plant Engineering Industry K. Matyas, A. Rosteck, W. Sihn Institute for Management Science, Vienna University of Technology Theresianumgasse 27, A-1040 Vienna, Austria [email protected] Abstract Industrial Services adapted to the requirements of the customers and aligned with the specifications of the physical products are becoming increasingly important for companies in the engineering industry in order to remain competitive and to ensure long-term economic success. An empirical study in the Austrian machinery and plant engineering industry should systematically evaluate and document corporate practice with regard to services. The current service landscapes in companies and the development of customer requirements have been investigated. The result of the study is a general idea of what the future needs for action in these companies as well as in applied research are. Keywords: Industrial Services; Engineering Industry; Service Development, Service Management

1 INTRODUCTION Machinery and plant engineering will in future be characterised by increasing global competition. The growing competition and cost pressure from the Far East in particular will make it more difficult for machinery manufacturers to differentiate themselves from competitors simply on the basis of their products. The service sector in machinery and plant engineering is a very important success factor for the companies with an increasing impact. Industrial services are no coproducts of the producton sector. Innovative industrial services are independent corporate strategies. In these general conditions it is especially important for companies to recognise industrial service as one of the factors for success in the coming years and to undergo the transformation from just a producer to a "producing service provider". What is of decisive importance in this process is to have a clear strategy for the area of service rooted in the company and to support and implement these on the basis of standardised and practiced processes and methods. Innovation in service also offers considerable opportunities to achieve differentiation from the competition and to increase customer loyalty. On average, service products achieve higher returns than the tangible product "machine" or "plant" and thus make a substantial contribution to ensuring a company's long-term success and competitiveness. This study deals with the business practice of service, i.e. with the organisation, processes and development of services and with future challenges, and presents a clear picture of the current situation in the industry and of the direction in which companies must develop. 2 OBJECTIVES AND METHODOLOGY 2.1 Objectives The study "Establishing the requirements for service concepts in Austrian machinery and plant engineering" was conducted by the Institute for Management Sciences of

CIRP IPS2 Conference 2009

40

the Vienna University of Technology on behalf of the FMMI (Association of Austrian Machinery and Metalware Industries). The study was conducted in the period from October 2007 to May 2008. The aim of the study was to investigate current business practices in the area of service in companies engaged in machinery and plant engineering and to identify future trends. In addition, recommendations should be given on what steps should be taken to improve service performance. At the centre of interest were the member companies of the FMMI, in particular in the sectors "machine tools" and "woodworking machines". The main questions asked in the study are: •

What services does your company offer?



What is the significance (turnover, revenues, strategy etc.) of the service area in your company?



How important is the development of services?



What are the future trends and developments and where do the companies see the need to act in view of these trends?

2.2 Methodology The design of the study was performed according to the principles of a combined investigation. On the one hand a quantitative investigation based on a questionnaire was carried out to achieve a higher sample size with the same employment of resources.[1] On the other hand also a qualitative investigation in form of semi-structured interviews with selected companies from the sectorial groups "machine tools" and "woodworking machines" was conducted to gain more detailed results. [2] The methodology in detail: •

Dispatch of the questionnaires to companies from the sectors in focus



Follow-up by telephone



Conducting semi-structured interviews at selected companies in the sectorial groups. Study participants:

having a large proportion of service employees among the total workforce. Approx. 20% of the companies contacted in the course of the study provide no service at all. The return on the service business is on average higher than on the traditional business with new machinery:



Total number of FMMI member companies contacted: approx. 700





Response rate approx 10% (including 7 interviews)

Two thirds of all companies generate a maximum return on services twice as high as that on the traditional business



10% of companies even achieve a return on service business that is 6 times as high as that on traditional business

Conducting structured telephone interviews with FMMI companies involved in machinery and plant engineering



3 RESULTS OF THE STUDY 3.1 Group of participants (companies) The study's focus was the machinery and plant engineering sector in Austria, which accordingly accounted for 90% of the companies polled. The corporate structure of the firms participating in the study reflects the corporate landscape in Austria: More than half (56%) of the responses received come from small and medium-sized enterprises as defined by the European Union (EU). 55% of the companies have a turnover of ≤ 50 million euros (SMEs), 20% are small enterprises with turnover of ≤ 10 million euros. The reason why so many small and medium-sized companies were considered in the study is the distinct situation of companies in Austria. More than 60% of the employees are working in small and medium-size enterprises due to the fact that less than 1% of the Austrian companies exceed a size of 250 employees.



The net returns that can be achieved vary within the portfolio of services offered, for example: o spare parts business: 10%-30% o training: 2%-5% What are the prospects, in the opinion of the companies, for the development potential both for turnover and for returns on service business in the medium term (over the next 4 years)? The majority of companies (>60%) see no or only limited opportunities for increasing turnover and returns in the area of services in the next 4 years (see figure 2). However, none of the companies expects returns or turnover from services to decrease, and the expected growth rates are independent of the size of the company.

3.2 Importance of service to the companies Figure 1 shows that on average approx 9% of the total workforce works in the area of service in the companies, straight service businesses excepted. A direct correlation between the proportion of service staff compared with the total workforce and the success of the company in the area of service (measured by the operating margin in service) cannot be established. Figure 2: Expected growth in service turnover in the next four years The underrated opportunities for increasing returns in the area of service suggest in part a lack of strategic direction in service, which is something that will be demonstrated later.

no

no

no

no

yes

yes

yes

Figure 1: Number of service employees compared to the total number of employees The average turnover generated by the companies with service, straight service businesses excepted, amounts to approx. 11%. This value is towards the lower limit in an international comparison. In comparison: in Germany approx. 27% of turnover in machinery engineering is accounted for by service [3]. In this context it is also worth noting that a large proportion of the firms interviewed stated that a goal of 15% was the maximum that was aimed for. The best in this class achieve over 30% turnover with service and the worst around 1%, with the best in class

yes

Figure 3: Organisation of services

41

With regard to the way services are organised it can be stated that more than 66% of the companies have their own organisation for service – in most cases within the normal organisation. Figure 3 attempts to illustrate that for 64% of the companies service is a strategic field of business that should be addressed with a clearly formulated service strategy. However, the relatively low growth expectations (see Figure 2) show that actions are not completely aligned with strategy or that the company does not have a clear strategy. 23% of the companies run their service business under a different serve brand (competitive differentiator, customer brand identification,...). A correlation can be established between a positive response to the questions and the turnover in the area of services. Large companies thus have a more distinctive service organisation than small ones. Operating the service business under a separate brand is more likely to be carried out by larger companies. 3.3 Services offered in the companies In principle, the current range of services offered by the companies includes services, product upgrades and customer support as shown in figure 4. •

"Classic services", i.e. services characterised by low innovation, still play the main role for the companies surveyed.



Maintenance and servicing or repairs together with the spare parts business are the most widespread services [4]

are not very widespread among the companies surveyed. Only approx. 6% say that they offer operator models. Currently, results-oriented models only play a subordinate role in comparison with the classic services. The question regarding the three strongest selling services in the company reveals that the most frequently offered services are those which generate the highest turnover for the companies: repairs, spare parts and maintenance are the services generating most turnover for more than 75% of the companies. Only very limited turnover is currently generated with more innovative service concepts or business models. 3.4 Service development In figure 5 it is shown that the most important triggers for developing new services are customer requirements and requests. Accordingly, most ideas for new services originate from customer enquiries. The companies are primarily customer-driven and are in particularly not proactive and innovative when it comes to developing new services. It is only the secondary answer "technical potential" that indicates that services are also developed on the companies' own initiative.



Individual and special services are frequently neither offered as standard nor promoted but provided on an ad hoc basis in response to a customer's request. There is a basic correlation between a company's size and the number of services it offers: large companies offer a higher number of services than small ones. No correlation can be established between the number of services offered and the success of the company (e.g. return on the service).

Figure 5: Triggers for developing services In order to be able to develop services successfully it is important to have clearly defined, standardised and practiced development processes in the company that are supported by appropriate methods. In this area, more than two thirds of the companies state that they have standardised development processes with written documentation. However, a systematic methodology-based approach is not very widespread and special methods for developing services are not widely employed. When methods are used, they are split approximately equally between creativity techniques and process modelling methods. When performing comprehensive service development there are four fundamental aspects that must be defined:

Figure 4: Services offered 25% of the companies surveyed indicate that they offer results-oriented business models and/or operator models [5]. Here in detail the most important business model, with a share of 19%, is machine availability. TCO models (TCO – Total Cost of Ownership) do not currently play a large role for the companies (approx. 4%) and operator models

42



product model – describes the scope of services



process model – formulates the processes that are necessary to deliver the service



resource model – determines the resource required to deliver the service



marketing concept – specifies the strategies and instruments used to market the service The survey results show a recognisable fundamental importance for systematic and structured service development for the companies, but the variance of the answers among the individual activities of service development is high, i.e. the companies rate this very differently.

Î A structured and continous development process for industrial services is hardly implemented in most of the companies.

3.6 Challenged and future developments As shown in figure 7 the biggest difficulties that companies have to face when providing services are primarily: •

3.5 Marketing and customer communications As described above, customer requests are the most common trigger for the development of new services. This explains why the most important sales channel for the companies is customer contact by sales representatives. Other marketing methods, apart from each company's website, play a subordinate role. Defining a marketing strategy during the development of the service is of decisive importance for its subsequent economic success. What factors make a service successful from the point of view of the companies selling it? The quality and advantages of the service are critical for its success on the market; in a similar manner customer communication is seen as an important success factor, with a well defined interface to the customer being essential. The primary communication channel to the customer (approx. 45%) is the sales representative; less than 20% of the companies use the instrument of regular and standardised customer surveys. A well functioning complaints management process is indispensable as a part of service quality management and is underrepresented in the companies' answers ( 1,000) companies. 4.3 Survey graphics When reading the results, one should have in mind that there are a different number of answers that were received from each of the different countries. This is especially important to note from the German part of the survey, which only includes answers from a maximum of 4 participants. This could in some cases give a somewhat biased graph for the German companies. These issues

75

are solved to some extent by also showing the total response for all countries in the same graph. Therefore, the responses for each country are complimented with an overall black bar called “All Countries” to the right in the graphs. The number in brackets denotes how many answers were received for the specific question.

establish closer and long-lasting relationships with business customers. Other identified driving forces, besides those connected to the customer, include (in decreasing order); increased competition, decreased costs, and improved company brand. It is interesting to see that increased competition is a minor reason for Japanese companies in comparison to other countries’ companies, especially Swedish companies.

4.4 Driving forces for PSS Figure 1 shows that, in all four countries, many incentives for PSS business exist and are connected to the customer. This result points out companies’ attention to customer satisfaction, and also refers to the possibility to Sweden (12)

Japan (8)

Germany (3)

Italy (6)

All Countries (29)

100% 90% 80% 70%

Share

60% 50% 40% 30% 20% 10%

isc el la no us M

le dg e ow Kn

er C

Ac hi ev e

Im

Pr od uc t

C C

D

us to m

ec r

ea se

en ta l nm

pr ov e

Im

on ne ct io n

os ts

ag e Im

Br an d En vir o

om

pr ov e

C

th e Se cu re

pa ny

Af te rm

ar k

et

Pr of it ro du ct

ai n

La rg

er P

C G

In cr ea se d

C

us to m

er D

om

em

an ds

pe ti t io n

0%

Figure 1. Driving forces for Product/Service Systems providers. (The graph shows each country’s response along with the aggregated total, n=29.) Sweden (12)

Japan (9)

Germany (3)

Italy (7)

All Countries (n=31)

100% 90% 80%

60% 50% 40% 30% 20% 10%

R Ta ke -b ac k

Figure 2. Contents of Product/Service Systems. (The graph shows each country’s response along with the aggregated total, n=31.)

76

Li ce ns es

es po ns ib ilit y

ep ai rs R

ai nt en an ce M

Co ns um pt io n

on su m C

En er gy

oo ds pt io n

G

pe ra to rs O

ro du ct s

0%

Ph ys ic al P

Share

70%

4.5 Contents of PSS In the survey, contents of PSS offers were also investigated. Figure 2 shows that responses from all companies were quite similar. The three main parts of the offers seem to be physical products, maintenance, and repairs. The figure also shows that it is only the German companies that also often included energy consumption in their PSS; however, the average share for all countries which had energy consumption included in their offers was only 16%. Furthermore, it is interesting to see that operators are included by the PSS providers in more than 20% of the PSS offerings.

Sweden (12)

Japan (8)

4.6 Actors in the PSS development When investigating which departments were involved in the design of product/service offers, it was found that staff from product development, marketing and after-sale was most common for most countries (Figure 3). However, in Italy there was also a major inclusion of the production department. Answers for “Other” include sales department and product line management. It is interesting to see that both the product development and the marketing departments in German companies were often included in the development work but, as seen in Figure 6, the physical products were always standard products. Almost the same pattern shown in Figure 3 is obtained when asking participants which department holds the

Germany (3)

Italy (6)

All Countries (29)

100% 90% 80% 70%

Share

60% 50% 40% 30% 20% 10% 0% Product Development

Marketing

Finance

Production

After Sale

Other

Figure 3. Participating departments in the Product/Service System development. (The graph shows each country’s response along with the aggregated total, n=29.) Sw eden (11)

Japan (9)

Germany (3)

Italy (6)

All Countries (29)

80% 70% 60%

Share

50% 40% 30% 20% 10% 0% Product Development

Marketing

Finance

Production

After Sale

Other

Figure 4. Responsible departments for the Product/Service System development. (The graph shows each country response along with the aggregated total, n=29.)

77

Sweden (12)

Japan (9)

Germany (3)

Italy (9)

All Countries (31)

100% 90% 80% 70%

Sh are

60% 50% 40% 30% 20% 10% 0%

Your Company

The Customer/User

3rd Party Company

Figure 5. Ownership of physical products used in Product/Service Systems. (The graph shows each country’s response along with the aggregated total, n=31.) Sweden (12)

Japan (9)

Germany (3)

Italy (7)

All Countries (31)

100% 90% 80% 70%

Share

60% 50% 40% 30% 20% 10% 0% Standard Products

Standard Products Adapted for PSS

Products Designed for PSS

Figure 6. Adaptation of physical products used in Product/Service Systems. (The graph shows each country’s response along with the aggregated total, n=31.)

responsibility for the development of product/service offers, as Figure 4 shows. 4.7 Ownership of physical products used in PSS The ownership of the customer reduces the possibilities and potential profits for the manufacturer to adapt their physical products for the use and remanufacturing phases, since improvements for these phases will not gain any profits for the manufacturer. In addition, the control over the products during use is worse if the ownership is transferred to the user. Looking at the ownership of ingoing physical products included in the PSS offer when the contract had been signed, Figure 5 shows that the traditional sales-type contracts seemed to be the norm in Sweden, Japan and Germany, while other various types of contracts were found in Italy. 4.8 Adaptation of physical products towards PSS Figure 6 shows the manner in which ingoing physical products are adapted for PSS. The result shows a mixed

78

picture, e.g. companies in Sweden and Germany mostly used standard products for their product/service offerings, while in Italy and Japan, the companies more often specifically adapted or designed products for PSS. The reason to why these differences between the countries exist is not clear. A strong involvement from the responsible departments “product development” and “production” in Japan and Italy could have had a strong influence on the product adaptation (see also Figure 4). The participants were also asked what kind of adaptations they had made to their physical products. Here are some of their answers: •

“The products are adapted for our business strategy.”



“We have designed surveillance systems in order to reduce production stops for the customer and to maximize the maintenance intervals to reduce our costs. Machines all over the world are monitored.”



“Realization of on-line service by designing a new device to integrate subsystems and adding communication functionalities”

5 DISCUSSION In sections below, the results obtained from the survey are discussed and summarized with some conclusions. 5.1 Method, response rate and companies The low number of respondents (n=34) makes it hard to draw any general conclusion of PSS providing companies based on only this survey. However, it is a good starting point for further studies, and the results give an indication of how a number of manufacturing companies in Sweden, Japan, Italy and Germany are working with PSS. The results from the Japanese and Swedish part of the survey are judged to be the most reliable, since the authors of this paper have best overview over those two countries’ industry. Although the questionnaire was sent out to the recipients in two different ways (web-based and paper-based), the authors are convinced that accurate answers were received from the recipients. This is because the authors had previous contact with the companies, and because the surveys were sent out in their own language. The relative high response rate indicates that the selected approach has been successful. One of the challenges the authors had when starting the survey was to identify companies offering PSS to their customers. The PSS concept was relatively new, and few companies had experience with PSS. This is also quite obvious when reviewing the early literature in this area, where some company examples are raised over and over again. Another challenge, still existing, was that even though companies work with PSS, they very seldom use the term PSS, and it was and is still therefore tricky to identify them. However, since the survey, the concept has become more mature, popular and widespread in industry, making it easier to identify companies. The authors’ experience from projects [11] and seminars with companies is that companies’ interest in participating in research about PSS has increased between 2005 to 2007. This implies that today it would have been easier to find more participants for this type of survey. This have also been indicated by the possibility to start up a company network of large-sized companies in Sweden [12]. It is interesting to note that the Swedish and Japanese surveys mainly consisted of companies with more than 1,000 employees. Since the authors before the surveys in Japan and Sweden did their best to identify companies working with PSS, the domination of major companies indicates that large companies have been the first adopters of PSS [9]. Not only is this PSS survey dominated by large companies, but so are other studies from this time, see e.g. Windahl [13] and Lakemond Ebbers and Magnusson [14]. A reason why large companies are dealing with PSS is that they can afford to spend some resources on testing new concepts like PSS, and also because the knowledge transfers from academia to major companies is faster and easier than to small companies, since major companies and researchers often have experience working with each other. The Italian survey is an exception, since that survey mainly consisted of small companies with fewer than 100 employees. 5.2 Driving forces for PSS The presented result shows that the main incentive for companies to work with PSS is connected to their customers, i.e. these companies’ attention to customer satisfaction in order to increase the possibility to establish closer and longer-lasting relationships with business customers. The highlighted driver “improved company brand” is in line with this. A good company brand is

important when developing longer and closer relationships. At the same time, a closer and longer relationship has the potential to imply that the amount of contact between company and customer will increase. This also implies that the providing company will have more opportunities to improve their customers’ understanding of them. Having a close relation between the manufacturer and the customer has been shown as the most preferable relation to have if the manufacturer wishes to remanufacture its products [15]. This is due to several reasons; one is that better control and knowledge of the product being used is achieved. The connection between PSS and remanufacturing was elucidated early by researchers [5] but and in recent years the connection have been adopted by industry e.g. for PSS sales of forklift trucks and soil compactors in Sweden [16] and photocopiers in Australia [17]. Both previous drivers, “the possibility to establish closer and longer-lasting relationships with business customers” and “more opportunities to improve their customers’ understanding of them” are related to the driver caused by increased competition. A closer and longer-lasting relation with the customer, and a stronger company brand, are means to increase a company’s own competitiveness and decrease the opportunities for other competitors to approach a company’s own customers, since they are more closely tied through the PSS offer. Furthermore, there seems to be diverse incentives (see Figure 1) for manufacturers in these four countries to have a PSS business which is in line with other PSS research studies e.g. [18]. Decreasing costs are another and quite often mentioned driver for companies to work with PSS. However, this driver is quite obvious and not unique for, PSS since cutting costs is a core activity in more or less all companies. This also is in line with other concepts like e.g. lean product development [19] and lean production [20, 21], which have also become popular in recent years within the industry since those concepts match companies’ drivers to cut costs. Even though many authors, e.g. Mont [3], have highlighted PSS from an environmental point of view, these results show companies do not consider improving environmental image and achieving more product knowledge to be important driving forces for PSS. However, since the results show that a major driver for PSS is to build up a good level of customer satisfaction, this lack of interest in environmental image might change if the customer starts to focus more on environmentally related issues. Since these studies were done, there has been quite an increased focus from customers and society on environmentally related issues; it would, therefore, have been interesting to study further if and how this driver has changed. 5.3 Contents of PSS As shown in the results, the three main parts in PSS offers are physical products, maintenance and repair (Figure 2). It should be noted that when conducting maintenance and repair, physical products are used and subsidiary products are consumed. Also, all other alternatives answered shows that physical products and/or consumables are provided. This is also in line with Lindahl and Ölundh [4]. The results also show that the content of a PSS offer can vary significantly, depending on how the contracts are written between the PSS provider and the customer. A contract can e.g. beside the pure physical products include more or less maintenance, spare parts, operators and energy consumption.

79

5.4 Ownership of physical products used in PSS It is interesting to note that the normal case in the studied companies’ PSS offers is that the ownership of the physical products are moved away from the providing company to the user or a third party. This is interesting since some researchers, e.g. Mont [7] and Baines et al. [22] consider that ownership of products should not be transferred to the customer for it to be a PSS, while others consider different types of product-related services as a PSS. For example, Baines et al. [22] state that “with a PSS, asset ownership is not transferred to the customer”. The structure of the ownership is important for the control and logistics of the physical products during use and takeback. If the ownership stays with the provider, there are many benefits to achieve. For example, the product takeback and remanufacturing issues are much more easily dealt with [15]. 5.5 Adaptation of physical products used in PSS From this survey and others similar to it e.g. Ölundh and Ritzén [18], one can conclude that the included physical products were standard products not adapted for PSS. The Italian companies, along with Japanese companies (as opposed to the German ones), seem to be advanced at adapting their physical products for PSS. One potential problem with having PSS offers with non-adopted products, i.e. standard products, is e.g. if they not are designed to enable easy service and maintenance and this is included in the contract. The incitement for adapting the PSS offerings’ ingoing physical products increases if the manufacturer maintains ownership during the use phase. The issues of how physical products could be adapted for PSS and remanufacturing is further described in general in [23] and more in detail in [16]. 5.6 Actors in development of PSS The results indicate that PSS are developed in a quite traditional way, i.e. more or less in the same way as traditional product offers. Participating departments are the traditional ones, with the product development department acting as the stronger and often responsible actor. This might also be a reason why most companies still use standard products in their PSS offers (Figure 6) In the Italian-German part of the survey two departments, product development and marketing were primarily involved in the product/service offer development, whereas among the Swedish and Japanese companies, the department of after sales had an equally large share of the involvement. In Sweden, the after sales department also had responsibility for the development of the product/service offering at one-third of the companies, in contrast to the results from Germany and Italy. Further, the mix of people in the development teams surely affects the degree of adaptation of both physical products and services of the PSS; this is for example due to the team members’ personal experiences and their own departments’ incentives. In Sweden, the authors recently have noticed a trend among PSS providing companies that are trying to incorporate more knowledge from service/maintenance and product take-back phases even more now than before when having traditional product sales. Some challenges found for this are the company departments’ different goals of their business [12]. 6 CONCLUSIONS AND FUTURE RESEARCH This paper contributes with quantitative data and discussion regarding companies’ work with PSS offerings, and gives a good starting point for further studies. It shows that the development of PSS offers has many

80

similarities to ordinary product development, since driving forces are alike. The result can be used as a base for developing PSS methodologies for Industry. The objective was to find answers to the following research questions. These answers are provided in brief below. RQ1 – What driving forces do PSS providers have? – Customer connection, customer demands and increased competition were the main driving forces. RQ2 – What is included in PSS offerings? – Most PSS include physical products, maintenance and repairs. For some (40%), PSS consumption goods are included, and in rare cases (15%) times energy consumption is included. RQ3 – Who are involved and responsible for PSS development? – The departments of “Product Development”, “Marketing” and “After-Sales” are the major actors in PSS Development. In most cases, the “Product Development” department was in charge of the PSS development. RQ4 – Who owns the physical products used in PSS offerings? – In most cases the ownership is transferred to the customer/user using the PSS similar to traditional sales. RQ5 – How are physical products adapted to PSS? – Standard products are mostly used in PSS, although the Japanese and Italian companies are more likely to provide products designed/adapted for PSS. Furthermore, since the number of companies providing PSS has increased during the last few years, it would be interesting to follow up this survey and also include some of the new issues that have been identified in order to get more reliable and stronger data about how companies are working with PSS. Comparison with the activities in the tertiary industry would be interesting as well. The researchers of this paper will continue to conduct surveys regarding PSS in their respective countries. In parallel, the authors of this paper invite other researchers to initiate complementary surveys in line with this survey in their own countries. Future research may also include specific investigation of the operational maintenance methods to be offered by the providers (e.g. breakdown or corrective maintenance, planned maintenance, condition-based maintenance). This includes what kind of support which is needed during the life-cycles of product/services. 7 ACKNOWLEDGMENTS The authors would like to express their gratitude to Professor Mats Björkman, Department of Management and Engineering, Linköping University, Sweden for support when developing the questionnaire, and Professor Massimo Tronci and Dr. Nicola Napolitano of the University of Rome “La Sapienza” for helping us to conduct the survey in Italy. Also, we would like to thank Professor Birkhofer of the Institute for Product Development and Machine Elements (PMD) of Darmstadt University of Technology, for the support to conduct the survey in Germany. All companies participating in the surveys are acknowledged. This research was mainly supported by the Swedish Governmental Agency for Innovation Systems (VINNOVA), the Swedish Association of Graduate Engineers and partly supported by a Research Fellowship Program by Alexander von Humboldt Foundation in Germany. 8 REFERENCES [1] Goedkoop, M.J., C.J.G. Van Halen, H.R.M. te Riele, and P.J.M. Rommens, Product Service Systems, Ecological and Economic Basics. 1999, VROM: Hague, the Netherlands.

[2]

[3]

[4]

[5]

[6] [7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

Tischner, U., M. Verkuijl, and A. Tukker, First Draft PSS Review. SusProNet Report, draft 15 December. Available from Econcept, Cologne, Germany; TNOSTB. 2002: Delft, the Netherlands, or www.suspronet.org. Mont, O., Special Issue on Product Service Systems and Sustainable Consumption. Journal of Cleaner Production, 2003. 11(8): p. 815-933. Lindahl, M. and G. Ölundh. The Meaning of Functional Sales. in Life Cycle Engineering: Challenges and Opportunities: 8th International Seminar on Life Cycle Engineering. 2001. Varna, Bulgaria: CIRP. Sundin, E., M. Björkman, and N. Jacobsson. Analysis of Service Selling and Design for Remanufacturing. in Proceedings of IEEE International Symposium on Electronics and the Environment (IEEE-00). 2000. San Francisco, CA, USA. Tukker, A. and U. Tischner, New Business for Old Europe. 2006, Sheffield: Greenleaf Publishing. Mont, O., Product-service systems: Panacea or myth?, in The International Institute for Industrial Environmental Economics (IIIEE). 2004, Lund University: Lund, Sweden. p. 233. Lindahl, M., Engineering Designers' Requirements on Design for Environment Methods and Tools, in Industrial Engineering and Management. 2005, KTH: Stockholm, Sweden. Sundin, E., M. Lindahl, Y. Shimomura, and T. Sakao. Need for New Engineering Design Methodologies for Functional Sales Business -An International Survey Concerning the Experiences of the Business Concept within Japanese and Swedish Industries. in Proceedings of the 15th International Conference on Engineering Design (ICED05). 2005. Melbourne, Australia: ICED. Sakao, T., N. Napoletano, M. Tronci, E. Sundin, and M. Lindahl, How Are Product-Service Combined Offers Provided in Germany and Italy? – Analysis with Company Sizes and Countries. Journal of Systems Science and Systems Engineering, Accepted to appear. Lindahl, M., E. Sundin, Y. Shimomura, and T. Sakao. An Interactive Design Model for Service Engineering of Functional Sales Offers. in Proceedings of the International Design Conference - Design 2006. 2006. Dubrovnik, Croatia: Design Society. Sundin, E., G. Ölundh Sandström, M. Lindahl, A. Öhrwall Rönnbäck, T. Sakao, and T. Larsson, Industrial Challenges for Product/Service Systems: Experiences from a large company network in Sweden, in Proceedings of CIRP Industrial Product/Service Systems. 2009: Cranfield, The United Kingdom. Windahl, C., Towards Integrated Solutions - Alfa Laval and the Wastewater Industry, in Division of Industrial Management, Department of Management and Economics. 2004, Linköpings Universitet: Linköping, Sweden. Lakemond Ebbers, N. and T. Magnusson, Creating value through integrated product-service solutions: Integrating service and product development, in IMP. 2005: Rotterdam. Östlin, J., E. Sundin, and M. Björkman, Importance of closed-loop supply chain relationships. International Journal of Production Economics, 2008. 115(2): p. 336-348.

[16] Sundin, E., M. Lindahl, and W. Ijomah, Product Design for Product/Service Systems - design experiences from Swedish industry. provisionally accepted in Journal of Manufacturing Technology Management, Special issue on Product Service Systems, 2009. [17] Kerr, W. and C. Ryan, Eco-efficiency gains from remanufacturing - A case study of photocopier remanufacturing at Fuji Xerox Australia. Journal of Cleaner Production, 2001. No. 9: p. 75-81. [18] Ölundh, G. and S. Ritzén. How do functional sales affect product development and environmental performance? in International Conference on Engineering Design, ICED 03. 2003. Stockholm. [19] Morgan, J. and J. Liker, The Toyota Product Development System: Integrating People, Process, and Technology. 2006: Productivity Press. [20] Ohno, T., Toyota production system: beyond large scale production. 1988, Cambridge, Mass.: Productivity press. [21] Womack, J.P. and D.T. Jones, Lean Thinking – Banish Waste and Create Wealth in your Corporation. 1996, New York: Simon & Schuster. [22] Baines, T.S., H.W. Lightfoot, S. Evans, A. Neely, R. Greenough, J. Peppard, R. Roy, E. Shehab, A. Braganza, A. Tiwari, J.R. Alcock, J.P. Angus, M. Bastl, A. Cousens, P. Irving, M. Johnson, J. Kingston, H. Lockett, V. Martinez, P. Michele, D. Tranfield, I.M. Walton, and H. Wilson, State-of-theart in product-service systems, in Proceedings of the IMECHE Part B Journal of Engineering Manufacture. 2007, Professional Engineering Publishing. p. 15431552. [23] Sundin, E. and B. Bras, Making Functional Sales Environmentally and Economically Beneficial through Product Remanufacturing. Journal of Cleaner Production, 2005. 13(9): p. 913-925.

81

Service and manufacturing knowledge in product-service systems: a case study 1

A. Doultsinou1, D. Baxter1, R. Roy1, J. Gao2, A. Mann3 Decision Engineering Centre, Cranfield University, Bedfordshire, MK43 0AL, UK 2 School of Engineering, University of Greenwich, Kent, ME4 4TB, UK 3 Edwards Ltd., Shoreham-by-Sea, West Sussex, BN43 6BP * [email protected]

Abstract In the developing Product-Service Systems (PSS) field, an emerging research challenge is supporting the PSS design activity. This paper presents a case study in which manufacturing and service knowledge is captured and classified in order to support the design activity. A knowledge capture exercise took place to identify manufacturing and service knowledge applied in the design process. A design knowledge capture exercise led to the creation of a design process model. The case study reports on the proposed structure for the application of manufacturing and service knowledge to a conceptual and a detailed design task. The knowledge framework is implemented using the Protégé knowledge base editor. PSS design requires an integrated system level approach to design, and therefore a system level knowledge structure is required. The detailed case study indicates where manufacturing and service knowledge is applied in the design activity, which is divided into ‘conceptual’ and ‘detailed’ stages. Keywords: Product-service systems, service knowledge, manufacturing knowledge, design

1 INTRODUCTION The emerging paradigm of Product-Service Systems (PSS) focuses on the integration of products and services to deliver customer value. The design of PSS is not simply product design followed by service design. An integrated approach taking service into account at the earliest stages of design is necessary. PSS should reflect an optimised product and service system: as such, that system becomes the focus for the design effort. This paper presents the results of an industrial case study, which took place during a 3-year industrial project and where manufacturing, design and service knowledge were captured and represented in a common knowledge base using Protégé software. The overall project aim is to develop a methodology to capture, represent and reuse knowledge to support product development in a collaborative enterprise context. The three core elements are: design knowledge, manufacturing capability knowledge, and service knowledge. The project aims to develop a means to ensure authorised access to the right knowledge for the different work functions in the product development process. Recognising crossover and synergy between design, manufacturing and service is a key aspect. The original aim of the project did not explicitly reflect the PSS design challenge, however it has been identified that this research can contribute to the developing area of technical Product-Service Systems through recognising the need for system level design and developing a knowledge framework to support it. The particular contribution of this paper in relation to the project is a description of the knowledge reuse framework, and applying a model of the design process as a central mechanism for knowledge reuse. In particular, this paper will describe how service and manufacturing knowledge is applied within the framework, in terms of when it is applied according to the design process and how it is

CIRP IPS2 Conference 2009

82

structured according to the ontology. Recognising the codevelopment of products, services, and their corresponding processes is contributing to the understanding of life cycle design knowledge support requirements. The process of developing the knowledge base as well as how it can be used and help designers at the conceptual and detailed product design is described in this paper. The future research agenda will then be outlined, which includes the final validation of the knowledge structure and a description of its possible applications and limitations. 2 LITERATURE FINDINGS 2.1 Service and PSS As far as technical services are concerned, several types of service activity can be identified, including: planned maintenance, unplanned maintenance, service exchange, product repair and overhaul, retrofitting and upgrading, product installation, commissioning and monitoring [1]. Product-Service systems extend the traditional functionality of a product by integrating additional services [2]. There are three different types of PSS that can be found in the literature [2]: 1. Product-oriented PSS: traditional sale of a product with the addition of services, like warranty, repair, maintenance, upgrades, re-use and recycling. The ownership is transferred from the supplier/ manufacturer to the customer. 2. Use-oriented PSS: sale of the use or availability of a product (e.g. leasing). The ownership is not transferred to the customer. 3. Result-oriented PSS: sale of the result or capability of a product. The ownership of the product is retained by the company and the customer pays only for the delivery of the agreed results.

2.2 PSS design methodologies PSS represents an integrated product and service offering. This needs an integrated development approach for products and services [1]. Therefore, a methodology to support PSS design is required. Aurich and Fuchs present three approaches to product and service design [4]. These approaches are presented in Figure 1. The first approach stems from the traditional view of manufacturing companies, whose core competencies are the development and production of innovative and highly reliable products. In this case, the product design process is systematically structured, whereas the service design process is carried out in an intuitive fashion. This approach is called ‘liability driven’, since the objective is to minimise in-service problems. The second approach supports service enhanced products. Systematic product design leads to the development of product variants, each of them supported by a service package. These service packages are also developed using a systematic service design process. Products and services are not regarded as separate artefacts; they can be combined based on the customer requirements. This approach can be called ‘function driven’, where still the focal point is the physical product, but its function is accomplished through services. The focus on service increases as it becomes a more integral component of the business strategy. The third approach supports the development of individualised product-service solutions for each customer. It is referred to as ‘use driven’. In this case products and services are indivisible artefacts. Consequently, the design of each solution requires the integration of service elements into the product design process. Aurich et al. [5] suggest that in companies, product design is typically performed by technical staff. Service design, however, is typically carried out by marketing and distribution personnel. PSS design requires both aspects: an understanding of the product, along with the delivery supply chain and service environment. Similarly, Morelli states that the design of a PSS is a challenge from the designer’s perspective because an extension of their traditional know-how into new areas domains is necessary [3]. The designer needs to take into account various customer needs, and to develop a solution as a result of their synthesis. The customer perspective of the product-service experience is a central

theme in PSS design. Another example of a PSS design methodology is the MEPPS handbook [6], which was created under the Fifth Framework Programme supported by the European Commission. The MEPPS handbook describes the PSS design in terms of selection, design and development of PSS business model. MEPSS aims to impart active support during actual phase-by-phase execution of PSS innovation projects of organisations by suggesting a comprehensible modular structure and giving management and design support. Product-service systems are related with systems that consist of several actors (producers, service providers and users) which altogether offer and consume products and services. For this, system analysis is essential because it constitutes a vital basis for the understanding of the system variables, the relations between the stakeholders involved, and potential development alternatives using the current market situation as a starting point. Therefore, the main difference between MEPPS methodology and the other literature approaches is that it takes into account the system as a whole and its variables, the market requirements, sustainability and it recognises the value of the early involvement of stakeholders. The MEPPS handbook is largely focused on the communication that takes place, rather than the knowledge used, during the design of a new PSS. Tukker and Mont suggest that the emerging variety of PSS design methodologies conflicts with the potential for a generic methodology. They advise that selected generic principles will always be applicable and then they recommend that each company needs to find and apply its individual practical approach. They highlight the necessity to focus on the system perspective in PSS design [7]. In summary, the current limited number of approaches to PSS design available in the literature all emphasise the requirement for an integrated system level design of products and services. Few approaches are supported by industrial cases, with the exception of Aurich et al. None of the existing PSS design approaches provide a framework for integrating the various sources and types of knowledge required; in fact, since the area is relatively new there are no papers describing in detail the types and sources of knowledge required in technical PSS design. 2.3 Knowledge integration frameworks Since the integrated knowledge framework is the main outcome presented in the paper, literature related to

Figure 1: Product-service design strategies (Aurich and Fuchs, 2004)

83

Figure 2: System knowledge structure knowledge integration frameworks has been investigated, and is presented here. BadiI and Sharif [8] present a framework for optimising knowledge integration. Their approach relates to an internal, social process that facilitates dialogue and reflection. Huang and Newell [9] also propose a knowledge integration process, again focusing on social processes. It is the intention of this paper to develop a technocratic solution, which is intended to support an integrated design effort in the context of a mature domain. This information systems approach (in contrast to an approach focusing on interpersonal communication) requires a framework to provide structure for knowledge storage and reuse. Young et al [10] propose an information and knowledge framework that can be applied to various life cycle contexts (design, manufacturing, operations) with the product as the central element. Using the product as the central element limits its use for PSS design, since the development of the PSS concept requires a system level view. Various elements of the framework can be adopted, such as the manufacturing resource descriptions; however a broader ‘upper level’ focus is required. Lee et al [11] developed an object-based knowledge integration system for product development. They apply an object-oriented architecture, using XML as a data interchange protocol. Their integration system is primarily for message exchange during product development, rather than for knowledge support for designers. Chen and Liang [12] developed a collaborative engineering information system intended to support the integration, management and sharing of engineering information, including product and manufacturing process information as well as a description of the engineering process. The framework is not strongly focused on a central product model, so could be applied to a system level design problem; however it is lacking a service component. As such, it does not provide a complete platform for PSS design support. Sudarsan et al [13] describe a product information modelling framework to support “the full range of PLM information needs”. The framework is based on various standards, including the NIST core product model, open assembly model, design analysis integration model and the product family evolution model. Whilst ‘product in use’ is recognised as part of the life cycle view, the approach to managing in-use data is not presented. With no clear references to system level design, it is not clear

84

how this framework could be applied to a PSS design problem. 3 INTEGRATED FRAMEWORK DEVELOPMENT An attempt has been made by the researchers to create a framework that integrates the various knowledge types from a lifecycle system perspective both for manufacturing and service. In order to achieve this, a detailed case study has been carried out. In combination with the literature findings, this has supported the proposal for a knowledge integration framework. The case study research consisted of approximately 30 semi-structured interviews with the collaborating company. Interviewees with various job roles took part, including service engineers, designers, manufacturing engineers, quality engineers, project managers and service managers. Notes were the primary mechanism for recording the interviews. 8 of the early interviews in the service and manufacturing contexts were recorded and transcribed; following this, a thematic analysis was carried out. Two knowledge structures were developed and validated: one for manufacturing and one for service Baxter et al., 2008). In order to extract the service knowledge types from the interview data, the following procedure was followed Doultsinou et al., 2008): 

10 semi-structured interviews (using the critical incident technique); the main service knowledge types were identified



5 semi-structured phone interviews: the initial service knowledge types were modified and enriched



6 interviews: the final service knowledge structure was developed The service structure is described in detail in this section. Two main classes were considered: the ‘product’ and the ‘service organisation’. Then, these two classes were divided into subclasses, so that the detailed service knowledge types can be illustrated in the structure. Product

Service organisation

Service feature

Training

Product attribute

Personnel

Subsystem

Facility

Component

Spares

Service process

Logistics

Maintenance strategy Operation

Table 1: Classes and subclasses Protégé software was the tool used for the development of the knowledge structure. Therefore, according to the requirements of this software the subclasses needed to have ‘slots’, which describe the properties of the classes and subclasses. For example, ‘can be recycled’, ‘part of product’, ‘part of spares kit’ are the slots of the ‘component’ subclass. Then, ‘availability’, ‘cost’, ‘definition’, ‘spares ID’ are the slots of the ‘spares’ subclass. Considering the structure of the service knowledge base, two issues that were faced during the service of the case study product can be described: 1. Long time to disassemble due to the big number of water pipes: This is implemented in the KB using the original structure: this is a serviceability issue of a specific product than can be recorded. It also relates to the subsystem issues (i.e. water system), where the issue and the actions taken to tackle it can be recorded. 2. Small pump is harder to move. This issue is described as part of the manoeuvrability class, where the service engineers can report any issues regarding this specific product attribute. The two knowledge structures were then integrated, taking into account the requirements for a system level design perspective. The final (generic upper level) structure is presented in Figure 2. Lower level classes are implemented for the specific case example, including component types, module types, requirements categories and design features. Figure 2 illustrates the generic knowledge structure, providing a mechanism to structure and store various knowledge instances that need to be considered from a manufacturing and service perspective when the whole lifecycle of the product needs to be designed. The top level class ‘life cycle system’ is comprised of three key classes: product, process and resource. The other classes, e.g. requirements, behaviour, logistics, operating methods and installation environment, are used to describe the system. After having created the class-subclass-slot structure, some instances were created using one of the collaborator’s products as the main focus (Figure 3).

Figure 3: Case study product

An instance, which has been created and is related to the product selected, has the following structure: 1. Product name 2. Has Bill of Materials 3. Product process 4. Description 5. Product ergonomics 6. Recorded failures 7. Has architecture 8. Operating system 9. Similar products 10. Has maintenance strategy 11. Product requirements 4 INDUSTRIAL CASE STUDY The case study company is a leading manufacturer of vacuum pumps. A description of their design process is illustrated in Figure 4. It is divided according to the conceptual and detailed stages. The black boxes (square corners) represent activities. The blue boxes (rounded corners) represent the datasets which are used as inputs to the activities. These datasets are also activity outputs. As a result of the case study, the researchers have suggested that the manufacturability analysis activity should be formally applied at the early stages, providing an input to the performance modelling activity. The output from the manufacturability analysis activity – the feature list – is comprised of module clearances, derived from component tolerances. A critical element of these activities is the association between machining tolerances and product cost, which is calculated using the normal component cost plus the expected scrap rate. This supports the commercial decision relating product performance to product value, and supports the comparison of expected value with expected cost to determine expected profit. This activity is currently applied in the company; however there is not currently a systematic approach with a supporting knowledge structure. Having conducted interviews with service personnel, designers, manufacturing engineers and people involved in the new product development process, the service elements identified as contributing to the design process have been positioned according to specific design process activities. This refers to aspects of service such as service package design and service location decision. Service location is an input to the machining process design, and later contributes to test system design. Service package design, in the process model shown, is incorporated into the requirements specification. A detailed level service package design process is required as an extension to this model. Currently, manufacturing location is decided in the early stages, however service location is not formally considered. As such, the constraints of those service facilities are not known at the conceptual stage. The detail of the requirements specification activity has been extended to include a range of service requirements. More specifically, the researchers have adapted the Hooks and Farry requirements classification from the literature [14], and combined it with the existing requirements specification format for the case study product at the collaborating company. This represents a combined requirements specification, detailing both product and service. All the requirements were matched with the service knowledge types identified through the interviews with the collaborating company. Table 2 shows

85

these relationships and the location of each requirement in the ontology that was developed.

An investigation is carried out by service personnel (strip and rebuild) at the prototype build and test stage. This informs the design team on physical issues relating to pump disassembly and rebuild. Alongside the manufacturing process design, service process design needs to take place. The detailed description of service process design will be carried out as a future research activity. Service activities include clean, disassemble, inspect, rebuild and test. An example service process was captured, and is shown in Figure 4. The process is implemented in the Protégé knowledge editor tool. It shows the sequence of the activities as well as the resources used by the process. In addition to the service processes, detailed knowledge captured and represented in the Protégé system includes requirements, design features, manufacturing features, machining processes, machining best practices, inspection processes, manufacturing resources (tools, machines), product descriptions, module descriptions, and component descriptions. When the control system and interfaces are designed, the application environment needs to be described and taken into account, as it plays an important role and can affect the performance of the pump. The design process is also incorporated in the Protégé knowledge base. In the detailed implementation there is a class called ‘installation environment’, which describes the environment that the pumps will be installed in terms of heat, humidity and space requirements.

An additional service related input at the conceptual stage is service failure reports. An investigation into the availability and format of failure report data is required in order to specify the content of the failure reports. It is envisaged that the designers could be presented with failure reports to understand the main causes of failures, issues faced with the product and how these were tacked in the past, so that they do not repeat the same mistakes and more importantly, not to ‘re-invent the wheel’ in cases where a solution has been found in the past. An example application is identify whether a relationship exists between pump, previous bearing types, and product failure. Such analyses are complicated by several factors, including the ability to identify single factors or components in a pump failure, and the ability to identify root cause. For example, a thermal seizure may be caused by contact between the rotor and stator. That contact may be due to bearing wear, shaft bending, different rates of thermal expansion, rotor deformation, or a variety of other factors. The service engineer may not be able to correctly identify the root cause. As such, further work is being carried out relating to the capture and presentation of failure data. So far, it is apparent that reports on components that were replaced, descriptions of the applications in which they were used, and any known differences between actual and expected life could support design improvements.

Conceptual design

Project Team

Detailed design

Requirements Specification

Product family definition

Engineering Requirements

Service Location decision

Service location

Manufacturing Location decision

Manufacturing location

Performance Modelling

Manufacturing (machining) process design

Process plan (+NC code)

Performance model (product conceptual model)

Machining process simulation

Simulation report

Prototype build & test

Prototype analysis report

Module structure definition Dynamic modelling

Manufacturability analysis

Shaft dynamics

Feature list – tolerances vs. scrap rate

(module) Cartridge layout

Cartridge model

(module) layout ..x

Service engineer strip & assess Product model

(module) layout ..y

Figure 4: Design process with manufacturing and service inputs

86

Service engineer report

Figure 5 : Integrated ontology screenshot 5 CONCLUSIONS This research set out to develop a knowledge framework to support PSS design. Existing PSS design methodologies in the literature are available to promote communication, from both technocratic and behavioural knowledge management perspectives. There are currently no knowledge frameworks which support the requirement for system level design – a fundamental element of PSS design. The proposed framework enables designers to take into account manufacturing and service knowledge when developing a new product or PSS, both at the conceptual and the detailed design stage. This is enabled by the integration of manufacturing and service knowledge, and the relationships identified between those knowledge elements and design process activities. Our detailed case study shows that manufacturing and service knowledge can be integrated in the design of a technical product and demonstrates how to. It also reveals the difficulties met in the attempt of combining two different areas, i.e. manufacturing and service, due to the difference in the level of abstraction. Manufacturing is mostly product/ component focused and it takes the system slightly into account, whereas in the service area system is taken predominantly into account and a little focus is given on the product/ component. Since the existing design process does not include the systematic development of an integrated technical PSS, our examples are product plus service design rather than true PSS design. The framework developed takes into account the requirement for a system level approach, in order to support true PSS design. 6 FURTHER WORK In order for this research to be completed, two main tasks need to take place. Firstly, the combined ontology (manufacturing, design and service) needs to be validated

by the collaborating company and it should be proved that the framework can provide benefits to the company by the integration of service elements in the new product development process. Secondly, the proposed framework requires validation by another manufacturing company so that it can be generalisable and not just applicable to the collaborating company. 7 ACKNOWLEDGMENTS The authors would like to acknowledge the support of the EPSRC through the Cranfield IMRC in funding this research. The continuous enthusiastic support given by Edwards is also acknowledged. 8 REFERENCES

[1] Doultsinou, N., Roy, R., Baxter, D. & Gao, J. "Identification of Service Knowledge Types for Technical Product-Service Systems", In: 4th International Conference on Digital Engineering Technology (DET 2007), Bath University, UK, September 17-20th 2007, pp 549-555 [2] Baines, T., Lightfoot, H., Evans, S., Neely, A., Greenough, R., Peppard, J., Roy, R., Shehab, E., Braganza, A., Tiwari, A., Alcock, J., Angus, J., Bastl, M., Cousens, A., Irving, P., Johnson, M., Kingston, J., Lockett, H. & Martinez, V. "State-of-the-art in productservice systems", Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, 221 (10), 2008, pp 1543-1552 [3] Morelli, N., (2003), “Product-service systems, a perspective shift for designers: A case study: The design of a telecentre”, Design Studies, Vol. 24, pp73-99 [4] Aurich, J. and Fuchs, C. (2004), "An Approach to Life Cycle Oriented Technical Service Design", in DeVries, M. F. (ed.), Anals of the CIRP, Vol. 53/1/2004, 2004, USA, pp. 151

87

Generic requirements

Service knowledge types

Location in the ontology

Functional and performance

machine uptime, failure types

As instance in ‘incident’, ‘product-failure’ classes

Physical: size, colour, styling

paint

Product feature

Materials, processes and parts use

service process (spare parts use), planned and unplanned maintenance

Strategy->maintenance strategy

Interface requirements

interface issues

As instance in ‘product module’ (product architecture)

Ergonomics + Usability

accessibility, tooling, personnel skills, training

As instance in ‘ergonomics handling’ (product attribute>design feature)

Logistics

logistics (shipping)

System->logistics system

Packing, storage and mobility

packaging

Packaging

Serviceability

serviceability

Product feature

attribute->design

Maintainability

maintainability

Product feature

attribute->design

Availability

machine uptime, product failure record

System behaviour->measured product performance

Reliability

reliability

System behaviour->measured product performance

Safety

safety record / accidents / near misses

‘incident’ slot

Cost: service

service cost

Environmental conditions Reuse and refurbishment

temperature, humidity component (reusability, yes or no), subsystem, service cost, spares

Installation environment Product architecture>component, product module

Disposability

packaging, component, recyclable materials)

Product architecture>component, product module

Test equipment+ testability

service facility + equipment (resources)

Resource->equipment->test equipment

Service features

Service features

System behaviour->measured service features

subsystem

(non-

attribute->design

Table 2: Mapping between generic requirements, service knowledge types and location in the ontology [5] Aurich, J., Fuchs, C. & Wagenknecht, C. "Life cycle oriented design of technical Product-Service Systems", Journal of Cleaner Production, 14 (17), 2006, pp 1480-1494 [6]MEPPS webtool, http://www.mepss.nl/handbook_part1/3UNIQUENESSOF MEPSSAPPROACH.html, Accessed 13th August 2008, 1400 hrs

[7]

Mont, A. and Tukker, A. "Product-Service Systems: reviewing achievements and refining the research agenda", Journal of Cleaner Production, 14 (17), 2006, pp 1451-1454 [8] Badii, A. & Sharif, A. "Information management and knowledge integration for enterprise innovation", Logistics information management,) 16 (2), 2003, 145-155 [9] Huang, J. & Newell, S. "Knowledge integration processes and dynamics within the context of crossfunctional projects", International Journal of Project Management, 21 (3), 2003, 167-176 [10] Young, R.I.M., Gunendran, A.G., CuttingDecelle, A.F. & Gruninger, M. "Manufacturing knowledge sharing in PLM: a progression towards the use of heavy weight ontologies", International Journal of Production Research, 45 (7), 2007, 1505-1519

88

[11] Lee, C., Lau, H. & Yu, K. "An object-based knowledge integration system for product development", Journal of Manufacturing Technology Management, 16 (2), 2005, 156-177 [12] Chen, Y. & Liang, M. "Design and implementation of a collaborative engineering information system for allied concurrent engineering", International Journal of Computer Integrated Manufacturing, 13 (1), 2000, 11-30 [13] Sudarsan, R., Fenves, S.J., Sriram, R.D. & Wang, F. "A product information modeling framework for product lifecycle management", Computer-Aided Design, 37 (13), 2005, 1399-1411 [14] Rios, J., Roy, R., Sackett, P., (2006), Requirements Engineering and management for manufacturing, Society of Manufacturing Engineers (SME), Blue Book Series, Michigan, USA [15] Doultsinou, A., Roy, R., Baxter, D., Gao, J., & Mann, A. “Developing a service knowledge reuse framework for engineering design”, Journal of Engineering design, Accepted December 2008 [16] Baxter, D., Doultsinou, A., Roy, R., Gao, J. “A PLM framework to integrate design, manufacturing and service knowledge”, In: 5th International Conference on Digital Engineering Technology (DET 2008), Nantes University, France, October 22-24th 2008

Comprehensive Complexity-Based Failure Modeling for Maintainability and Serviceability K. T. Meselhy, H. A. ElMaraghy and W. H. ElMaraghy Intelligent Manufacturing Systems Centre, Department of Industrial & Manufacturing Systems Engineering University of Windsor, Windsor, Ontario, Canada [email protected] Abstract:

Failures are the primary triggers for repair and maintenance actions. A clear definition of failure events is important in order to improve maintainability and serviceability. A comprehensive complexity-based mathematical definition of failure is introduced. The applicability of the developed failure model to different complexity definitions is discussed. A new metric is introduced to capture the change in complexity associated with function degradation. A case study is presented to illustrate the application of the new failure definition and metric. The developed approach for failure modeling can be used for maintenance planning. Keywords: Failure, Maintenance, Serviceability, Complexity, Manufacturing Systems

1.

INTRODUCTION:

1.1. Problem statement Current research in failure-induced investigated two main issues:

maintenance

1) Many maintenance activities in manufacturing systems are triggered by machine’s failure. This failure is normally interpreted as physical failure that is easily identified. But, in many cases, the machine fails to perform its intended function, such as maintaining a certain dimensional tolerance, without a noticeable physical failure. This type of functional failure is not precisely defined in literature. However, it may cause significant production losses and lead to increased operational complexity. 2) Machines have two main modes of failure, sudden and gradual. In the sudden failure mode, a machine switches from an operating state directly to the failure state. But in the gradual mode, the machine experiences many in-between states before reaching failure. Nevertheless, most of the reliability and maintenance related research in manufacturing systems use a failure rate model based on the two states assumption. This assumption neglects the nature of the actual machine failure, which leads to ineffective maintenance strategies. 1.2. Literature survey The term “failure” is widely used in daily life and in the branch of reliability and maintainability engineering. From a manufacturing perspective, machine failure is the trigger for corrective maintenance. Therefore, it is important to detect failures as, or even before, they occur. Thus, modern manufacturing systems need reliable failure detection mechanisms. This fact has been emphasized by including the diagnosis ability as one of the key characteristics of new types of manufacturing systems such as flexible or reconfigurable manufacturing systems

CIRP IPS2 Conference 2009

89

[1]. The effect of the used failure detection mechanism on the manufacturing system performance depends on the adopted failure definition [2] : “A common element that is vastly ignored but is rather critical to a sound reliability specification is the definition of equipment failure. Even the most vigorous reliability testing program is of little use if the equipment being tested has poorly defined failure parameters”. There are physical and operational approaches for failure definition found in both academic literature and industrial practice. The physical approach in defining machine failure has been widely accepted where failure is defined as “an undesirable and unplanned change in an object, machine attribute or the machine structure” [3]. Therefore, failure is synonymous with breakdown [4]. The breakdown is characterized by a physical change in any of the machine modules or machine parameters such that the machine is totally unable to continue performing its function. A breakdown of any of the machine tool modules (heads, controls, etc.) is an example of this failure type. The second approach in defining failure is based on the machine operation. Fashandi and Umberg [2] defined failure as: “Any unplanned interruption or variance from the specifications of equipment operation”. An example of the application of this failure definition is used in the quality control charts where it is indicated that a machine is in need for repair if the process carried out by that machine is out of control [5]. Some researches consider operational failure as a symptom of physical failure such as Umeda et al. [3], who defined a failure symptom as the function that has not been performed due to a failure. Physical failures normally lead to operational failures; however, the reverse is not necessarily true. A machine operational failure can happen without being preceded by physical failure. For example, a cutting tool breakage (physical failure) would certainly lead to machine functional failure, while deterioration of machining precision to a level below specifications (operational failure) can happen without any physical failure in the machine or the tool. This concept of functionality versus physical state has been considered by Umeda et al. [6]. They developed a new concept of maintenance where

maintaining the system functionality is emphasized instead of its physical state. Based on this concept, Umeda et. al. [7, 8] developed the Self-Maintenance Machine (SMM) that keeps performing its basic functions even during periods of physical failure. It is clear from this discussion that functional failure of any module is the triggering event for either functional delegation or control action. Nevertheless, a precise definition of the functional failure is still needed.

m

I sys = −∑ log 2 Pi

(1)

i =1

Where : Isys

information content of the system

Pi

probability that FRi is satisfied

m

number of FRs

The previous review shows that there is no unified and precise definition of physical and operational failure in the manufacturing. This ambiguity about failure may lead to ineffective fault detection and hence loss of production capacity.

The information content is a direct measure of the uncertainty of satisfying the function requirements. This uncertainty is therefore a measure of the system complexity [12]. Complexity is expressed mathematically by the following equation:

2- FAILURE DEFINITION Grall et al. [9] assumed that the deterioration condition of any device can be modeled by a stochastic ageing process such that when the system is new, the ageing variable equals zero and when the ageing variable reaches a predetermined level, called failure level (L), the system is deemed to have failed. This model is shown in Figure (1).

Complexity CR = log2 1/Pi= -log2



f ( FR )dFR

design Range

(2) Accordingly, as complexity increases the uncertainty of satisfying the functional requirements also increases. Thus, complexity is a measure of system functionality. Therefore, it is proposed to use the complexity as the machine state variable in the failure model. Consequently, the definition of functional failure can be stated as: “The manufacturing system fails to perform its intended function(s) when its Complexity reaches a predetermined threshold level F”. The determination of the failure threshold level F is a strategic management decision where many factors are to be traded off including cost, product quality, and system availability. This definition is shown in Figure (2), which shows the Complexity change for a typical machine tool as it increases with time.

Figure 1 Failure Threshold Definition The curve in Figure (1) represents the system states at different time instances and shows that the system state variable increases with time till it reaches the failure threshold (L). This model precisely defines the failure by a threshold level of a system state variable beyond which, the system is considered failed even if it is still working. However, Grall et al. [9] did not specify the system state variable on which the failure threshold should be based. Hence, their model is not considered complete. From the literature survey, it can be concluded that both types of failures, physical and functional, lead to the same result; loss of system functionality. Therefore, the system functionality is a suitable system state variable for defining failure. The concept of system functionality is modeled in the Axiomatic Design and Complexity theory introduced by Suh [10]. The design world is assumed to consist of four domains; customer, functional, physical and process domains such that the design process is an interplay between those domains where the design is described in each domain by certain parameters. They are respectively customer wants, functional requirements, design parameters, and process parameters whereas system functionality is described by the Function Requirements (FRs). Suh [11] defined the information content of the system as:

Figure 2 Functional failure definition According to the proposed definition, the machine is considered good for production as long as the Complexity is less than the failure threshold, and as the complexity exceeds the failure threshold (the shaded region), the machine is deemed to have failed. The application of the proposed failure definition can be explained using the example presented by ElMaraghy et al. [13]. Assume that the functional requirement of a machine is to satisfy a specified production demand. This determines the design range to lie between the two extremes of the expected demand. When the machine is new, the machine availability distribution lies completely within the functional design range, hence, the demand would certainly be satisfied and the complexity would be zero. As the machine ages, the failure rate increases and the availability distribution shifts away from the design range and the certainty of fulfilling the demand decreases. Hence, the complexity increases. Assuming the minimum acceptable demand satisfaction certainty is 90%, the Failure threshold = -log2 0.9 = 0.152. Therefore, when the availability Complexity reaches 0.152, the machine is considered functionally failed.

90

Although the developed failure model relies on an uncertainty-based Complexity measure, the model can also be applied to other Complexity definitions. For example, ElMaraghy and Urbanic [14] derived a relationship for process Complexity factor as a function of physical and cognitive efforts of the process tasks. In this case, the process Complexity factor can be considered in order to define the machine functional failure threshold. As the machine ages and its functionality deteriorates, the physical and cognitive effort required by the worker increases in order to maintain the production quality and volume. In this case, a complexity factor threshold can be defined such that when it is surpassed due to the increased required effort, the machine functionality should be restored. 3- FAILURE FORMS Two machine failure forms have been identified in the literature; sudden and gradual. Typically, sudden failure occurs randomly and its time of occurrence is modeled by an exponential distribution of a mean denoting the failure rate [15]. This model assumes that the system has two discrete states; operation and failure [16]. This assumption is inapplicable to gradual failure where the system gradually experiences many states between the two extremes of operation and failure. Therefore, the traditional failure rate is not suitable for modeling it. It is suggested to model the gradual failure by a performance parameter of a value that timely represents the system functional status. Since the complexity is a measure of system functionality as early explained, it is proposed to model the gradual failure by the rate of complexity change, which is named “Complication Rate”. This metric quantifies the machine/manufacturing system functionality deterioration per unit time. Assume that the Complexity at time t is C(t), then

complication rate, υ (t ) =

dC (t )

(3)

dt And based on the developed complexity based failure definition, the failure occurs when the complexity reaches a threshold level, F. this condition is expressed

Figure 3 Complexity Change in Sudden and gradual Failure Sudden failure is explained by the dotted line where complexity increases gradually with time till a random failure suddenly occurs, which causes a significant complexity increase that surpasses the specified failure threshold. This type of failure is modeled by the rate of failure occurrence or simply the failure rate (λ). Gradual failure is modeled by the continuous line where the Complexity gradually increases until it surpasses the failure threshold, which causes system functional failure. 4- CASE STUDY Ott et al. [17] introduced a case study of producing an “air-receiver magnetic assembly”. Samples of size 5 were taken from the production line every shift. The depths of cut of 25 samples were collected (shown in Appendix A). According to the customer wants analysis, it was determined that the producing machine has one functional requirement; low deviation of the depth of cutting with a deviation design range of [-1, +1] mm. Traditionally, such a problem is analyzed using quality control charts. Therefore, the X control chart will be constructed first. Then, the developed Complexity model will be used to analyze the same problem. Ott et al. [17] constructed the X chart as shown in Figure (4) where the upper and lower control limits are: UCL, LCL= 159.6616±3*0.1343.

t

mathematically as

∫ υ (t )dt = F . 0

The total system failure rate would be function of the failure rate λ, the complication rate υ and the failure threshold F. The total failure rate is not expected to be simply the summation of the sudden and gradual failure rates because in most real cases these two failure modes are dependent. Therefore; the total failure rate λT would

generally be expressed as: f ( λ , υ , F ) where the exact relationship is case-specific and its determination requires historical failure and performance data. This proposed new failure rate relationship captures all failure modes. The Complexity changes in sudden and gradual failures are illustrated in Figure (3).

Figure 4

X

Control Chart for the Depth of Cut

Analyzing this control chart according to Western Electric rules, [18] indicates that there is an out of control signal at sample 15. Therefore, a corrective maintenance action has to be performed to restore the machine functionality and bring it back to an in-control state. However, the control chart does not reveal any information about the machine functionality. Therefore, the control charts cannot be used to plan for preventive maintenance.

91

Using the same sample readings, the proposed Complexity-based functional failure metric application would be explained. The system range at each sampling point can be determined; assuming the samples are drawn from a normally distributed population and since the sample size is relatively small (5), then the samples readings follow the student t distribution. System range will be represented in Figure (5) at each sample point by a line segment from X -3S to X +3S where S represents the sample standard deviation. The design range is represented by the shaded area in the deviation range [-1, 1]. Figure (5) illustrates the system range changes with time relative to the design range.

Figure 5 Change of system Range with time This depiction shows any changes in either distribution mean or dispersion, which helps the decision makers understand the change in machine/ process functionality. Using these results, the machine complexity at each sampling point can be calculated using the following steps: Step 1: calculate t values of upper and lower design range limits:

tU = i

tL = i

1− Xi Si

(4)

−1 − X i Si

Step 2: calculate the probability associated with the design range:

Pi = F (tU ) − F (t L ) i

(5)

i

where F(t) is the student t distribution cumulative function Step 3: calculate the machine functional complexity as follows:

Ci = − log 2 Pi

(6)

Figure 6 Complexity Change with Time The regression analysis indicates that the Complexity can be modeled by the indicated equation where x represents the sample number (indication of sample time) and y represents complexity at the time of sample x. Since the samples are drawn from the production line at the beginning of each shift, then, the complication rate of this machine is 0.01 per shift. Therefore, assuming the machine failure threshold is set to be 0.1, then, the machine complexity is expected to exceed the pre-defined 0.1 − 0.012 = 8.8 shifts. Therefore, a threshold at 0.01 preventive maintenance should be planned before that time. Therefore, if this machine has a multi-level maintenance strategy, the duration between any two successive preventive maintenances should be less than 8.8 shifts. If the machine is in a plant that operates 2 shifts per day, 5 days per week, then the least preventive maintenance frequency is every 4.4 days ≈ 1 week. 5- SUMMARY AND CONCLUSIONS A new model for defining functional failure is presented based on the complexity theory. Its main advantages are the formulation of a mathematical failure definition that is applicable to all failures types. The proposed model uses the machine Complexity as a measure of functionality and determines a failure threshold for Complexity. This threshold is case-specific and is determined by experienced decision makers as a trade-off between cost, quality and availability. The “complication rate” term is introduced to measure machine functionality deterioration and gradual failure. It represents the rate of change of Complexity. The complication rate combined with the failure rate completely defines the machine failure behavior. This new approach of failure modeling captures and reveals the behavior of machine functionalities. It can be used to enhance preventive maintenance planning in order to keep desired machine functionalities above certain predetermined level/threshold. The proposed novel complexity-based functional failure metric is applicable to individual products, machines and systems. REFERENCES [1]

The results of these calculations are shown in Figure (6). A linear regression analysis is performed to construct a Complexity trend line as shown by the straight line in Figure (6).

92

Koren, Y., Heisel, U., Jovane, F., Moriwaki, T., Pritschow, G., Ulsoy, G., and Van Brussel, H., 1999. Reconfigurable manufacturing systems. CIRP Annals - Manufacturing Technology, 48(2): p. 527540.

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9].

[10] [11] [12]

[13]

[14]

[15] [16]

[17]

[18]

Fashandi, A. and Umberg, T., 2003 Equipment failure definition: a prerequisite for reliability test and validation. in 28th International Electronics Manufacturing Technology Symposium. San Jose, CA, USA: IEEE. Umeda, Y., Shimomura, Y., Tomiyama, T., and Yoshikawa, H., 1994. Design methodology for control-type self-maintenance machines. Seimitsu Kogaku Kaishi/Journal of the Japan Society for Precision Engineering, 60(10): p. 1429-1433. Hajji, A., Gharbi, A., and Kenne, J. P., 2004. Production and set-up control of a failure-prone manufacturing system. International Journal of Production Research, 42(6): p. 1107-1130. Jensen, W. A., Jones-Farmer, L. A., Champ, C. W., and Woodall, W. H., 2006. Effects of parameter estimation on control chart properties: A literature review. Journal of Quality Technology, 38(4): p. 349364. Umeda, Y., Tomiyama, T., Yoshikawa, H., and Shimomura, Y., 1994. Using functional maintenance to improve fault tolerance. IEEE Expert, 9(3): p. 2531. Umeda, Y., Tomiyama, T., and Yoshikawa, H., 1992 A design methodology for a self-maintenance machine. Edinburgh, UK: IEE. Umeda, Y., Tomiyama, T., and Yoshikawa, H., 1995. Design methodology for self-maintenance machines. Journal of Mechanical Design, Transactions of the ASME, 117(3): p. 355-362. Grall, A., Dieulle, L., Berenguer, C., and Roussignol, M., 2006. Asymptotic failure rate of a continuously monitored system. Reliability Engineering & System Safety, 91(2): p. 126-30. Suh, N. P., 2001, Axiomatic Design, Advances and Applications: Oxford University Press. Suh, N. P., 2005, Complexity, Theory and Applications, ed. MIT: Oxford University Press. Lee, T., Complexity Theory in Axiomatic Design, in Mechanical Engineering. 2003, Massachusetts institute of Technology: Massachusetts. p. 182. ElMaraghy, H. A., Kuzgunkaya, O., and Urbanic, R. J., 2005. Manufacturing systems configuration complexity. CIRP Annals Manufacturing Technology, 54(1): p. 445-450. ElMaraghy, W. H. and Urbanic, R. J., 2004. Assessment of manufacturing operational complexity. CIRP Annals Manufacturing Technology, 53(1): p. 401-406. Ebling, C. E., 1997, An Introduction to Reliability and Maintainability Engineering: McGraw Hill. Kenne, J. P. and Nkeungoue, L. J., 2008.Simultaneous control of production, preventive and corrective maintenance rates of a failure-prone manufacturing system. Applied Numerical Mathematics, 58(2): p. 180-194. Ott, E. R., Schilling, E. G., and Neubauer, D. V., 2005, Process Quality Control. fourth ed: ASQ Quality Press. Perry, M. B., Spoerre, J. K., and Velasco, T., 2001. Control chart pattern recognition using back propagation artificial neural networks. International Journal of Production Research, 39(15): p. 33993418.

93

APPENDIX A: CASE STUDY SAMPLES DATA [17] Subgroup

Sample Sample Sample Sample Sample 5 1 2 3 4

1

160.0

159.5

159.6

159.7

159.7

2

159.7

159.5

159.5

159.5

160.0

3

159.2

159.7

159.7

159.5

160.2

4

159.5

159.7

159.2

159.2

159.1

5

159.6

159.3

159.6

159.5

159.4

6

159.8

160.5

160.2

159.3

159.5

7

159.7

160.2

159.5

159.0

159.7

8

159.2

159.6

159.6

160.0

159.9

9

159.4

159.7

159.3

159.9

159.5

10

159.5

160.2

159.5

158.9

159.5

11

159.4

158.3

159.6

159.8

159.8

12

159.5

159.7

160.0

159.3

159.4

13

159.7

159.5

159.3

159.4

159.2

14

159.3

159.7

159.9

158.5

159.5

15

159.7

159.1

158.8

160.6

159.1

16

159.1

159.4

158.9

159.6

159.7

17

159.2

160.0

159.8

159.8

159.7

18

160.0

160.5

159.9

160.3

159.3

19

159.9

160.1

159.7

159.6

159.3

20

159.5

159.5

160.6

160.6

159.8

21

159.9

159.7

159.9

159.5

161.0

22

159.6

161.1

159.5

159.7

159.5

23

159.8

160.2

159.4

160.0

159.7

24

159.3

160.6

160.3

159.9

160.0

25

159.3

159.8

159.7

160.1

160.1

Informatics-Based Product-Service Systems for Point-of-Care Devices 1

O. Ajai1, A. Tiwari2, J.R. Alcock1 Materials Department; Manufacturing Department, Cranfield University, Cranfield, Bedfordshire, MK43 0AL, UK [email protected] 2

Abstract Informatics related to point-of-care devices denotes the ability to translate stand-alone biological data into meaningful information that can be interpreted to enable and support users in taking the most appropriate steps to aid in managing their health. This paper considers small point-of-care devices used outside healthcare environments, and presents glucometers as an example. The paper seeks to evaluate the current level of servitization of point-of-care testing devices and considers whether they are, or could form, the product-core of a product-service system. The type of product-service system, its informatics requirements, and the services such a system could provide are also considered. Keywords: Informatics, Point-of-care testing, Medical Device, PSS, Product-Service System

1



INTRODUCTION

1.1 Point-of-care biomedical devices Medical devices are used to diagnose, screen, monitor or treat patients. Their primary aim is not ‘pharmaceutical activity’, rather a tool to ‘deliver a service’ [1]. Point-of-care systems offer, according to the National Institutes of Health, ‘laboratory and other services to patients at the bedside’ which may include ‘diagnostics and laboratory testing’ [2]. Point-of-care testing (POCT) has been defined as ‘diagnostic testing at or near the site of patient care’ [3]. Point-of-care devices form a sub-class of medical devices used to carry these tests. This paper concentrates on such devices. The benefits of point-of-care testing rely on the increase in the speed of processing and analysis of biological samples, the speed at which data from the tests may be obtained by the user, or healthcare professional, and therefore, the more timely use of this information as an aid to reach diagnosis and treatment. Examples of commercially available POCT devices, and their associated services, include: 

 



Blood gas monitoring devices: to monitor blood pH, oxygen concentration, carbon dioxide concentration and the concentration of certain electrolytes, as an aid in diagnosis for several medical conditions. Blood glucose monitoring devices: to measure the concentration of glucose as an aid to diagnosing hypoglycaemia or hyperglycaemia. Cardiac marker monitoring devices: to identify markers in the blood as an aid in diagnosing patients with acute coronary syndrome, venous thrombo-embolism and congestive heart failure [4]. Haemoglobin monitors: to measure haemoglobin concentration in the blood, as an aid to the diagnosis of anaemia.

CIRP IPS2 Conference 2009

94

Hand-carried cardiac ultrasound devices: for obtaining echocardiographic images for the assessment of cardiac function [5]. Point-of-care devices can be used in different settings including: the healthcare environment, the home environment and remote locations such as at the scenes of accidents and in battlefield situations. Point-of-care devices used within the healthcare environment tend to be large and fixed in a permanent location. Those used in the home and in remote locations tend to be smaller, compact, transportable and sometimes disposable. 1.2 Medical informatics ‘Medical informatics’ or ‘health informatics’ is the application of computational methods to aid in maintaining the general well being of the body. Informatics is the application of computational methods to data in order to:   

Classify them. Store in a repository once classified. Retrieve the data in an efficient manner when needed. The method of storage will also ensure that they can be retrieved in an efficient manner, e.g. by creating indexes in the data or making associations in the data. Efficiency in this context denotes speeding up the rate at which the process occurs to deliver the information needed.  Disseminate the data effectively to the resource requiring it. 1.3 Product-service systems A product-service system (PSS) ‘is an integrated product and service offering that delivers value in use’ [6]. In a PSS, the product and the service are considered as a single offering. A PSS can be classed as a special form of ‘servitization’ in that it emphasises utilization or performance more than simple product ownership. Three sub-classes can be discerned within PSS: productoriented PSS – the product is sold to the customer but with additional services; use-oriented PSS – the use or availability of the product is sold to the customer, not the

product; result-oriented PSS – a result or capability is sold to the customer, not a product, though the product is still required to support this capability [6]. A service can be described as something done in relation to a product. It may come in the form of maintenance or the supplying of extra products/parts. Baines, et al, describe it as an activity done for others with an economic value [6]. Services provided through point-of-care devices should aid in monitoring and improving the health of the user. ‘Servitization’ is defined as the development ‘of product identity based on material content to a position where the material component is inseparable from the service system’ [7]. PSS can be seen as an example of servitization of products. In PSS terms, point-of-care devices would be the product in the offering. For such products, the degree of servitization of a PSS offering would strongly depend on the level of health informatics support for services associated with the device. This may govern whether a point-of-care device has the capability to form part of a PSS without redesign, and if so, at which level (productoriented, use-oriented or result-oriented) the PSS may function. 2 AIM This paper concentrates specifically on the concept of PSS for those POCT devices designed for use in the home environment. It investigates the following questions: 

What is the current level of servitization of POCT devices for home use?



How does this level of servitization compare to that required for a result-oriented PSS offering?



What informatics resources are in place or would be required in these devices to support a resultoriented PSS?



What would be the benefits to users of such new services?

3

THE CURRENT MODEL OF USAGE OF HOME ENVIRONMENT POINT-OF-CARE DEVICES Blood glucose measurement devices (blood glucometers) provide an example of the current model of usage of POCT devices in the home environment. Glucometers are used on a regular basis by diabetics to monitor their blood glucose concentration and to identify if it is within an appropriate range. Both hypo and hyperglycaemic glucose concentrations carry severe medical risks. Glucometers serve as an interesting example of the current model of use of POCT because: 

They are relatively ubiquitous as examples of point-of-care devices.



Their rate of innovation as products is relatively high. For example, Weitgasser, et al compared four old generation and four new generation models of glucometers in terms of their ‘analytical’ performance. They found that newer devices were smaller, more aesthetically pleasing, easier to use and gave more accurate results [8]. They noted that the improvement in functionality could be attributed to technical improvements in glucometers and the lower blood volumes used for measurement.



They provide a clear example of a situation in which informatics-based services would be of

benefit to the user. Nobel has discussed how improved information generation and exchange is needed to help reduce the ‘morbidity and mortality’ of diabetes [9]. Diabetic care is particularly suited as an application for informatics because ‘its management is characterised by quantifiable outcomes’ [9]. Informatics would aid in improving those methods that may currently be manual and unautomated. Calculations needed to ascertain a diagnosis could be computed quickly thus allowing treatment to be administered to patients at a quick rate. Currently, glucometers are bought as a complete, packaged product consisting of the glucometer, a lancet, test strips and control solutions. Operating instructions are provided through a user manual. Therefore, users generally do not need technical support when using glucometers apart from when the glucometer malfunctions. An example of this is when incorrect units of measurement, for a particular user’s country, are displayed on the glucometer. In such a case, the user is simply instructed by the manual to return the glucometer to the manufacturer for an exchange. In order to ascertain the degree of servitization of glucometers, the state-of-the-art in informatics in glucometers was evaluated in a recent study by the authors [10]. This established that many glucometers provided information in addition to blood glucose concentration. The majority of the glucometers provided error messages indicating that part of the testing procedure had not been followed properly. Certain glucometers displayed a graphical representation of blood glucose concentration plotted against time so that users could monitor trends. A number of glucometers had data management software, which provided off-device data analysis. This included statistical analysis of the data and the facility to generate trend graphs and the ability to easily identify outliers in the data. An ‘electronic logbook’ facility was also present to allow the users to record their results and medications [10]. Some glucometers were supplied with extended measurement functionalities above those of blood glucose measurement. In a survey of 100 glucometers, 11% provided other functionalities (see Figure 1). 8% allowed a blood pressure measurement to be obtained. 1% measured the Ketone level in the blood, while 2% measured Uric acid.

Graph showing the glucometers with multiple functionality 9 8 7 6 Number of 5 glucometers 4 (%) 3 2 1 0 BP

Ketone

Uric Acid

[Functionality]

Figure 1: Glucometers with multiple functionality. The observations noted above suggest that glucometers are currently operating as stand-alone products and do not yet form the product core of product-service systems. However, the extension of informatics and measurement

95

functionalities by a number of manufacturers suggests that the degree of sophistication of this example of POCT devices is tending towards the point where their future incorporation into PSS could become a reality. 4

POINT-OF-CARE TESTING DEVICES IN A RESULTORIENTED PRODUCT-SERVICE SYSTEM

4.1 Drivers for change Wakefield noted that there were multiple drivers for change in healthcare [11]. Although they were mentioned in relation to the nursing sector, they are also applicable to POCT devices. They included: 



Cultural diversity – society is becoming increasingly multicultural with more integration between different races and ethnic origins. Consequently, healthcare services are being tailored to meet this transition. Aging population – the trend in the age of members of society is generally becoming older. The World Health Organization (WHO) notes that an aging population is a challenge that will impact the current century and requires ‘joint approaches and strategies’ [12]. A WHO report noted that healthcare for older people should ensure that they remain independent and continue to play a role in their families and communities [13]. Thus, healthcare ought to be adapted to look after this group of the society, as they will need long-term care. As a result, care can be provided to patients in a number of alternative settings rather than the traditional healthcare environments.



New services and technologies – Wakefield noted that the challenge of time and distance within healthcare was ‘irrelevant’ because of the use of technology [11]. Information technology has been successfully applied in other industries such as banking, travel and communications; however healthcare informatics is still lagging behind [14]. For instance, people are able to monitor their bank accounts and carry out transactions online through the internet. The healthcare environment also needs to be streamlined so that services are delivered accordingly. In an information-driven society where people desire information at their fingertips in any location, informatics will help to improve access to information by allowing the wireless transfer of data between sources etc. Following these drivers for change, a shift in the perspective of POCT devices to a result-oriented PSS whereby customer value is achieved through the provision of services, rather than the purchase of a product is proposed. 4.2 Examples of services Whitney [15] notes that the development of concepts for providing information services within healthcare will be enhanced if issues are considered from a combination of the users and the companies. It is a necessity to recognise the needs from both perspectives in order to ensure that the point-of-care device and the service provided are suited to users. Whitney also provides two interesting examples of generic future services: a ‘medication management system’ and a ‘diet assistance program’ for patients with cardiac conditions. In the medication management system example, a service would be provided that would allow users to monitor the

96

medication they are taking. It would remind the user to take the medication, and notify them of when it had been taken, in order to prevent missed doses or overdosing. In the diet assistance example, an integrated application that monitored diet, exercise and medication for a user was envisaged. This would form the basis of a service customised to individual users. The service would offer advice, based on the individual’s monitored information, on how the user could achieve and maintain a healthy lifestyle [15]. Returning to the example given in the previous section it is interesting to consider what services could be provided based on a POCT device such as a glucometer. Several services that could be provided using a PSS approach are: 

Reminders for the patient to carry out their test (medication management).  The result of the blood test (currently state-ofthe-art).  An option to allow users to input the diet and current lifestyle habits they have as health professionals can provide advice to users on ways to improve this (diet assistance). (A commercial example of this approach currently exists and is briefly discussed in the next section).  An automated report on a regular basis (either weekly or monthly), providing feedback to user of trends in their result. This should be accompanied with advice from health professionals, or information systems, on the progress of the user and how the user can maintain and improve their health condition (condition monitoring).  The logging of extreme results so that if there is a regular occurrence, the user is prompted to contact their health professional. Alternatively the health professional may be alerted through an automated process to contact the patient. Streamlining this would provide real-time advice to patients (acute condition alerts). The first two services already exist in glucometers, and the three latter services have emerged through the gaps found in the literature. In principle, the above services could also be applied to other POCT devices. 5

INFORMATICS RESOURCES REQUIRED FOR POCT DEVICES TO SUPPORT A MOVE TOWARDS A RESULT-ORIENTED PSS A core part of the result-oriented PSS approach towards POCT devices will involve the use of informatics. Health informatics on POCT devices used within the home care environment will need to process data quickly in order to underpin service provision. However, owing to insufficient medical expertise at the point of use, a bottleneck in service-provision is likely to occur. Data will then need to be transferred to an external location where advice may be obtained from health professionals who must then contact the user. A candidate for an approach to solving the problem of this bottleneck in service provision is the ‘telemedicine’ approach. Telemedicine may be defined as ‘the delivery of healthcare and the exchange of health information across distances’ comprising reaching a diagnosis, providing treatment, transferring knowledge and skills to other health professionals and enhancing ongoing research [16, 17]. Bryant, et al [18] have proposed a ‘medical monitoring and patient advisory service’ within the home. This

system would offer medical advice through a rule-based decision support system. GlucoCom (manufactured by Cardiocom) is an example of a glucometer that has adopted the telemedicine approach. The glucometer is linked to another device, which then transfers the blood glucose results to an online system. Health professionals have access to their patients’ data and they can offer ‘timely’ advice concerning changing their medication and monitoring their health [19]. From the examples, one can deduce that the following additional health informatics resources are needed to enable POCT devices to move to a higher degree of servitization. They include:

example from a health professional’s perspective. When health professionals are attending to patients, many issues and questions are considered in order to make a diagnosis. Although information resources are available, access to them may not necessarily be easy and efficient, as health professionals must enter specific search terms in order to find the correct information.

Raw patient data is stored here temporarily.

 

Secure databases to store the patient data. Fast and reliable long-range wireless technology such as GPRS to allow data transfer remotely.  An efficient computational algorithm for cleaning the raw data entered by the patient. Alternatively, the POCT device could be designed in a way that ensures data is entered in a standardised way thus reducing ambiguity and allowing efficient data/information transfer and exchange.  Effective synchronisation of data must also be ensured to prevent inconsistencies arising in the database.  Dedicated health professionals to provide on-demand advice to users. They may not necessarily be in a fixed location; however they do need to have access to communication resources enabling them to send messages quickly and efficiently to the patients. 5.1 A proposed example of an informatics system required for a result-oriented PSS based on POCT devices A proposed view of an informatics system for a PSS for home care point-of-care testing devices is seen in Figure 2. It involves the transfer of the user’s results remotely to a temporary database. A software application is then used to clean and analyse the data before it is saved to a central database. Health professionals are then alerted through the system whenever new results are added to the database that appear to be outliers or show an inconsistency to regular results. Advice is then provided to the patient via a wireless communication link. An important question for the use of POCT devices as part of a result-oriented PSS is whether it would possible to fully automate the actions of point-of-care devices and their supporting informatics. Based on the current level of informatics, it is likely that semi-automation would be possible but not full automation. In contrast to trained health-care professionals, current point-of-care devices do not have the ability to reason through all the available scenarios and options they are presented with. Furthermore, occasionally health professionals need to obtain a second opinion before making final decisions regarding a patient’s diagnosis and currently, this is unlikely to be within the capabilities of a fully automated system. 6

BENEFITS TO USERS OF RESULT-ORIENTED PSS FOR POCT DEVICES Kilbridge and Classen [20] noted that the benefits of informatics in healthcare are to improve ‘access to information, reduce reliance on memory, increase awareness’ and allow procedures to be standardized. Each one is considered from the perspective of point-ofcare testing devices. 6.1 Improving access to information Blumenfeld [21] discussed how information can be best utilized for point-of-care treatment and describes an

DB

User with a point-of-care device.

Advice given to patient

Computer program/software application to clean the data initially.

DB Database/ repository for storing formatted results. Health professionals dedicated to providing advice. They need to be equipped effectively for this.

Key

Data is transferred via Bluetooth, WiFi or GPRS depending on distance

Figure 2: A proposed view of PSS for home care devices. This view was echoed by Poon, et al [22], who, in addition, noted that due to the short amount of time available for doctors to spend with patients, some of the questions may remain unanswered. Blumenfeld [21] discussed knowledge bases and how they may be used to aid in decision support within the health context. One example was that of a knowledge base that could be used to check whether a new drug prescribed to a patient does not interfere with the actions of current medications. This is of particular importance to aging populations whose members are on multiple medications. Such a base may also provide knowledge as to whether the prescribed dose is suitable for the patient, thus also reducing the chances of an adverse drug reaction.

97

Although the example described above was related to point-of-care devices used by health professionals it could be envisaged that the approach could be used for devices used by non-health professionals. However, clearly, the informatics challenges would be considerably greater. 6.2 Reduce reliance on memory Point-of-care devices may incorporate data storage facilities thus reducing the information load on users of the device. For example, glucometers used by diabetics have memory facilities to store the blood glucose results. Therefore users do not need to manually keep a record of their results. Memory facilities on devices will reduce the pressure on users of devices thus facilitating a faster way of health monitoring. 6.3 Increase awareness Informatics enables the addition of alerts into electronic data. Incorporating automated features on point-of-care devices may help to alert users to events that may otherwise go unnoticed. Ford described how the introduction of an electronic system with wireless capabilities to a hospital, alerted health professionals when there were irregularities in patient data (i.e. if a patient’s results fall outside the normal specified threshold) [23]. 6.4 Standardize procedures Informatics enables the standardization of processes. Handheld devices such as PDAs, even though they are not point-of-care devices per se are an example of using informatics in healthcare to standardize procedures. Lapinsky, et al [24] gave an example through a case-study of how a Palm PDA was used in an ICU. Patients’ safety issues may arise when using automated processes. Kilbridge and Classen [20] described several cases where automated processes have been used successfully in healthcare. They are for reporting ‘safetyrelated’ events and for training health professionals. In addition, an automated process can be used to identify clear-cut outliers when analysing data. 7 LIMITATIONS AND EXTENSION TO PAPER This paper has concentrated primarily on POCT devices used in the home environment with a focus on glucometers. Many other POCT devices could have been included and this shows the limitations of this paper. In addition, there are limited publications on informatics relating to POCT devices. The research could be further extended to cover POCT devices that are used in healthcare and in remote environments. 8 CONCLUSIONS The current level of servitization for POCT devices, such as glucometers, has been considered in this paper. From the three sub-classes of PSS, it was initially ascertained that glucometers were currently supplied as products with some additional functionalities such as the ability to carry out other measurements for blood pressure, ketone and uric acid concentration. It was established that the services affiliated to POCT devices relate to the level of information that the device provided to its users. A “use-oriented” PSS model was not suggested, as in this model the usage of the POCT device is sold, i.e. the device is still owned by the manufacturer, and the customers lease the device. (This is not to say that the services required from these devices are not useroriented.) POCT devices used in a home care 98

environment are not suited to leasing as they are generally solely for personal use and are not designed for use by multiple users. This is a health and safety issue to aid in avoiding transfer of infections, which is key in monitoring health and caring for a patient. Hence, a result-oriented PSS has been proposed and its suitability discussed within the context of POCT devices. This was the most appropriate model applicable to improve the level of servitization for POCT devices such as glucometers. Whilst some services, such as providing reminders to carry out a test, already exist, POCT still has some way to go in order to reach the requirements for a result-oriented PSS model. The availability of informatics resources will aid in reaching this goal, as highlighted by this paper. Whilst this paper has concentrated on informatics requirements for POCT-device based services to users, it has been highlighted throughout that actors such as health professionals are likely to have an important role in the service provision. Therefore, there is an intimate relationship between the systems, and system constraints, within which these health professionals have to operate, such as local and national health organisations, and the possible and efficient provision of services. The interaction of health professionals with POCT devices therefore, forms an important future domain of study, without which it is unlikely that the service-potential of POCT-based PSS will be fully achieved. 9 ACKNOWLEDGMENTS The authors acknowledge the EPSRC and Cranfield IMRC for funding this research. 10 REFERENCES [1] Summerhayes, K. and Sivshankar, S. (2006), Introduction to medical device concepts - Diversity, in the challenges of conducting medical device studies, Institute of Clinical Research, pp. 2. [2] MeSH (2008), Definition of Point of care systems [Online]. Available at: http://www.ncbi.nlm.nih.gov/sites/entrez?Db=mesh& Cmd=ShowDetailView&TermToSearch=68019095& ordinalpos=6&itool=EntrezSystem2.PEntrez.Mesh.M esh_ResultsPanel.Mesh_RVFull (accessed 9th April 2008). [3] Kost, G. J. (2002), Chapter 1, in Goals, guidelines and principles for point-of-care testing. Philadelphia: Lippincott Williams and Wilkins, pp. 3-12. [4] Roche (2008), Roche Diagnostics Products [Online]. Available at: http://www.roche.com/home/products/prod_diag/pro d_diag_poc.htm (accessed 20th February 2008). [5] Salustri, A. and Trambaiolo, P. (2002), Point-of-care echocardiography: Small, smart and quick, European Heart Journal, vol. 23, no. 19, pp. 14841487.Sullivan, F. and Wyatt, J. C. (2005), ABC of health informatics: How decision support tools help define clinical problems, British Medical Journal, vol. 331, no. 7520, pp. 831-833. [6] Baines, T. S., Lightfoot, H. W., Evans, S., Neely, A., Greenough, R., Peppard, J., Roy, R., Shehab, E., Braganza, A., Tiwari, A., Alcock, J. R., Angus, J. P., Basti, M., Cousens, A., Irving, P., Johnson, M., Kingston, J., Lockett, H., Martinez, V., Michele, P., Tranfield, D., Walton, I. M. and Wilson, H. (2007), State-of-the-art in product-service systems, Proceedings of the Institution of Mechanical

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

Engineers, Part B: Journal of Engineering Manufacture, vol. 221, no. 10, pp. 1543-1552. Morelli, N. (School of Architecture and Design, Aalborg University) Product service-systems, a perspective shift for designers: a case study – The design of a telecentre. Des. Stud., January, 2003, 24(1), 73–99. Weitgasser, R., Gappmayer, B. and Pichler, M. (1999), ‘Newer Portable Glucose Meters--Analytical Improvement Compared with Previous Generation Devices?’, Clinical Chemistry, vol. 45, no. 10, pp. 1821-1825. Nobel, J. (2006), ‘Bridging the knowledge--action gap in diabetes: information technologies, physician incentives and consumer incentives converge’, Chronic Illness, vol. 2, no. 1, pp. 59-69. Ajai, O., Tiwari, A., Alcock, J. R. (2008), Evaluation of the state-of-the-art in informatics in glucometers. [Unpublished material submitted to Informatics for Health and Social Care journal Nov 2008]. Wakefield, M. (2003). Change drivers for nursing and health care, Nursing Economics [Online]. Available at: http://findarticles.com/p/articles/mi_m0FSW/is_3_21/ ai_n18615834 (accessed 13th August 2008). World Health Organization (2008), The world is fast ageing – have we noticed [Online]. Available at: http://www.who.int/ageing/en/ (accessed 13th August 2008). World Health Organization (2004), International plan of action on ageing: report on implementation [Online]. Available at: http://www.who.int/gb/ebwha/pdf_files/EB115/B115_ 29-en.pdf (accessed 13th August 2008). Altman, R. B. (1997), ‘Informatics in the care of patients: ten notable challenges.’, Western Journal of Medicine, vol. 166, no. 2, pp. 118. Whitney, P. (2008), Designing the experience of health care, Topics in Stroke Rehabilitation, vol. 15, no. 2, pp. 125-130. Eren, A., Subasi, A. and Coskun, O. (2008), A decision support system for telemedicine through the mobile telecommunications platform, Journal of Medical Systems, vol. 32, no. 1, pp. 31-35.

[17] Croteau, A.-M. and Vieru, D. (2002), Telemedicine adoption by different groups of physicians, Proceedings of the 35th Annual Hawaii International Conference on System Sciences, 2002. HICSS. 710 January 2002, Hawaii, IEEE, pp. 1985. [18] Bryant, D., Colgrave, O. and Coleman, R. (2006), Knowledge and informatics within home medicine (KIM): The role of a 'Home Health Hub', International Journal of Healthcare Technology and Management, vol. 7, no. 5, pp. 335-347. [19] GlucoCom (2008), Telemonitoring Tools description [Online]. Available at: http://www.glucocom.com/telemonitoring_device.htm l (accessed 8th August 2008). [20] Kilbridge, P. M. and Classen, D. C. (2008), The Informatics Opportunities at the Intersection of Patient Safety and Clinical Informatics, Journal of the American Medical Informatics Association, vol. 15, no. 4, pp. 397-407. [21] Blumenfeld, B. (1997), Integrating knowledge bases at the point of care., Health management technology, vol. 18, no. 7, pp. 44-46. [22] Poon, S.-K., Rocha, R. A. and De Fiol, G. (2006), ‘Rapid Answer Retrieval from Clinical Practice Guidelines at the Point of Care’, 19th IEEE International Symposium on Computer-Based Medical Systems, 2006. CBMS 2006. June 22-23 2006, Salt Lake City, Utah, IEEE Computer Society, pp. 143 - 150. [23] Ford, A., Wireless glucose results - the latest in realtime data [Online]. Available at http://www.cap.org/apps/cap.portal?_nfpb=true&cntv wrPtlt_actionOverride=%2Fportlets%2FcontentView er%2Fshow&_windowLabel=cntvwrPtlt&cntvwrPtlt% 7BactionForm.contentReference%7D=cap_today%2 Ffeature_stories%2F1007Glucose.html&_state=max imized&_pageLabel=cntvwr (accessed 3rd March 2008). [24] Lapinsky, S., Weshler, J., Mehta, S., Varkul, M., Hallett, D. and Stewart, T. (2001), Handheld computers in critical care, Critical Care, vol. 5, no. 4, pp. 227-231.

99

Service Information in the Provision of Support Service Solutions: A State-of-the-art Review 1 1 2 2 3 3 4 S. Kundu , A. McKay , R. Cuthbert , D. McFarlane , D. Saxena , A. Tiwari , P. Johnson 1 School of Mechanical Engineering, University of Leeds, Leeds, LS2 9JT, UK 2 Institute for Manufacturing, University of Cambridge, Cambridge, CB2 1RX, UK 3 Manufacturing Department, Cranfield University, Bedfordshire, MK43 0AL, UK 4 BAE SYSTEMS, UK

Abstract The transition from the delivery of physical products to the delivery of product-service systems demands new forms of information system that are designed to support the lifecycles of both physical products and associated services. Information requirements for service solutions are dependent on the nature of the offering and the underpinning service agreement. In this paper we provide a survey of current practice, highlighting examples of best practice, and review literature in information support for service support solutions. Results are being used to inform the definition of a blueprint for future service information systems. Early conclusions will be reported. Keywords: Service Information, Support Service Solutions, Product-Service Systems, Industrial Practices

1 INTRODUCTION In the move to product service systems, the delivery of engineering excellence demands the delivery of excellence in both physical products and associated through-life services. Emerging service products strive to deliver availability and capability to customers. As with physical products, the delivery of service excellence begins in the very early stages of the service lifecycle when contracts are developed and agreed. A key to the delivery of service excellence lies in defining contracts that are feasible for delivery. Once a contract has been agreed the service product is developed and then delivered. Access to high quality information (complete, correct, minimal and available to the right people at the right time a form that they are able to use effectively) is key at each phase of service development and delivery: contract definition, service definition and service delivery. This paper highlights key findings from a state-of-the-art review on the current state of service information. Reviews of key developments in the academic literature and a web-based survey of current industrial practice in information provision for services were carried out. Results of the literature review re summarised in Section 2 and an overview of a number of industrial cases highlighting areas of best practice are presented in Section 3. Finally, in Section 4, key lessons learnt are outlined. 2

STATE-OF-THE-ART REVIEW OF ACADEMIC LITERATURE Each broad phase of a service lifecycle (contract definition, service definition and service delivery) has its own requirements for information. A fourth phase, end of life, might well be key in that it is where learning from a service could be collated in a form that can be used to inform future generations of service products. As with all

CIRP IPS2 Conference 2009

100

information systems, they need to be designed to maximize the chances of them being fit for purpose. An early challenge in the development of information systems for service rather than artifact based products lies in the differences between physical and service products; these are summarised in Section 2.1. An overview of what constitutes service information is provided in Section 2.2. Traditional approaches to the design and development of engineering information systems involve analyzing engineering processes to identify information requirements and then satisfying these requirements by bringing to bear knowledge and expertise on the representation of products and the realization of information solutions. Different kinds of service information are needed at different phases of the service 1 lifecycle. For example, LCIA provides a framework for the development of information systems needed to support the delivery of service contracts; as such, it can be used in the establishment of requirements for contract and service definition processes. Service definition results in a definition of a service to be delivered, for example in the form of a service blueprint [1]. Key elements of a service blueprint are the information resources that sit at the bottom of the blueprint, providing information needed to deliver the service effectively, and the process definitions that can be used in the elicitation of requirements for these information resources. The establishment of information support systems requires understanding of the processes that are to be supported: for example, LCIA during contract and service definition and those captured in the service blueprint (or equivalent) for service delivery. A common early activity is to establish service information requirements, both in general and for specific service products; literature related 1

LCIA – Logistics Coherence Information Architecture – www.modinfomodel.co.uk

to service information requirements is outlined in Section 2.3. Once requirements have been established, information systems are developed. Information classification (Section 2.4) is an approach used to deliver as much commonality as possible across information system solutions; this reduces the need to build multiple solutions to the same problem (so reducing maintenance costs associated with the information system itself) and is an enabler for the delivery of high quality information. Literature on information quality is reviewed in Section 2.5. 2.1 Key characteristics of service products Intangibility, inseparability (or simultaneity) of production and consumption, heterogeneity (or variability), perishability and non-ownership are five key characteristics that have been traditionally used to distinguish between physical and service products [2]: •

Intangibility: Services are predominantly performances of actions rather than objects that can be perceived using any of the physical senses.



Perishability: Services must be consumed as they are provided. In general, they cannot be saved, stored, returned or carried forward for later use or sale.



Non-ownership: Largely as a result of their intangibility and perishability, customers do not obtain ownership of services; rather, they experience the delivery of the service.



Inseparability of production and consumption: Service products are typically produced and consumed at the same time - consumption cannot be separated from the means of production.



Variability: Service product quality is subject to variability because services are delivered by people to people. Two dimensions of variability have been identified [2], [3] the extent to which delivery standards vary from a norm, and -

the extent to which a service can be deliberately varied to meet the specific needs of individual customers.

Parallels between these variabilities and those of physical products can be drawn. The extent to which a delivered service varies from a norm is akin to the extent to which a dimension on a physical product varies with respect to its nominal dimension and tolerance band. On the other hand, the variation of a service to meet the needs of individual customers has parallels with mass customisation and the delivery of customised products. Engineering information systems to support the lifecycles of product-service systems need to accommodate these distinctions without compromising the need to preserve commonalities between physical and service products. 2.2 Service information Information has been described as ‘the lifeblood of the organization’ [4] and ‘the most valuable resource in industry today’ [5] but it is also recognized that information is an often undervalued resource because it is difficult to manage. However, if properly managed, the value of information can grow over time. Information is important in service development and delivery as a means of enhancing decision-making processes. Information per se has no direct value but the impact of improved information quality can both reduce costs and enhance service performance. In the context of product servicing, information can provide details about the condition and usage of the product. In a service delivery context, on the

other hand, information provides the contractual requirements of the customer to enable service delivery decisions to be made. For this paper, service information refers to all the information that is required to support the taking of decisions and actions in a service environment. A service information system is a system (which may itself be a collection of systems) which provides the information required to take key decisions and actions in a service environment [6]. 2.3 Service information requirements Information requirements have been discussed extensively in terms of engineering design and information system design. Indeed, in the context of engineering design Court [7] asserts, ‘a large volume of research has been undertaken in establishing these (information) requirements, but many have failed to identify exactly what they are. Much research has proposed to discuss these requirements but only provide details of the commonly used sources of information’. With such a background, it is unsurprising that the information requirements for service specification and delivery are, equally, not well understood. McFarlane [8] asserts that the information requirements for service support solutions are multifaceted and highly dependent on the nature of the offering and the underpinning service agreement. McKay [9] observes that the transition from product to product-service system delivery requires that engineering information systems change to meet new demands to support product data needed for the effective delivery of lifecycle services, including data generated through the whole life of the product, and the rationale behind decisions that were made through life. This is because over the extended time-span of a product’s lifecycle, as opposed to its realisation, the people who created support for this information are increasingly likely to be unavailable to provide comparable support for the definition of both service as well as physical products delivery. Defining information requirements is perhaps the most neglected aspect of the information management process. Berkeley and Gupta [10] survey information required to deliver quality services involving high customer contacts. They classified information required for delivering quality services into three broad categories: input information, process information and output information. Input information refers to the information that are needed before the service is actually being delivered. Process information is the information actually required by the service provider and the service recipient while the service is actually being delivered. Finally, output information refers to information that is available after the service is delivered and as results of the service. Output information can be exploited for future use (e.g. as input information for the next cycle of service delivery, to judge the extent to which the service met customer expectations and needs or to inform the design of the next generation of services) [10]. Zeithaml et al. [11] identify five quality gaps that may occur in delivering services. One of the major reasons for service failure is an inability to bridge these quality gaps. Recommended by Zeithaml et al. [11] and distilled by Lovelock and Wirtz [1] are a number of managerial strategies that should be taken to close the service quality gaps. Several of them are related to proper management of service information. Perspectives on requirements of through-life information of product-service systems in delivering quality service

101

can be found in [12], [13], [14] and [15]. Using the product data framework proposed by McKay et al. [16], it can be argued that effective through-life support services of products requires both product data (i.e. product specification data, product definition data and data related to actual product) and service data (i.e. service specification data, service definition data and data related to actual service) be made available across the product life [14]. At each of the different stages of the lifecycle of a complex engineering product, the needs of the various stakeholders involved are different and distinct. From the viewpoint of general information provision, each of the different stakeholders (with different sub-problems and goals) in a product’s lifecycle has different knowledge requirements [15]. Also, since these stakeholders have a variety of information needs, it is likely that they would make different demands of a knowledge and information management system, such as Product Lifecycle Management (PLM) [15]. In order for knowledge management systems to provide efficient lifecycle support it is necessary to understand their knowledge requirements and the information flows between different stakeholders. A major challenge lies in the generation and maintenance of the flow of appropriate information across and within diverse communities of stakeholders. Designing, servicing, maintaining and upgrading a product are all knowledge intensive activities. However, the information on which these activities depend is often informally captured and may become pertinent information as the design process and lifecycle of the product continues [15]. Often potentially valuable lifecycle information is typically created, gathered and owned by a range of organizations and stored in ways that renders it inaccessible to potential beneficiaries. Also, the quality of this information is not consistent and is highly dependant on the individual agent. McKay et al. [12] argue that the strategy of establishing future-proofed product information to support future lifecycle processes will fail in situations where the information requirements of the processes are not anticipated far enough in advance, usually during product realization when the majority, if not all, of product definition data is created. To address this weakness, McKay et al. [12] propose an integrated product, process and rationale model that allows, throughout the life of the product, the definition of product structures (with associated process enterprise and life-cycle rationale information) that can be superimposed onto relevant aspects of existent product definitions. The product structures can be suited to the lifecycle stage and people concerned rather than predefined earlier in the lifecycle; the inclusion of process enterprise and rationale information allows the context within which information was created to be captured in a way that is comparable to design data. 2.4 Information classification Classification of information provides a means of determining the appropriateness of the information required as the types of information are directly related to the activities that use the information. There are two macro-types of information required in order to reduce the risk to the service provider of moving towards performance-based contracts. The first of these is provider related and aims to quantify how well the service performed against the contractual specification. This information enables the provider to understand where changes to the internal processes are needed in order to deliver the service to its specification. The second of

102

these is customer related and links with the customers’ perception of the service quality. Examples of the first macro-type of information which is provider related are Service Level Agreements (SLAs) and Key Performance Indicators (KPIs), which are usually used to provide performance metrics and gauge the adherence of the service delivered to the contractual requirements. SLAs are specific to identified features of the overall service and, while they provide an indication of the performance, may not give a complete representation of customer satisfaction. The second macro-type of information related to the customer seeks to gain an understanding of the service quality, or perceived service quality (as distinct from quality of service), which may also be described as satisfaction or quality delivered and involves a comparison of expectations with performance. Johnston and Clark [17] describe this, from an operations perspective, as an indication of whether the service specification is being met, and, from a customer perspective, as a mismatch between the customer’s expectations of the service and their perception of its delivery. Classifications of information can be based on internal and external use or sources or into the categories of functional and organizational information [18]. This provides different players, from the service providing organization, with the information required for them to carry out their function. This concept proposes other characteristics relating to the information accuracy, detail, time interval to which it relates and timeliness. Functional information, for example, must be accurate, detailed and provided over short time intervals whereas management information will be less accurate, less detailed and cover a longer time frame. 2.5 Information quality The quality of information is subject to the use of the information [19] and, therefore, the use and quality will define the value of the information. The extent to which a dimension of information quality is important will depend on the purpose for which the information is used. Garvin’s [20] five perspectives on quality can be used to understand elements of service information quality. Berry and Parasuraman [21] suggest quality dimensions for service information based around how relevant, precise, useful, in context, credible, understandable and timely the information is to the user. Wand and Wang [19] propose a set of information quality dimensions which include reliability, timeliness, currency, completeness and consistency. Parlikad and McFarlane [22] also consider similar dimensions in the context of RFID evaluation. Berry and Parasuraman [21] assert that information quality test for these dimension are not absolute and improvement of information quality is a journey of trial and error, experience curve effects, user feedbacks, and new knowledge. Raghunathan [23] investigates how the quality of both the information and the decision-maker impact the quality of the decision. The work shows that the decision quality will only improve with higher information quality when the decision maker has knowledge of the context and problem variables. This is reinforced by the fact that there is a greater significance to information than the knowledge in itself conveys. This is derived from its association with other existing knowledge and implies a dynamic organization of knowledge based on that which is known already [24]. This is supported by the description that information is data with a context.

2.6 Summary of academic developments in service information It is widely recognized that a key to the delivery of excellence through service products lies in the availability of high quality information related to both the service being delivered and the artifacts through which service performance is realized. Key differences between physical and service products (reviewed in Section 3.1) influence the requirements of service information systems. A number of authors have written in service information (Section 3.2) and associated requirements (Section 3.3) in general but establishing a detailed understanding of the information requirements for specific service products still demands understanding of the processes that are to be supported, for example, LCIA for contract and service definition and service blueprints for service delivery. In assessing the quality of service delivery, two kinds of information have been identified in the literature: information related to customer perceptions of the service and information that allows service performance to be quantified in terms of performance indicators (Section 3.4). Finally, in Section 3.5, literature on information needed in service delivery which heavily influences both performance and perceptions of service delivery were reviewed. 3 SURVEY OF CURRENT INDUSTRIAL PRACTICE This section presents key findings from a survey of six industrial cases; the survey was based on information in the public domain either in the literature or on the worldwide web. Table 1 captures uses a common framework to provide an overview of the service systems surveyed. One of the main objectives of the survey of was to identify areas of best practice in delivering support services. Key observations are grouped into three main categories: emphasis on requirements capture and service design (Section 3.1), feedback loops to enable evaluation (Section 3.2), and maintaining competitive advantage (Section 3.3). 3.1 Emphasis on requirements capture and service design Clear and unambiguous requirements and service process definitions lay foundation for efficient management of service information. Rolls-Royce 2 MRMS® (Mission Ready Management Solution), ABB Full Service® 3 , Civilian IT Service Provider and BT’s ‘Shaping New Markets in the Digital Networked Economy’ 4 (SNM-in-DNE) programmes emphasize the need to capture requirements and use systematic ways of defining service processes and offerings. Rolls-Royce captures and communicates customer requirements through SABRe (Supplier Advanced Business Relationship). SABRe is mandatory for all the suppliers and partners who provide products or service that impact upon Rolls-Royce and its customer requirements. SABRe enables Rolls-Royce to assure quality of the products or services delivered to the customers against the contracts by formally communicating Rolls-Royce’s requirements (plus those of the customers & regulatory bodies) and expectations to the supply chain, both in terms of performance and improvement.

2 3 4

www.rolls-royce.com/service/defence/default.jsp

ABB’s Full Service® provides a methodology for defining service processes and offerings. Using collaborative efforts between ABB and the client and following a stage gate process map, Full Service® methodology enables ABB to screen customer’s requirements & business opportunities, identify feasible solutions, develop partnership, define service, define implementation steps of the defined service and manage contracts. In the case of the civilian IT service provider, an area where practice was seen to be of a high standard was the well defined service design process. This involved detailed work between the customer and the supplier at several stages with formal sign off following these phases. The supplier puts a significant amount of effort into mapping the customer’s output requirement to the service providing company’s input requirement in terms of types of information required. The aim of this is to minimize the gaps between the specification process output and the delivery process input. In addition, where the customer specifies services from additional service providers, the organizations liaise early in the design specification phase to ensure that the offerings are compatible and combined to provide the service required by the customer. BT’s ‘Shaping New Markets in the Digital Networked Economy’ programme emphasizes on the importance of in-sourcing and shared risk and responsibility. While on the one hand in-sourcing provides the service provider an opportunity to identify the service needs of the customer better, on the other hand it allows the customer to be a part of the service design and development process and hence, to influence it to the desired ends. For truly strategic partnership and collaboration, the risk and responsibility need to be shared. This is a pre-requisite towards seamless sharing of service information. 3.2 Feedback loop to enable evaluation Another area of good practice identified from the survey of the industrial cases was the presence of feedback mechanisms in the service development process. This enables better evaluation of the service processes and offering against contracts. In the case of the financial service provider, the introduction of what is termed the control centre, drives the use of a capability contract position and drives the development of the IT service as a strategic objective for the company. It enables a full feedback of the performance throughout the service process. The challenges ahead for the service providing company regard the sustainability of the system in place and its scalability with the growth of the company or the inclusion of third party providers in the service supply chain. 3.3 Competitive advantage Service information strategy should be formulated so that it can help service provider gaining competitive advantage over its rivals. Delivery of responsiveness, customization of service offerings and assurance of quality/excellence in delivered service were identified as some of the key factors that can augment competitive advantage of the service provider. Delivery of responsiveness In service operation, the delivery of responsiveness (especially, its call-to-repair or call-to-support commitments to its service customers) and global availability of service levels to its Care Pack Service Customers provides HP huge competitive advantage against its rivals for the similar kinds of support services.

www.abb.com/service/us/9AAC125937.aspx?country=GB www2.bt.com/static/i/media/pdf/cinema_visa_cs.pdf

103

Contract Type

Nature of Offering

Frequency of Delivery

Planned or On-demand

SLA/KPI

Multi- / Single Provider

Formal Process Description

Availability of service design provision.

Provision of IT services; e-mail systems, servers, relocation of company’s IT, etc.

Ongoing with duration of availability contract

Service request is on demand. Delivery is ongoing.

Time to key stage gates such as agreement of requirements and issue of formal proposal. Case by case SLAs with specific contracts.

Often other providers are involved in service design and delivery

Formal process for the design of the service exists.

Availability

Service offering deals with issues related to the issuing and acquiring of debit and credit cards

Ongoing with duration of availability contract

On-demand

SLAs are reviewed monthly. The contract specifies the business rules which drive the SLAs.

Elements of the service may be outsourced by IT but the main service provision is controlled internally

Formal process for the design, delivery and evaluation of the service exists

Availability

Technology to enable financial service

Five years

Both

SLA/KPI in place. Measured by the customer.

No

Process description exists

Either discrete maintenance /coordinated partnership/ availability/ capability

Ensuring operational readiness for air defence

On an ongoing basis for the duration of the contract

On-demand basis for discrete maintenance. As planned for availability and capability.

SLAs/KPIs depend upon individual contract. Typical KPIs are time, cost, quality and responsiveness.

RR global business units with their suppliers, partners, and representatives provide the service

Formal process description exists for a range of services

Capability

Maintenance and improvement of production equipment of industrial plants

Delivered on an ongoing basis for the duration of the contract

Services are delivered as planned

SLAs/KPIs are contract specific. A frequently used KPI is overall equipment effectiveness.

ABB is the only provider of the Full Service

A stage-gate process model includes five principal stages. Formal process description exists for all of them.

Either discrete maintenance / availability

A spectrum of support services to maximize uptime and availability of IT products

Delivered for the duration of the contract

As planned (proactive) or on-demand (reactive) depending upon service contract and choice of service pack

Typical SLAs/KPIs are response time, ease and flexibility, technology coverage (i.e. end-to-end consistency), global availability of service level, and competitive pricing

Multiple providers often HP authorized representatives and providers (who are sometimes HP’s competitors) are involved in delivering services

HP Care Pack Services include 30 standard offerings (or service levels) for an entire IT infrastructure. Process description exists.

HP Care Pack Services

ABB Full Service®

Rolls-Royce MRMS® BT’s SNM-inDNE

Financial Service Provider

Civilian IT Provider

Title

Table 1: An overview of the service systems surveyed.

104

Customization of service offering In ABB Full Service®, ABB and the client jointly screen customer’s requirements & business opportunities, identify feasible solutions, develop partnership, define service, and define implementation steps of the defined service. The joint effort by ABB and the client in these key stages and the presence of an effective backstage IT support service mechanism act as key enablers for delivering highly customized services for industrial plant maintenance. Assurance of quality/excellence In aviation industry, assurance of the quality of the products or service extremely critical. For Rolls-Royce, SABRe acts as the key tool to assure quality/excellence of the products or services delivered to the customers against the contracts. SABRe is the outward facing element of Rolls-Royce’s quality management system. Through SABRe, Rolls-Royce formally communicates its requirements (plus those of the customers & regulatory bodies) and expectations to the supply chain, both in terms of performance and improvement. The requirements in SABRe are about how suppliers interact with RR through their quality systems rather than the detail of what supplier quality systems “should” be like.

5

The following areas of good practice have been identified as being of potential interest to the industrial collaborators, in moving to the delivery of product service systems (shown in Table 2 below).

4 SUMMARY OF FINDINGS In this section, some key lessons from the survey of current industrial practice and review of literature are drawn out as being of potential interest to the industrial collaborators: •

CONCLUSION

The existence of a clearly defined process for the design of the service will provide clarity in the objectives of the delivery phase. Also, it is a good practice to become systematic in the way services are designed and developed.

Good Practice

Supporting Evidence

Do

focus on relationships between customer and supplier organisations

Do

accept heterogeneity in the networks that deliver PSS

RR do this through SABRE – the requirements in SABRE are about how suppliers interact with RR through their quality systems rather than the detail of what supplier quality systems “should” be like

Do

be systematic in the way PSS are designed and developed

ABB have a stage gate process Civilian IT provider has a clearly defined development process

Do

ensure goals of customer are aligned with goals of the supplier …

Screening in the ABB process checks this …

Do

… and if goals are not aligned then stop early.

… and the subsequent stage gate provides a stop point if needed

Do

provide a range of service offerings to suit the needs of different customers

HP do this in their care packages



Having feedback mechanisms into the service development process may help to early diagnose and address gaps and shortcomings in the service offering.

Do



The use of measurable KPIs which provide an evaluation of the service against the specification is key to determining the performance.

have measurable KPIs

BT provides an example of this.

Do

Financial sector provider does this



For services depended on partnership relationships, it is critical to focus on relationships between the customer and supplier organisations and accept heterogeneity in the supply networks that deliver the service. Also it is important to ensure that goals of customer are aligned with goals of the supplier and if goals are not aligned then to stop early.

build feedback mechanisms into the service development process

Do

… unless you make them so generic that they are useless

As models become more general they become more difficult to test, which impacts their reliability

Do

classify information

It could enable quick analysis of the potential scale of service solutions early in the PSS development processes

Do not

assume that there is one set of information requirements for all services …

There is no evidence to suggest there is one set of requirements. There isn’t one set of requirements for all physical artefacts so why would we expect there to be one for all services?



Maintaining alignment of the in-house IT provider with the business “mission critical” processes is shown to improve the delivery process.



The creation of a control centre provides a key point of contact to manage incidents and prioritize the order in which these should be sorted.



Classification of information could enable quick analysis of the potential scale of service solutions early in the service development process.



The notions of ‘seamless sharing/transfer of service information’, ‘in-sourcing’ and ‘shared risk and responsibility’ and also the importance of driving a ‘shift in comfort-zone’ (i.e. shift from cash-based to cashless-based economy and attitudinal changes needed by the customers and other service players) of both the customer and the provider for any major transformation in the delivered service could be of potential interest to the industrial collaborators.

Table 2: Areas of good practice and supporting evidence. These are being used to inform the definition of Service Information Requirements and a blueprint for Future Service Information. With this in mind, the following initial

105

information requirements for the development of information support for product service systems have been identified: •

When establishing requirements for a given service offering, it is advisable to consider customer-supplier dyads and the needs and capabilities of existing information systems in each organisation.



As in the development of information systems for physical products (e.g., in the ISO10303 development methods), key information flows from which requirements are typically elicited might be extracted from PSS development process definitions. In later lifecycle stages, analogous process definitions might be beneficial to the identification of information requirements.



[12]

[13]

[14]

PSS information requirements need to be aligned with the strategic intents of both customer and supplier, and with their delivery capabilities (current and planned). Information systems development might be usefully phased against these capabilities.

6 ACKNOWLEDEGEMENTS The research reported in this paper was carried out under the auspices of the S4T project. Support Service Solutions: Strategy and Transition (S4T) project is jointly funded by EPRSC and BAE SYSTEMS through grant no. EP/F038526/1. 7 REFERENCES [1] Lovelock, C., and Wirtz, J., 2007, Services Marketing: People, Technology, Strategy, 6th ed., Prentice Hall, Upper Saddle River, NJ, USA. [2] Zeithaml, V.A., and Britner, M. J., 2003, Services Marketing: Integrating Customer Focus across the Firm, 3rd ed., McGraw-Hill, Irwin, USA. [3] Palmer, A., 1998, Principles of Service Marketing, 2nd ed., McGraw-Hill, London. [4] Scarrott, G. G., 1985, Information: The Lifeblood of Organization, The Computer Journal, 28(3): 203205. [5] Hollingum, G., and Jowell, R., 1978, Implementing an Information Strategy in Manufacture: A Practical Approach, Springer Verlag and IFS Publications Ltd, UK. [6] McFarlane, D., and Cuthbert, R., 2008, Meeting Notes from WP2 Case Study Workshop, S4T WP2 Case Study Workshop, Farnborough, UK, 28 May. [7] Court, A. W., 1995, The Modelling and Classification of Information for Engineering Designers, PhD Thesis, University of Bath, UK. [8] McFarlane, D., 2007, Service Information Systems, DSI Workshop, Pittsburgh, PA, USA, May. [9] McKay, A., 2006, The Product-service Paradigms: Implications for Information Systems Design, in Proceedings of the IMechE Seminar on Knowledge and Information Management: The Challenge of Through Life Support, IMechE, London, 26 September. [10] Berkeley, B. J., and Gupta, A., 1995, Identifying the Information Requirements to Deliver Quality Service, International Journal of Service Industry Management, 6(5):16-35. [11] Zeithaml, V. A., Parasuraman, A., and Berry, L., 1990, Delivering Quality Service: Balancing

106

[15]

[16]

[17]

[18]

[19]

[20] [21]

[22]

[23]

[24]

Customer Perceptions and Expectations, The Free Press, New York. McKay, A., Kundu, S., and de Pennington, A., 2008, An Integrated Product, Process and Rationale Model for Through-life Information Management, 2nd KIM Conference, University of Reading, Reading, UK, 2-3 April. McKay, A., Kundu, S., and de Pennington, A., 2008, Product Development and the Transition to Product Service Systems, in Proceedings of the IMechE Seminar on Knowledge and Information Management Through, IMechE, London, 24 June. Kundu, S., McKay, A., de Pennington, A., Moss, N., and Chapman, N., 2007, Implications for Engineering Information Systems Design in the th Product-service Paradigm, in Proceedings of the 14 CIRP Conference on Life Cycle Engineering: Advances in Life Cycle Engineering for Sustainable Manufacturing Businesses, Waseda University, Tokyo, Japan, Springer, Shozo Takata, Yasushi Umeda (editors), 11-13 June: 165-170. Eckert, C., Jowers, I., and Clarkson, J. P., 2007, Knowledge Requirements Over Long Product Lifecycles, in Proceedings of the 16th International Conference on Engineering Design (ICED'07), Paris, France, 28-31 August: 913-914. McKay, A., Bloor, M. S., and de Pennington, A., 1996, A Framework for Product Data, IEEE Transactions on Knowledge and Data Engineering, 8(5): 825-838. Johnston, R., and Clark, G., 2005, Service Operations Management: Improving Service Delivery, 2nd ed., Pearson Education Ltd., UK. Flynn, D., 1998, Information Systems Requirements: Determination and Analysis, 2nd ed., The McGrawHill Companies, London. Wand, J. Y., and Wang, R., 1996, Anchoring Data Quality Dimensions in Ontological Foundations, Communications of the ACM, 30(11). Garvin, D. A., 1988, Managing Quality: The Strategic and Competitive Edge, The Free Press, New York. Berry, L., and Parasuraman, A., 1997, Listening to the Customer: The Concept of a Service-quality Information System, Sloan Management Review, 38(3): 65-76. Parlikad, A. K., and McFarlane, D., 2007, RFIDBased Product Information in End-of-life Decision Making, Control Engineering Practice, 15(11): 13481363. Raghunathan, S., 1999, Impact of Information Quality and Decision-maker Quality on Decision Quality: A Theoretical Model and Simulation Analysis, Decision Support Systems, 26(4): 275-286. Marsh, J. R., 1997, The Capture and Utilization of Experience in Engineering Design, PhD Thesis, The University of Cambridge, UK.

An Infodynamic Engine Approach to Improving the Efficiency of Information Flow in a Product-Service System C. Durugbo, A. Tiwari, J.R. Alcock, School of Applied Science, Cranfield University, Cranfield, UK [email protected], [email protected], [email protected]

Abstract In this paper, literature was used to identify and present a worked example of the use of an Infodynamic engine. It shows that: (a) Product-Service System (PSS) information flows can be characterised by such an engine; (b) the engine can measure the efficiency of information flow in a PSS; (c) the engine can be used as a tool to make recommendations about improvements in information flow efficiency in a PSS. Keywords: Product-Service Systems, Information flow model, Efficiency

1 INTRODUCTION Manufacturers in an attempt to stay competitive continually seek opportunities for innovations and value creation. Product-Service Systems (PSSs) are provisions which offer customer value by highlighting the benefits of integrated offerings of product and service artefacts. PSSs are also ‘social constructs’ for realising goals, delivering results and solving problems following information gathering [1]. These social constructs consist of actors and their roles as well as possible scenario and representations. These representations could be used to depict and model components for the delivery of a service or information exchanges within and between systems. Established links and communication paths can then be used to support information flow especially in integrated processes involving flow of materials [2]. Modelling this information can be crucial in identifying repeatable and inefficient system processes as drivers for improving quality, efficiency, and financial performance in accordance with ISO 9000 and ISO 14000 [3]. This paper begins with an overview of the PSS concept. The Infodynamic engine approach will then be introduced. This information flow model offers a high-level abstraction of information exchange. Its selection was based on three criteria: efficiency, adaptability and reusability. Information flows in a PSS will then be demonstrated based on a worked example of the Infodynamic engine. Finally, recommendations based on the findings during the worked example will then be used to suggest improvements in information flow efficiency in a PSS.

hardware/software (HW/SW) codesign methodologies. Codesign or collaborative design recognises the need for designs to realise a goal [6]. Design activities in this approach can run concurrently, improving the time-tomarket and optimising solutions. A PSS is a process system which offers avenues to shift company and organisational focus from ‘product thinking’ to ‘system thinking’ in an enterprise [7]. It is a system consisting of product and service systems as well as productized services and servitized products (Figure 1). Servitization in PSS is a practice which closely links and incorporates services with offered products for servitized products just as productization does the converse for services. Products are tangible and domain specific whereas the service offering is made up of intangible artefacts or services such as upgrades and recycling.

2 PRODUCT-SERVICE SYSTEMS A PSS is a concept which offers a medium to foster competitiveness and promote sustainability for the manufacturer and the environment. It is a functionoriented business model [4], in other words it is a technical approach, which seeks to offer commercial value for manufacturers while adding value for customers. Function oriented design is a strategy which involves the decomposition of a system into interacting units [5] while the business model offers value creation. The PSS approach promotes the need for codesign of product and service subsystems much like traditional

The domain of application, investment and resources are important considerations for choice of service strategy. A service is an activity offered by the one who provides to affect or influence the one who receives. Tomiyama [8] identifies three important features of a service: service environment, service content and service channel. He identifies material, energy and information as forms of service content which are consumed by the service channel within the service environment. Much like a service, information is non-physical but unlike a service, information flow and exchanges especially in organisations and companies usually requires some form of integrity, confidentiality or security. Information models

CIRP IPS2 Conference 2009

107

Figure 1: PSS model.

in these institutions could aid in identifying patterns and manners of information exchange. 3 THE INFODYNAMIC ENGINE An ‘Information Engine’ or ‘Infodynamic Engine’ approach to model information flow, with considerations of efficiency in systems, has been presented by Sundresh [2]. This model was selected for use here, because it is based on the Carnot Cycle; which is the most efficient, known thermodynamically cycle supporting reversible processes [9].

refer to system configurations or arrangements which combine to give a system’s macrostate. For instance, fluids (gas and liquid) can be thought of as a macrostate, for which concentration of samples (analytes) and fluid dynamic phenomena (fluidics) can be microstates. ‘Information potential’ is a concept of the model used to describe information held by either A or B, which the other needs to complete a task. Entities with a higher information potential can satisfy the needs of those with a lower information potential. ‘Information need’ describes this concept in relationship to a particular application. It is defined as ‘additional information’, i.e. over and above that already existing in the system, required for realising a goal or performing a task. This is equal to the probability of successfully performing a task without possession of information, m, i.e. [2]:

N  P (m )1

Figure 2: Information Engine Generation Cycle [2] This Infodynamic engine can be applied for two cooperative, interconnected or integrated processes (or systems) A and B; integrated for the joint realisation of a goal or task by information exchange. The ‘Engine’s’ aim is ‘to generate a new piece of directly usable information from the information already in the possession of A and B’ [2]. 3.1 Measures and operation ‘Physical entropy’ and ‘information need’ [2] are the two measures employed in the Infodynamic engine. The physical entropy can be computed as the sum of the algorithmic randomness and the statistical disorder of a given data d i.e. Sd  K (d )  Hd

(1)

Where Sd is the physical entropy, K(d) is the algorithmic randomness (or Kolmogorov complexity) of d and Hd is the statistical disorder which is dependent on the conditional probability of the data set. Zudek [10] refers to the statistical disorder in terms of missing information and the algorithmic randomness with regards to the known randomness (or disorder). The algorithmic randomness (K(d)) for a given data d with length d* is defined as the shortest path which produces d and is calculated as K (d )  d 

(2)

The statistical disorder (Hd) provides information about microstates which are unavailable regardless of the presence of d. This statistical disorder can be computed as [10]: Hd  

P

kd

log2 Pk d

(3)

k

Where Pk|d is the conditional probability of the microstate k given macrostate d Each microstate k can be described in terms of this conditional probability (Pk|d) with respect to a macrostate d which it constitutes. A microstate is used in this context to

(4)

Where N is the information need, ¬ means ‘without’ or ‘not’ and P is probability. From the model shown in Figure 2, four phases can be identified. In the first phase (a-b), A is the source and B the recipient of the data. Next, B starts to process the data in isolation. The next stage, b-c, corresponds to an information transformation stage in which the information is processed with respect to the task to be done. At point c the data has now been transformed and begins the process of being separated, ‘crystallized,’ into the parts usable and unusable for the specific application being considered. In the final step, A and B, ‘reconcile’, i.e. come up with a piece of information that B needs, based on the prior processes. The reader is referred to reference 2 for the detailed information entropy and complexity arguments which underlie this model. 3.2 Analysis of information flow In terms of analysis, some characteristics of information exchange can be used to study the information flow such as value of the ‘usable information’ and ‘information flow efficiency’. The value of the usable information is the first simple measure, which is determined from the product of the information content and the information need i.e. (H 2  H1 )  (N1  N2 )

(5)

Information flow efficiency can be determined from three concepts of efficiency [2]. These concepts are:  Information generation efficiency (IGE) – comparison of the algorithmic information generated with the total information available;  Information utilisation efficiency (IUE) – comparison of the algorithmic entropy for satisfied need with the initial need;  Information system efficiency (ISE) – comparison of the information value generated with the total maximum information value in the data received. It compares actual use to possible use of information. These concepts are based on the entropy change between H1 and H2 as well as the information need value difference between N1 and N2. They can be computed as follows [9]: IGE  1  H1 H 2

(6)

IUE  1  N2 N1

(7)

ISE  (H 2  H1 )  (N1  N2 ) (H 2  N1 )

(8)

108

4. Information Flow in a Product-Service System For PSS, with integrated processes for product P1 and services S1, an Infodynamic engine can be demonstrated. The Infodynamic engine can be either based on a predictive model of eitherP1 or S1 as dictated by the domain of application. For instance, in the transport sector, identifying a new transport route service (S1) in a town can be used to decide the means of transportation (P1) such as ferries or double-decker buses. Similarly, new high speed trains (P1) may be acquired and its facilities can be used to decide its use for service provision such as carrying postal freight or intercontinental connections (S1). An example of a PSS implementation based on the operation of a major healthcare provider will now be applied to demonstrate the use of the Infodynamic engine. This company offers a ‘service agreement’ based on providing mission-critical equipment for sale backed, with 24 hour service for remote clinical and technical expertise. Products sold include X-ray machines, CT, MR, ultrasound and nuclear medicine imaging equipment. Service support for software is provided by means of updates while services are offered to support hardware facilities by means of planned maintenance, exchange or parts replacement/ repair. Information exchanges will be identified for which the Infodynamic engine will be used to generate new pieces of directly usable information. This information can then be used for the delivery of a service. Information exchange is used in this context to describe a phase in which information are accessed or delivered between systems or within a system. Information flow on the other hand is used here to describe the phases in which a new piece of directly usable information is generated from information already possessed by the system(s).

3.3 A PSS worked example A simple case of an information exchange is presented based on the healthcare provider’s ‘service agreement’. For demonstration purposes, it is assumed that four forms of provisions (AS) can be offered in this PSS – sold products, planned maintenance, parts replacement and parts repair. Similarly, for a service to be provided, three pieces of information (IS) will need to be determined – the availability of service, the right to receive the service and the availability of a service team to provide the required service. The information flow centres on the need to replace a fault part in a CT (computed tomography) machine and to follow that up with a scheduled maintenance. As an example, for this case, four pieces of information, to uniquely identify a service request (UI), are obtained from a laboratory technician in the company hosting the CT: product name, model, company name and department. Four additional services (AOS), outside the scope of those of the service provision depicted in Figure 3a, will also be required to fulfil customer requirements: parts procurement, transportation, installation and disposal of old parts. Similarly, two additional pieces of information (AOI) are required: the location of the parts and the department to deal with request. Information exchange In this example, a laboratory technician requires replacement parts and maintenance for a CT machine (See Figure 3a). Information exchange can include the following steps: i. Technician requests assistance via phone support; ii. A request for service is made by a support staff; and iii. The service is delivered by the service team and the system updated by a support staff.

Figure 3: Path in an information exchange for service provision (a.) actual exchange (b.) Infodynamic engine representation

109

The actual exchange is represented in Figure 3a. In database management, indicated by dotted arrows, each arrow indicates that the support staff either feeds in inputs or gets outputs from the stored data. For telephone conversations the broken arrows imply a call to or from a source. The path involving, Request service → Check service → Check orders → Update orders in Figure 3a equates to step i where the technician requests assistance. In step iii, service delivery and system update is accomplished by Place order for service → Feedback (to technician) → Feedback (from service team) → Update record. Step ii, in which the request is made, is realised by, Check service → Check orders → Update orders → Update record. The ‘check service’ is used to access the availability of a service and to check if the company for which the laboratory technician works can be provided with the service. The ‘check order’ on the other hand is used to access the availability of the service team. Information Flow This example will focus on the information flow between the support staff and laboratory technician. This can be mapped unto the Infodynamic engine as shown in Figure 3b. This is because the Infodynamic engine is initiated by an ‘obtain data’ (a-b) phase as shown in Figure 2. This equates to the ‘Request service’ in Figure 3a. The ‘assimilation’ (b-c) phase in which information is generated corresponds to the Check service → Check orders → Update orders. The crystallised data during phase c-d is used to accomplish Place order for service → Feedback (to technician) → Feedback (from service team). Similarly, phase d-a in which P1 and S1 are reconciled corresponds to the ‘Update record’ process in Figure 3a. In an ideal Infodynamic engine, as shown in Figure 2, phase a-b matches c-d i.e. the entropy difference between P1 and S1 when service data is obtained must match the reversed process when the service data is crystallised into usable and unusable parts. Similarly, information need variance at phase b-c must match d-a. At phase b-c, information need moves from high to low as information is processed. This is then reversed at d-a, i.e. low to high to reflect the presence of a higher information potential. This high information potential highlights the need for additional information to complete a task. The macrostate in this case would be the service to be provided while microstates are encompassed by AS and OAS. Frequency of request for these services can be automatically logged and used to determine conditional probabilities (earlier identified in eqn. 3). For instance, if the support staff receive 50 requests for service (the macrostate) and 10 of these requests are for planned maintenance; then the conditional probability for the planned maintenance is given as 10 out of 50 (or 0.2). Table 1 presents the conditional probabilities which have been simulated for use in this worked example. Service Type Artefacts Supplied (AS)

Other Artefacts Supplied (AOS)

Service Component sold products

Conditional Probabilities 0.6

planned maintenance

0.2

parts replacement

0.1

parts repair

0.1

Transportation

0.5

Parts procurement

0.2

Installation

0.2

Disposal of old parts

0.1

Table 1: Service request frequencies for worked example

H1 is the entropy at the point of service request and k (see eqn. 3) denotes the individual microstates for the service (or macrostate) requested. H2, in contrast, is the entropy at the point before it is delivered and k denotes the individual microstates required to realise the service. N1 is information need at the point of request and the value of m (see eqn. 4) is the number of information items required to distinguish a service receiver. N2, on the other hand, is the information need at the point when a service has been identified and is ready to be delivered. At this point, m can be computed from both the information required to identify a particular service receiver and any other additional or internal information in an organisation used to identify a particular user. For the laboratory technician (as the receiver), data about two services is being sought i.e. K(d) at H1 is equal to 2. These services, planned maintenance and parts replacements have service request frequency (or conditional probability) values of 0.2 and 0.1 respectively (see Table 1). Consequently, from eqn. 3, H1 can be calculated as  0.2 log2 0.2   0.1 log2 0.1  2  2.8

For the support staff (as the provider), two services and four additional services (OAS) are identified i.e. K(d) at H2 is equal to 6. Using Table 1 as before, the service request frequency values can be identified. From these parameters, H2 can be computed as 0.2 log2 0.2  0.1 log2 0.1  0.5 log2 0.5     6  8.6  0.2 log2 0.2   0.2 log2 0.2  0.1 log2 0.1

Total information in the exchange is the sum of information about the availability of a service (IS), the receiver (UI) and information of additional services required to realise service requested (AI) i.e. 3+1+2 = 6. For this case, it is assumed that all information required to complete a task have the same probability. For information need, the laboratory technician provides a unique ID from which the name, model, company and department is used to identify the product in P1. Additional information is then obtained to aid the delivery of the service by S1. Consequently m at N1 equates to 0.5. From this information, N1 can be calculated as

 1  0.51  2 Similarly, the value of N2 can be computed as

 1  0.1667 1  1.2 The value of m at N1 is 0.1667 (or one sixth) because the unique ID supplied is the only data obtained from P1 by S1. 3.4 Reengineering for efficient information flow As demonstrated, information exchange in the provision of the service for the laboratory technician can be mapped unto the Infodynamic engine. However, the information exchange process could be further optimised based on the identified information flow path. For this reengineering process, two approaches could be applied: i. Aggregating and restructuring information to reduce processes and exchanges; or ii. Reducing the number of actors and scenarios in the information exchange for which processes can be automated. Both approaches as shown in Figure 4 adopt separate schemes to handle redundancy inherent in the processes involving 3-4-5-6 which is shown in the Infodynamic engine representation in Figure 3b.

110

Figure 4: Re-engineering the information flow (a.) aggregating and restructuring information (b.) reducing actors and scenarios (c.) optimised information flow Restructured Information Exchange In the first approach (aggregating and restructuring information), the information system can be configured to perform order processing and to provide and receive feedback to and from the laboratory technician and the service support team as shown in Figure 4a. For the second approach (reducing actors/scenarios), adequate automation facilities can be used to reengineer the information exchange in the PSS (see Figure 4a). The information system could be automated to check and update orders when a check for service is performed. The World-Wide Web or a virtual private network could be used to link systems at the request and delivery ends of the service. Information flow, in both approaches, as shown in Figure 4b, is established in four steps. Step i involves a request for service which may be automated. This request would also advise on the available orders and could allocate available resources as required. Step ii would involve alerting the service team of the need for a service provision. Step iii (or 7 as in Figure 4c) involves feedback from the service team indicating service delivery. The final step, step iv, would involve updating the database to reflect the completed service provision. Efficiency comparison of Reengineered Information Flow In the original example, the value of the usable information based on information need and physical entropy can be computed as (eqn. 5):

(8.6  2.8)  ( 2  1.2)  4.6 The Information generation efficiency (IGE), Information utilisation efficiency (IUE) and Information system efficiency (ISE) can likewise be computed as (eqn. 6,7,8) 0.67, 0.4 and 0.27 respectively. In the reengineered information flow (Figure 4), the efficiency measures applied are based on schemes to improve information need. In these schemes, the generation of information is maintained but the utilisation is improved to increase efficiency of the information system.

111

For the first approach (aggregating information), three reports containing aggregated information can be generated. These reports could contain information about the availability of a service (IS), the receiver (UI) and information of additional services required to realise service requested (AI). For this approach, the value of m is 1 and 2 for N1 and N2 respectively. This is because both UI and AI are required to make the service available while only UI is required to deliver the service. The value of the usable information can then be calculated as:

(8.6  2.8)  (3  1.5)  8.7 The new IUE and ISE can also be computed as 0.5 and 0.34 respectively. For the second approach (reducing actors/scenarios), the information flow and exchange is restructured. In this new scenario, the information flow is now focused on the support staff and the networked information systems. In terms of information need, N1 is now based on IS and AI while N2 is concerned with only AI. The values of N1 and N2 are calculated as 6 and 2 respectively. The value of the usable information can then be computed as:

(8.6  2.8)  (6  2)  23.2 The new IUE and ISE can also be computed as 0.67 and 0.45 respectively. 4 DISCUSSION Designing and managing the life cycle of a product can be a complex task because considerations have to be made for issues such as obsolescence, costs risk and uncertainties, development costs, process management and concurrent design/development. Integrating design for service in these activities can further complicate the design and management process. In PSS, where value creation is used to drive a functionoriented architecture, the need for customer focused models and efficiency driven tools are also particularly important. The networks realised from an established

information flow model can be used to define the information system in a PSS. These models however have to be developed with flexibility in mind, so as to ease time and financial costs associated with organisation shifts [7].This flexibility can also have the added benefit of fostering the development of conceptual models. Applying an information model such as the Infodynamic engine in PSS can be beneficial in terms of efficiency and reusability. As demonstrated in the worked example, a path for information flow can be established based on information exchanges. This efficiency is driven by increasing value of generated usable information. Additional optimisation schemes can be implemented to improve the physical entropy to further increase the overall efficiency of the information system.

5 SUMMARY This paper has attempted to identify information exchange in a Product Service System (PSS) and ways of modelling the flow of information. The concept of PSS was first presented as a function-oriented principle offering value for both manufacturers and customers. The Infodynamic model was then presented as a high-level approach to modelling information flows. The Infodynamic engine was then applied in a PSS for information processes as a way of generating new directly usable information for the delivery of a service. The potentials of the tool for recommending improvements were also highlighted and measured with simple information flow efficiency metrics in a PSS.

4.1 Infodynamic engine mapping process As noted in the worked example, identifying and mapping information exchanges for the Infodynamic engine is an activity which requires human judgement. A key consideration for this activity is the source of data which starts the information generation process. In the worked example, data was obtained based on an interface created for information flow between the support staff and laboratory technician. Once this interface is determined the rest of the mapping process for information exchanges can be achieved by collaborating teams of domain experts, system analysts and even stakeholders. When these exchanges have been identified, the operation of the Infodynamic engine can then be automated. In the real world, support for the mapping process in organisations is realised by tools known as decision support system (DSS) [11]. These tools are required to identify the information flow from information exchanges. They are also useful in terms of matching customer requirements based on manufacturing capabilities. A DSS can also be configured to capture technical specification of customer requirements; generate and select alternative configurations to rapidly provide accurate technical solutions; and schedule resources for service commencement. For PSS, as in modern organisations, the Infodynamic engine can be particularly useful. This is because modern organisations are characterised by a collaboration of human judgement and computational facilities [11]. Computation facilities are required for coping with possibly large information quantities while human judgement in this respect refers to customer needs being fulfilled with solutions that satisfy their requirements without jeopardising business objectives of manufacturers 4.2 Limitations of the Infodynamic engine The Infodynamic engine is a tool for modelling information flow at a high-level of abstraction and is limited to a sequential representation of information flow extracted from information exchanges. The engine examines information need in the delivery of a service by a PSS but criticality or timing of the information for the delivery of a service is not identified or supported. It is also vital to note that the Infodynamic engine is not a tool for comprehensively modelling information in a PSS. Neither is it a tool for the implementation of a system or reengineering a PSS for increased servitization or productization. These limitations of the Infodynamic highlight the need for comprehensive information modelling down to the lowerlevels of data modelling and even language specification for concrete/detailed model to offer support for systems implementers. Further considerations could include sourcing avenues for ensuring integrity, privacy and confidentiality of information exchange between consumers and manufacturers.

6 ACKNOWLEDGMENTS The authors would like to extend their sincere thanks to the EPSRC, for its support via the Cranfield IMRC, towards the work carried out in the preparation of this paper. 7 REFERENCES [1] Morelli N, 2006, Developing new product service systems (PSS): methodologies and operational tools, Journal of Cleaner Production, vol. 14, issue 17, pp. 1495-1501 [2] Sundresh TS, 1997, Information complexity, information matching and system integration, IEEE Transactions on Systems, Man and Cybernetics, vol. 2, pp. 1826-1831 [3] Castka P and Balzarova MA, 2008, The impact of ISO 9000 and ISO 14000 on standardisation of social responsibility—an inside perspective, International Journal of Production Economics, vol. 113, issue 1, pp. 74-87 [4] Tukker A, 2004, Eight types of product-service system: eight ways to sustainability? Experiences from SusProNet, Business Strategy and the Environment, vol. 13, no. 4, pp. 246-260 [5] Sommerville I, 1992, Software Engineering, 4th edition, Addison Wesley, London [6] Wolf W, 2003, A Decade of Hardware/Software Codesign, IEEE Computer, vol.36, no.4, pp 38-43 [7] Baines T, Lightfoot HW, Evans S, Neely A, Greenough R, Peppard J, Roy R, Shehab E, Braganza A, Tiwari A, Alcock JR, Angus JP, Bastl M, Cousens A, Irving P, Johnson M, Kingston J, Lockett H, Martinez V, Michele P, Tranfield D, Walton IM, & Wilson H, 2007, The state-of-the art in Product Service Systems, Proceedings of the I MECH E Part B Journal of Engineering Manufacture, vol. 221, no. 10, pp. 1543-1552 [8] Tomiyama T, 2001, Service engineering to intensify service contents in product life cycles, Second International Symposium on Environmentally Conscious Design and Inverse Manufacturing, pp. 613 - 618 [9] Sundresh TS, 2000, Information flow and processing in anticipatory systems in IEEE Transactions on Systems, Man and Cybernetics, vol. 1, pp. 271-276 [10] Zurek WH, 1989, Algorithmic randomness and physical entropy, Physical Review, vol. 40, issue 8, pp 4731-4751 [11] Patel A, Meade R, O'Sullivan D and Tierney M, 1996, Information modelling for service provisioning - the DESSERT experience, Computer Standards & Interfaces, vol. 18, issue 2, pp. 175-18

112

A Periodicity Metric for Assessing Maintenance Strategies K. T. Meselhy, W. H. ElMaraghy and H. A. ElMaraghy

Intelligent Manufacturing Systems Centre, Department of Industrial & Manufacturing Systems Engineering University of Windsor, Windsor, Ontario, Canada [email protected] Abstract

The maintenance policy in manufacturing systems is devised to reset the machines functionality in an economical fashion in order to keep the products quality within acceptable levels. Therefore, there is a need for a metric to evaluate and quantify function resetting due to the adopted maintenance policy. A novel metric for measuring the functional periodicity has been developed using the complexity theory. It is based on the rate and extent of function resetting. It can be used as an important criterion for comparing the different maintenance policy alternatives. An industrial example is used to illustrate the application of the new metric. Keywords Maintenance, periodicity, complexity, manufacturing systems

1

machines, systems or industry. The maintenance actions are normally classified as shown in Figure (1) [3] :

INTRODUCTION 1.1

Problem description

Manufacturing systems are planned, controlled and maintained with the objective of supplying products with predetermined quality level and maximizing capacity utilization. The system features and capabilities are designed a priori. At the beginning a manufacturing system performs as designed. As time passes, the machines age and un-planned failures occur causing the system performance to drift away from its initial state. Therefore, the function of a manufacturing system must be periodically restored, which is practically achieved by the maintenance operations. This periodic resetting is performed to ensure that the machines are kept in an acceptable condition throughout their useful life. Therefore, it can be said that one of the main outcomes of the maintenance action is introducing periodicity into manufacturing systems and re-initialize their functional state. Nevertheless, there is a lack of reported literature that assesses maintenance from a periodicity perspective. Furthermore, a metric is required to evaluate the effectiveness of maintenance strategies and support decisions regarding designing a new maintenance policy or re-designing an existing one in response to changes in the manufacturing system. Such strategies and metrics should be simple to use to facilitate its application in today’s changeable and reconfigurable manufacturing environment [1]. 1.2 Literature Review Maintenance in manufacturing systems has to provide the required machine reliability, availability, efficiency and capability [2]. The current research is focused on the administrative maintenance actions at the policy level rather than its detailed technical aspects at the machines level and proposed general approaches and high level metric, which are not restricted to certain type of

CIRP IPS2 Conference 2009

113

Figure 1 Classification of Maintenance Actions Corrective maintenance includes all actions performed as a result of failure to restore an item to a specified working condition, while Preventive Maintenance (PM) includes all actions performed on operating equipment to restore it to a better condition [4]. The maintenance strategy is a structured combination of the above mentioned maintenance actions [2], which describes the events (e.g. failure, passing of time, certain machine condition, etc.) and the type of action they trigger (i.e. inspection, repair, maintenance or replacement). The literature contains more than thousand maintenance policies/strategies [5], which can be categorized as follows: Age dependent PM polices: the PM actions (minimal, imperfect or perfect) are triggered by the age of the component such as (T, n) policy [6]. Periodic PM policies: the PM are pre-planned at fixed time intervals [7] and [8]. Sequential PM policies: PM is carried out at age dependent decreasing time intervals [8, 9]. Repair number counting and reference time policies: the maintenance action depends on the number of previous failures and the item’s age at the time of maintenance [5]. Failure limit policies: the PM is carried out when the unit failure rate or reliability indices reach a predetermined

level. The intervening failures are corrected by repairs [10]. Reliable evaluation methods are needed to compare the effectiveness of these numerous and diverse maintenance policies/strategies. Different criteria have been used in the literature to assess maintenance strategies such as cost, [11], Availability [12], Reliability [13], and quality [14]. This brief overview of the existing maintenance evaluation methods and criteria highlights the need for a new criterion to evaluate the main role of maintenance strategies in defining the required and sufficient frequency and extent of the maintenance actions. 2

MAINTENANCE STRATEGIES

AND PERIODIC

SYSTEM COMPLEXITY

considered as a time dependent complexity. Hence, carrying out maintenance actions would serve to re-set the system performance characteristics. It is clear from this discussion that periodicity is important for the long-term system functionality. However, how often a system should be reset, to what extent the design parameters should be re-set, what is a desirable level of periodicity and at what cost, remain un-answered questions. Another important question is how much periodicity does exist given a certain maintenance regime and how much periodicity is needed to achieve the desired functionality goals? Therefore, a metric to quantify the amount of required periodicity is needed to help design new and effective maintenance strategies and evaluate existing ones. 3

The main task of maintenance is to periodically reset the manufacturing system either by repairing failures or by PM. This resetting should be defined in terms of specific parameters that are related to functionality such as production rate or available capacity, and/or physical parameters such as machine tool power efficiency. The notion of periodic resetting in the functional domain has been defined by Suh [15] as a mechanism to reduce complexity and to restore the desired state of operation. Suh [16] explained that introducing functional periodicity transforms the combinatorial complexity into periodic complexity which serves to ensure long term stability for engineered and natural systems. In this context, complexity is defined as a measure of the uncertainty in satisfying the system functional requirements and is measured by the information content. It is defined as follows [16]:

In order to develop a periodicity metric for evaluating the maintenance strategies, it is necessary to establish a model to define the maintenance strategies. To date, there is no systematic and mathematically consistent method for modeling the maintenance strategies. The known policies are currently described in a textual nonmathematical form. Therefore, there is a need to develop a mathematical methodology for modeling maintenance strategies. In this research, the main focus is placed on the timebased maintenance strategies because they are the most commonly used in industry. The developed Maintenance policies, reported in the literature, and the maintenance strategies applied in industry, indicate that there are two sub-strategies for any maintenance policy: The failure repair sub-strategy describes when to repair the failure and the level of repair.

m

I sys = −∑ log 2 Pi

MAINTENANCE MODELING

(1)

-

i =1

Where Isys Information content of the system m Number of functional requirements, FRs Pi Probability that a function requirement, FRi, is satisfied. The complexity is categorized by Suh [17] as shown in Figure (2):

The PM sub-strategy describes the number of PM classes and their levels.

Therefore, the maintenance strategy can be fully determined by defining the five criteria shown in Table 1. The first two criteria determine the failure repair substrategy and the last three determine the PM substrategy. Since the failure is normally repaired when it happens (assuming that a perfect failure detection system is in place), the first parameter is excluded from the proposed model. The repair / PM level is represented by a continuous real variable in [0 , 1] range where 0 means restoring the machine to its state just before failure and 1 means restoring it to the original new state [19]. Maintenance . Strategies Subcategories Failure Repair

Figure 2 Complexity Categorization The ‘periodicity’ causes the deteriorating functions to exhibit a cyclic behavior that restores their desired characteristics periodically. Therefore, periodicity reinitializes the system functionality to a “like new” state, which assures a high degree of functional certainty. Hence, introducing periodicity reduces, if not eliminates, uncertainty and consequently decreases the complexity associated with combinatorial complexity. Lee [18] introduced many examples of periodicity in systems from different fields including manufacturing systems. The deterioration of manufacturing systems performance is characterized by a time varying system range and may be

PM

Defining Criterion

Maintenance Policy Parameter

When to repair a failure? Repair Level

RL

Number of PM classes

N

Frequency of carrying out each PM class

PMF1,…,PMFN

Level of each PM class

PML1,.,.,PMLN

Table 1 Maintenance Strategy Parameters

114

4

PERIODICITY MODELING

complexity before resetting−complexity after resetting complexity before resetting (3) complexity after resetting =1− complexity before resetting

Resetting Extent=

The periodicity is a result of a resetting plan. Each resetting action re-initializes the system functionality, according to a certain pattern (plan). A system may mean a single machine or a whole manufacturing system but the current research considers the case of single machines, therefore, the resetting plan means the machine maintenance policy. The two words machine and system will be used interchangeably in the remainder of this article. First, the case of a single resetting plan is introduced and the formula for the resulting periodicity is developed. Then, the periodicity resulting from multiple resetting plans is investigated. 4.1 Periodicity of Single Resetting Plan

For simplicity, we will assume that the complexity increases linearly with rate υ, however, other patterns may be used. The complexity change in the presence of a resetting plan with time between resetting T and resetting extent χ is shown in Figure (5).

A resetting plan has two essential dimensions that completely define it: Frequency of resetting. -

Extent of resetting which expresses the level of re-initialization

These two aspects are illustrated in Figure (4) where different resetting levels are shown with a time between resetting, T.

When there is no periodicity (pr = 0, combinatorial complexity case), the complexity continues to increase without resetting as represented by the dashed line (line of zero periodicity). The other theoretical extreme is when the system is fully reset at infinitesimal time periods such that its complexity is always zero (pr =∞,) and is represented by the line of full periodicity. Therefore, as the periodicity increases, Area B increases and area A decreases. The periodicity is, therefore, expressed as:

Figure 4 Resetting Frequency and Level The resetting level is represented in the figure by the ratio

a L

where L is the total system complexity before the

resetting process, and a is the amount of complexity actually recovered by the resetting process. The resetting frequency represents the number of resettings per unit time which is expressed as 1/T. It assumes real values in the range [o, ∞] where 0 means no resetting at all and ∞ means system resetting at infinitesimal time intervals. The resetting extent quantifies the amount of resetting and it can be expressed by the following relationship:

Resetting Extent=

Figure 5 Complexity versus time with resetting policy (T, χ)

pr = lim t →∞

Area B

(4)

( Area A) × t

The area A can be expressed by the following relationship: ∞

υT 2

i =1

2

A = ∑ T χ ′Ci −1 + Where:

(5)

Ci complexity at time iT where i represents the number of resettings

χ′ =

(1- χ )

Where the complexity at any time iT is described by the following relationship:

amount of resetting amount of full re-initialization

(2) Where the amount of resetting for any machine functional parameter (such as production rate, availability, etc.) is defined as the difference between the value of the parameter before and after resetting. Furthermore, to make the resetting extent measure more generic and dimensionless, it is expressed in terms of the uncertainty of fulfilling the functional requirement, which represents complexity [16]. Therefore, assuming the complexity related to a defined functional requirement, such as availability [20] of a new system to be zero (i.e. designed system fulfills the specified functional requirement), then the resetting extent is expressed as:

Ci = χ ′Ci −1 + υT Where:

(6)

T

time between resetting Complication rate

υ

The complication rate is a new term introduced in this research to expresses the rate of increase of complexity. It is a property of each machine that depends on the rate of functionality deterioration. It is represented in Figure (5) by the slope of the complexity line. After mathematical manipulation, the following relation can be derived: pr =

1 1

1

− 1) 2T ( + 2 1− χ′

1

= T(

2 1− χ′

1

= − 1)

T(

2

χ

= − 1)

β 2

χ

(7)

−1

Where β = 1/T stands for resetting rate. The periodicity relationship is plotted in Figure (6) where each curve

115

represents the relation between resetting extent and the corresponding time between resetting at a certain periodicity level.

5.2 Preventive Maintenance Periodicity PM is the main source of periodicity when the maintenance strategy calls for minimal repair of failures. The machine is reset with each PM, therefore, the resetting rate is the same value as PM frequency (PMF) and the periodicity extent is expressed by the level of PM (PML). Therefore, the periodicity resulting from the PM of n classes can be described by the following relationship: n

prPM =



PMFi 2

i =1

(11)

−1

PMLi 5.3 Total Maintenance Policy Periodicity Figure 6 Resetting Extent versus Time between Resetting at Different Periodicity Levels

From Equations (10) and (11), the total system periodicity resulting from a given maintenance policy can be expressed as:

i =1

i =1

Ti (

1 2

χi

− 1)

(8)

RL

PERIODICITY-BASED

MAINTENANCE

POLICY

EVALUATION The periodicity due to maintenance programs, introduced by two independent sources; the Failure Repair and the Pre-planned PM, will be evaluated. 5.1 Failure Repair Periodicity The machine periodicity due to failure repair is calculated assuming the machine has a known failure rate λ. It is assumed that the machine is repaired (i.e. functionality reset) as soon as it fails, which is the common practice in industry. Therefore, the resetting rate is the same as the failure rate λ. Using Equation (3), it can be stated that: complexity after resetting = (1-Resetting Extent) ×complexity before reresetting

(9)

The resetting extent, in the maintenance context, is represented by the repair/maintenance level (more information about the different imperfect maintenance modeling approaches can be found in Pham and Wang [19]). Therefore, the failure repair periodicity can be expressed by the following equation:

prrep =

λ 2 RL

−1

PMLi

Therefore, given the maintenance policy parameters, RL, PMLi, and PMFi and the machine failure rate λ, the maintenance policy periodicity can be calculated. This calculated periodicity (pr) is a measure of the relative ability of the maintenance strategy to reset the machine functionality. 6

5

(12)

2

i =1

ILLUSTRATING EXAMPLE

In the following example, the maintenance policy used at an auto manufacturer assembly plant is used to illustrate the application of the proposed new approach. The plant maintenance policy can be described as follows: When a machine fails, it is instantaneously minimally repaired to quickly restore the production. The PM policy comprises four classes; between shifts: weekly, semi-annual, and annual. Each PM class has associated courses of action for each machine. The exact determination of maintenance level for each maintenance class requires historical data. Nevertheless, based on the courses of action in each maintenance class, the maintenance levels are estimated to be 0.05, 0.3, 0.6, and 0.95 respectively. The plant operates two shifts per day, five days per week. One shift is considered the time unit. Using the proposed maintenance modeling approach, the plant maintenance strategy can by fully described by the following parameters, given in Table (2): Repair Level

RL

0

Number of PM Classes

N

4

PM Frequency

PMFi

PM Level

PMLi

(10)

−1

0.05

0.3

0.6

Annually

i

−1

PMFi

0.002083

∑ pr =∑

2

n

+∑

0.004167 Semiannually

pr =

n

λ

Weekly

n

pr =

1.0 Between shifts 0.1

4.2 Multiple Reseting Plans Most real systems are reset using multiple resetting plans. These plans can be independent or dependent. In this research independent resetting plans are considered. The independency condition makes it possible to express the total periodicity as a summation of all the periodicity elements. Therefore,

0.95

Table 2 Original maintenance plan parameters This maintenance strategy applies to every resource/machine in the plant. One of these resources is a frame-welding robot which experiences random failures

116

Repair Level

RL

0

Number of PM Classes

N

4

PM Frequency

PMFi

0.5 Daily

0.025 Monthly

0.0083 Quarterly

0.002083 Annually

with an average of one failure/week. The amount of functional periodicity introduced by this maintenance strategy is calculated using Equation (12) to be 0.047. This periodicity measures the relative ability of the maintenance strategy to re-initialize the robot functionality. This measure is relative because it has no physical embodiment, but it is useful when used to compare different maintenance strategy alternatives. The company is considering a new alternative maintenance strategy described by the following parameters (Table 3):

PM Level

PMLi

0.1

0.3

0.5

0.95

[1]

[2]

[5]

[7]

[8]

[9]

[10]

[11]

DISCUSSION AND CONCLUSIONS

Maintenance in manufacturing systems introduces periodicity, which is required to keep the system functional stability throughout its life. A novel general metric for quantifying the periodicity has been presented and developed. A formula for calculating the periodicity introduced by a maintenance policy is derived. A new term called complication rate has been introduced to measure functional deterioration. The proposed periodicity metric can be used to quantitatively compare the resetting ability of different maintenance policies, which combined with other performance metrics like cost, availability and quality can vastly enhance decision making in selecting appropriate maintenance strategies. It has been shown that the calculation of periodicity introduced by a maintenance strategy, using the proposed model and formulation, is quite simple and makes it practically applicable in industry. Although the application of the developed periodicity metric has been discussed in the context of the engineering/manufacturing maintenance field; nevertheless, the method is general enough and can also be applied in any application that involves system resetting such as natural and political systems. 8

[4]

[6]

Table 3 Suggested maintenance plan parameters Using Equation (12), the periodicity of the new maintenance strategy is calculated to be 0.065. Since pr2> pr1, Therefore, the proposed new maintenance policy provides more periodicity of resetting the machine(s) functionality than the original policy. Hence, it is more capable of reducing the combinatorial complexity which leads to more stable system that is more certain to satisfy its functional requirements. It is important to notice that this conclusion is based only on the periodicity/complexity. But, in real life cases, there may be other criteria to be included in the decision. 7

[3]

[12]

[13] [14]

[15]

[16]

[17] [18]

REFERENCES Wiendahl, H. P., ElMaraghy, H. A., Nyhuis, P., Zah, M. F., Wiendahl, H. H., Duffie, N., and Brieke, M., 2007, Changeable Manufacturing - Classification, Design and Operation. CIRP Annals - Manufacturing Technology, 56(2): p. 783-809. Dekker, R., 1996, Applications of maintenance optimization models: a review and analysis.

117

[19]

[20]

Reliability Engineering & System Safety, 51(3): p. 229-240. Aurich, J. C., Siener, M., and Wagenknecht, C., 2006, Quality oriented productive maintenance within the life cycle of a manufacturing system. in 13th CIRP international conference on life cycle engineering. Dhillon, B. s., 1999, Engineering Maintainability: Gulf Professional Publishing. Wang, H., 2002.A survey of maintenance policies of deteriorating systems. European Journal of Operational Research, 139 (3): p. 469-489. Sheu, S.-H., Griffith, W. S., and Nakagawa, T., 1995. Extended optimal replacement model with random minimal repair costs. European Journal of Operational Research, 85(3): p. 636-649. Xiao-Gao, L., Makis, V., and Jardine, A. K. S., 1995. A replacement model with overhauls and repairs. Naval Research Logistics, 42(7): p. 1063-79. Nakagawa, T., 1986, Periodic and Sequential Preventive Maintenance Policies. Journal of Applied Probability, 23(2): p. 536-542. Nakagawa, T., 1988, Sequential imperfect preventive maintenance policies. IEEE Transactions on Reliability, 37(3): p. 295-298. Cassady, C. R., Bowden, R. O., Leemin, L., and Pohl, E. A., 2000, Combining preventive maintenance and statistical process control: a preliminary investigation. IIE Transactions, 32(6): p. 471-8. Moore, W. J. and Starr, A. G., 2006. An intelligent maintenance system for continuous cost-based prioritisation of maintenance activities. Computers in Industry, 57(6): p. 595-606. Wang, H. and Pham, H., 2006. Availability and maintenance of series systems subject to imperfect repair and correlated failure and repair. European Journal of Operational Research, 174(3): p. 170622. Demers, J.-M., 2005. Reliability centered maintenance. CIM Bulletin, 98(1086): p. 50. Ben-Daya, M., 1999. Integrated production maintenance and quality model for imperfect processes. IIE Transactions (Institute of Industrial Engineers), 31(6): p. 491-501. Suh, N. P., 2004. On functional periodicity as the basis for long-term stability of engineered and natural systems and its relationship to physical laws. Research in Engineering Design, 15(1): p. 72-5. Suh, N. P., 2005, Complexity in engineering. CIRP Annals - Manufacturing Technology, 54(2): p. 581598. Suh, N. P., 2005, Complexity, Theory and Applications, ed. MIT: Oxford University Press. Lee, T., Complexity Theory in Axiomatic Design, in Mechanical Engineering. 2003, Massachusetts institute of Technology: Massachusetts. p. 182. Pham, H. and Wang, H., 1996. Imperfect maintenance. European Journal of Operational Research, 94(3): p. 425-438. ElMaraghy, H. A., Kuzgunkaya, O., and Urbanic, R. J., 2005. Manufacturing systems configuration complexity. CIRP Annals Manufacturing Technology, 54(1): p. 445-450.

Development of an Extended Product Lifecycle Management through Service Oriented Architecture. J. Cassina, A. Cannata, M. Taisch Politecnico di Milano Piazza Leonardo da Vinci 32, 20133 Milano { jacopo.cassina; alessandro.cannata; marco.taisch } @polimi.it

Abstract The aim of this work is to define new business opportunities through the concept of Extended Product Lifecycle Management (ExtPLM), analysing its potential implementation within a Service Oriented Architecture. ExtPLM merges the concepts of Extended Product, Avatar and PLM. It aims at allowing a closer interaction between enterprises and their customers, who are integrated in all phases of the life cycle, creating new technical functionalities and services, improving both the practical (e.g. improving usage, improving safety, allowing predictive maintenance) and the emotional side (e.g. extreme customization) of the product.

Keywords: PLM, Extended Product, Service Oriented Architecture

1

INTRODUCTION

Nowadays, enterprises are ever more stressed and subjected to high market requests. Customers are more and more sophisticated in terms of products quality and related services. Thus, the “product” and its related management are becoming unavoidable key-aspect for the creation of a “product centric” or “product-driven” approach. This work aims at defining new business opportunities through the combination of the concepts of PLM, Product Extension and Avatar. Moreover it will be investigated the possibility of merging these ideas with the Service Oriented Architecture approach. The target is to allow a closer interaction between enterprises and their customers, who are integrated in all phases of the life cycle, creating new technical functionalities and services, improving both the practical (e.g. usage, safety, allowing predictive maintenance) and the emotional side (e.g. extreme customization) of the product. This paper deals with two completely different meanings of the same word: service. First of all, it will refer to the business meaning of functionality or features required by customers. On the one hand the word “service” will refer to a software component that can encapsulate business logic. In order to avoid misunderstandings we will refer to this meaning adopting the expression web service (or the acronym WS, and the known name of Service oriented architecture - SOA) instead we will use the term service by itself to refer to the other meaning. This paper will first of all explain the concept of Extended Product Lifecycle Management, that has been developed

CIRP IPS2 Conference 2009

118

starting from the concepts of PLM, Extended Product and Avatar. Then will be investigated the possibilities achieved by the adoption of Service oriented architecture (SOA) that uses loosely coupled web services (WS) to encapsulate and deploy business processes and products. For this area, the results of an European Project called Socrades will be analyzed, and the impacts of two main technologies such as embedded systems and wireless communication will be considered. The final aim is to propose a framework where products provided with enough intelligence can autonomously interact with other devices/humans within and outside the industrial plants through its overall lifecycle. Since each business process (production, logistics, maintenance, etc.) can be represented as a set of WS, such aspects as services or functionality can be naturally included and controlled in the framework. Finally, thanks to SOA paradigm, communication among different systems becomes naturally supported, being the base of the Extended PLM where products can take advantage of pervasive information more and more available in the coming Internet of Things.

2

PRODUCT LIFECYCLE MANAGEMENT:

The term Product Life-Cycle Management (PLM) defines the integration of different kind of activities, from the technical, organisational and managerial point of view, which are performed by engineering staff along the entire life-cycle of industrial products. This cycle covers the concept, the development and the design of the product, together with the manufacturing process planning, the

factory and supply-chain planning, till to the final disposal/recycling of the product itself. All these activities are strongly based and supported by engineering and production management information systems. Generally, PLM is an integrated approach for the management of product data along the product lifecycle (“from the cradle to the grave”). As such, it entails [1]: •

a strategic management perspective, wherein the product is the enterprise value creator,



the application of a collaborative approach to better use the enterprise competences distributed amongst diverse business actors,



the adoption of plenty of ICTs in order to practically establish a coordinated, integrated and access-safe product information management environment in the extended context. Then, PLM deals with the management of all the product data that are created, stored and managed along the lifecycle of a product, from its design to end of life. In such a way, it is possible to say that PLM is assuming a “holistic” role: according to Stark [2], “PLM brings together products, services, structures, activities, processes, people, skills, application systems, data, information, knowledge, techniques, practices, skills and standards”. Listening to the enterprise requests, several vendors, coming from different areas interested into the product and production management, are creating a PLM software market. Nevertheless, PLM is not primary an IT problem, but at first, is a strategic business orientation of the enterprise. From a strategic organisation point of view, the adoption of a “product centric” approach means a remodelling of all the relations established between the resources (people and equipment) involved into the relevant business processes oriented to a “product” lifecycle direction, with all that it concerns in terms of task allocations and measurement of the obtainable performances. In literature there are many different lifecycle models. The most common one presents three life cycle phases:

3

THE EXTENDED PRODUCT

The explosion of information & communication technologies has created a new kind of concept, defined as Extended Product, where the product is more then a simple artefact, but it is a complex result of tangible and intangible components. The extension is usually related to the functionality or a new business process around the product. According to Jansson [3] and Hirsch [4] tangible extended product can be intelligent, highly customized, and user-friendly; an intangible product is mostly the business process itself.

4

THE AVATAR CONCEPT

The Avatar is a virtual representation of the product, a digital shadow that contains all the information and knowledge of it. The Avatar is always connected with the physical item allowing all the stakeholders easier and better communications with it. The product-avatar that emerges is the sum of the physical product and an ICT shell of information, knowledge and intelligence. According to McFarlane [5] and Hribernik [6], the product avatar is characterized by: •

possessing a unique identity,



capability of communicating effectively with its environment,



can create, access transfer and operate upon information about himself,



deploys a language to display its features,



is capable of participating in or making decisions relevant to his destiny.



Beginning of Life (BoL): when the product is in the hands of the manufacturer.



Middle of Life (MoL): when the product is owned by the consumers.

5



End of Life (EoL): when the product has finished its useful life and has to be dismissed.

The Extended Product Lifecycle Management aims at creating new business opportunities through the combination of the previously explained concepts of PLM, Product Extension and Avatar [7] (Figure 1).

119

DEFINITION OF EXTPLM

Figure 1: ExtPLM Schema [7] It aims at allowing a closer interaction between enterprises and their customers, who are integrated in all phases of the life cycle, creating new technical functionalities and services, improving both the practical (e.g. improving usage, improving safety, allowing predictive maintenance) and the emotional side (e.g. extreme customization) of the product. In the previous schema it is possible to see how the Product, recognized using its id, is extended through product intelligence and is interfaced to the user, who interacts with it as an avatar. The ExtPLM aims at following this extended product through all its lifecycle phases, giving to the product enduser a set of services that will extend the usability and the utility of the product itself, improving the ownership experience. These services will also improve the market value of the product itself, and can be sold both to the customer (e.g. predictive maintenance) and to other companies (e.g. specific advertising).

120

Examples of applications could be a deep customization, predictive maintenance, customized manuals and FAQs (frequently asked questions), self adaptability to the user etc. To define all the possibilities of ExtPLM, a Delphi study has been carried out with the efforts of experts from all over the world [8], analyzing all the answers and then the whole, a list of ExtPLM possibilities, within the different lifecycle phases, emerged. These are explained in the following. During the BOL (Beginning Of Life), which is composed by the design, the manufacturing and the delivery phases, the customer will be able to define and redesign the product according to his needs and willing, changing the configuration as many times he wants, shaping the initial form of the avatar within his hands. During this first phase since the physical part still doesn’t exist, the customer directly interacts with the virtual part of the avatar. For example, through a web site, he is able to modify the parameters of the product he is aiming for, verify the result on a virtual model and being informed about price

changes due to his decisions. Moreover the avatar will also give him specific suggestions on possible improvements and particular discounts. This will allow the buyer to tailor the product on his very own needs, opening the possibility of a mass customization. Furthermore, the information on the behaviour of the customer within a configurator and the data from the products already sold (e.g. failures reports, most common upgrades, etc.) will be used from the designers and the engineers to improve the design making better and more desirable products. Then, if the customer will order the production of the avatar he shaped, he will be allowed to follow its idea during the production, being able to modify his decisions about the specifications if he wants, being informed if there have been changes in the process or in the raw materials available. During this phase the buyer will also be able to change some product characteristics like for example the colour of the product, e.g. if a new one is proposed by the producer. He will be able to modify actively the price, selecting different quality, or modifying the delivery date, or using recycled or remanufactured parts. During this phase the customer will also be able to follow and to see how his product is “growing”, being assembled. From the manufacturer point of view, the avatar will be ready to auto-negotiation processes within the shop floor, to optimize the usage of the resources, the machines and the manpower. The extended product will also be able to interact with the suppliers and the ERP, adding items to the list of products to purchase improving the management within the company itself. When the product has reached its final physical form, the buyer will also be able to modify the shipping while it is in process, modifying both the place and the time of the deliver During the MOL (Middle Of Life) phase, the extended product, using the sensors on the physical side and the intelligence of the ICT side, will collect data during the usage. These will allow self diagnose, that will lead - for example - to predictive maintenance; moreover it will be able to understand the needs and the habits of its user, suggesting way of using it better, or being able to answer questions of the customer. All these will lead to a better interaction with the product. The end user, who is the owner of the data, will be able to share the collected data

with the producer, being paid or with other benefits for that, allowing this way a continuous re-design process that will lead to continuous improvements. Furthermore the user will be able to access, suggested by the product itself, manuals, FAQs, wikis and forums where he could share his experiences with other users, creating a virtual community. Finally the data will allow the extended product to be able to propose to his user services, upgrades or information and news that it considers interesting for him through an analysis of the behaviour. The stored data will also improve the EOL (End Of Life), since it will be possible for the avatar to esteem the aging of its own components, suggesting these to reuse, remanufacture and recycle.

6

AN ARCHITECTURE FOR EXTPLM

As emerged in the previous sections, flexible information, knowledge management, communication and treatment are the basis to achieve an Extended Product Lifecycle Management. The ExtPLM concept requires that continuous changes and modifications could occur both from customer and company sides, highly flexible and reconfigurable information and manufacturing systems are needed to ace the new level of variability. Within the PROMISE project, that dealt with Closed Loop Lifecycle Management [9], an architecture for managing products during their whole lifecyle has been proposed. This architecture, represented in figure 2, allows great flexibility and connectivity and can represent the basis for the ExtPLM. Within the same project, it has been also developed a Product Data Knowledge Management Sistem Object Model (PDKM SOM), that has been then implemented using different tools (MySQL, MySAP-PLM) and an ontology [10].

121

Figure 2: PROMISE Connectivity [10] An improvement of this kind of architecture, using the same kind of data structure, can be achieved if it will be developed using the Service-oriented Architecture (SOA) as conceptual model for software organization.

7

SERVICE ORIENTED ARCHITECTURE

There are multiple definitions of SOA both in technological and business perspectives [11-12-13]. For the purpose of this document, the following will be used: “A serviceoriented architecture (SOA) is a set of architectural tenets for building autonomous yet interoperable systems” [14]. SOA can be considered more an architecture philosophy than a technology or a standard since it represents a set of good principles for designing, open and interoperable software. The proposed definition includes two keywords: autonomous and interoperable. The principal characteristics setting apart autonomous systems are that: they are created independently of each other they provide self-contained functionality, i.e., their functionality would be useful even if it was not associated with any higher level systems.

Interoperability is favoured by:

122

clearly abstracting the interface that a service exposes to its environment, from the implementation of that service; making this interface visible to others, together with policies and constraints for its use.

To achieve the ExtPLM vision through SOA, both high enterprise level and device level services have to be analyzed, defined and implemented. At enterprise level SOA concepts have been studied in the last years [15-16], fewer works and literature can be found on the adoption of SOA at low level (e.g. for supporting interaction between products and industrial devices/machines or else). The interaction between product and “production systems” or customer in a seamless way through WS-orientation is not yet deeply studied and developed; only some paper can be found on the adoption of web ontology for supporting design and engineering activities, not for supporting PLM nor extended product or manufacturing activities [17]. In order to let the product become an agent that can autonomously interact with such heterogeneous entities, there is a strong need for a common language such as WS that need to be implemented on both sides: manufacturing devices and products (i.e.: for shop floor control applications). By achieving combined autonomy and interoperability, SOA enables architectural approaches that can make easy to implement decision support systems services for all the lifecycle phases. The following features are the most prominent: Integration capability: services can be readily integrated with other services, either statically or

-

-

-

-

dynamically. Furthermore, services can be readily composed into higher level services. Owing to the abstraction between service interface and service implementation, services can be materialized on heterogeneous software and hardware platforms. Agility, flexibility and adaptability to change are greatly increased as services can be easily reconfigured or replaced, service deployment can be conducted incrementally and scaling can take place over time. Communicating entities can share and exchange resources and collaborate with each other through direct, peer to peer communication, i.e. without depending on the assistance and control of some higher level entity. Decision making can thus be driven down to the source of the information acted upon. This in turn enhances responsiveness and efficiency, while improving configurability. A decentralized mode of operation further adds resilience against failures by eliminating single point of failure hazards. Development cost is reduced as re use of services is facilitated and application programming is done at the highest possible level of abstraction.

Figure 3: Service Oriented Architecture Schema One relevant European project that is developing the implementation of SOA paradigm down to the device level is SOCRADES [18]. Its technical approach is to create a service-oriented ecosystem: networked systems are composed by smart embedded devices interacting with both physical and organizational environment, pursuing well-defined system goals. Taking the granularity of intelligence to the device level allows intelligent system behaviour to be obtained by composing configurations of devices that introduce incremental fractions of the required intelligence. This approach favours adaptability and rapid reconfigurability, as re-programming of large monolithic systems is replaced by reconfiguring loosely coupled embedded units. The use of device-level Service Oriented Architecture, contribute to the creation of an open, flexible and agile environment, by extending the scope of the collaborative architecture approach through the application of a unique communications infrastructure, down from the lowest levels of the device hierarchy up into the manufacturing enterprise's higher-level business process management systems. Moreover SOCRADES is developing and studying the adoption of embedded systems and wireless technology in order to provide industrial devices with the technical autonomy and independence needed for the implementation of such interoperable and reconfigurable

systems. The combined adoption of SOA paradigm, wireless technologies and computing capabilities (through embedded systems) enable the creation of an ecosystems of smart and autonomous devices that can independently cooperate and communicate. By adopting SOA paradigm, both industrial devices and products could be encapsulated into WS or described as a composition of WS. Products can be represented through semantic web detailing all its components and additional (more intangible) features. SOA allows two main important opportunities: Seamless interaction with “things” (i.e: enterprise systems, customers, other devices through the overall lifecycle, other product, etc.) Seamless inclusion of even more intangible aspects such as services or functionalities.

8

SERVICE ORIENTED ARCHITECTURE FOR EXTENDED PRODUCT LIFECYCLE MANAGEMENT

The adoption of Service-oriented Architecture, through the various available tools (DPWS, WSDL, etc.) is fundamental in order to describe and include product’s features and additional services as described in the concept of Extended Product enabling interoperability with heterogeneous “things” (devices, other products, etc.). Hence, thanks to the extreme flexibility of SOA paradigm, the previously explained concept of Avatar, that is the digital shadow of the product, should be seamlessly implemented through the adoption of WS technologies. Given to the products the necessary computing and communication capabilities provided through the adoption of embedded systems and wireless technologies, thanks to SOA paradigm as conceptual way of creating systems, products can effectively become autonomous agents with the capability of interacting with other devices at low level but also with higher enterprise systems or also with customers. Moreover, it could be possible to effectively represent and identify the services and the functionalities on the product itself, in order to let it become an Extended Product with a clear embedded representation of the Avatar through WS. Analyzed the current status of the proposed reference architecture for ExtPLM proposed within the PROMISE project and the under development architecture that is under development within the Socrades project, it is possible to see a desirable convergence toward a Service Oriented Architecture to be used for Extended Product Lifecycle Management. Such a kind architecture could implement the ideas of ExtPLM in a modular way. The structure for communication and product traceability could be set up, using the PROMISE results; in fact the developed middleware could be used a “traceability” service, while the PDKM SOM could serve as the data structure, the ontology to structure the data exchange within the system. This approach will have the advantage that then it will be possible to implement each specific service toward the customer and the product stakeholders as a web service, such a way it will be easy to add, manage and restructure services, since each one of them will have just to define its own required data within the given structure and its own way of working. Also the update of the system will be

123

easy since each WS will be seen by the system as a black box. Such a system, creating an easy interoperability of the product, could also allow it to interact with a multi layer decision making system, where decisions that have to be taken at lower levels (e.g. when to manufacture a product) can be redirected ad higher level where there is complete awareness of the whole system. This kind of issue could be handled thanks to SOCRADES general architecture since a unique communications infrastructure, down from the lowest levels of the device hierarchy up into the manufacturing enterprise's higher-level business process management systems, will be available.

9

CONCLUSION

[4]

[5]

[6]

[7]

This paper envisions that the adoption of SOA can support both from a conceptual and practical point of view all the aspects of Extended Product Lifecycle Management. Hence, the trend of products more and more provided with services (described by the concepts of Extended Product) can be effectively satisfied providing also all the advantages of a PLM point of view The ExtPLM concept requires a seamless integration that is possible to achieve through the SOA paradigm. The product will also require that communication will be enabled among multiple heterogeneous entities through a common language. The basis for this kind of ExtPLM architecture has been already developed within the PROMISE project that deployed a Product Data Knowledge Management ontology and an effective middleware. The Socrades project developed a Service Oriented Architecture that enables device level communication, shop floor management and multi layer decision making systems. Starting from these results, an architecture for ExtPLM could be developed such a way to be flexible, powerful and interoperable. Moreover, thanks to the WS structure and the WS metaexchange it will be possible to add easily new services to the architecture, creating and deploying continuously new services for the customers.

[8]

[9]

[10]

[11] [12]

[13]

[14]

[15]

Acknowledgments

[16]

This work has been developed merging two European Project; PROMISE PROduct lifecycle Management and Information tracking using Smart Embedded systems (No. IST-2004-507100) and SOCRADES …. . The authors thank all partners, and the European Commission for their support.

[17]

[18]

References [1]

[2]

[3]

Garetti M. Terzi S., 2004, Product Lifecycle Management: definition, trends and open issues, Proceedings at III International Conference On Advances In Production Engineering, 17 - 19 June 2004, Warsaw, Poland Stark J., (2004), Product Lifecycle Management: Paradigm for 21st century Product Realisation, Springer, ISBN: 1852338105 Jansson K., Thoben K.-D.; The extended products paradigm. An introduction. DIISM2002 - The 5th

124

International Conference on Design of Information Infrastructure Systems for Manufacturing 2002, Osaka, Japan. Hirsch B. E., Thoben K.D. and Eschenbaecher J., 2001, Using e-business to provide Extended Products, Automation, Automation days, Helsinki, Finland. McFarlane D., J. Sarma, G. Chirn, J. Wong, A. Ashton, 2003, Auto-ID systems and intelligent manufacturing control, Journal of Engineering Applications of Artificial Intelligence, 16 : 365 – 376 Hribernik K, Rabe L., Schumacher J., Thoben K.D., A Concept for Product-Instance-Centric Information Management, 11th International Conference on Concurrent Enterprising, Munich 2005, pg. 427-434. J. Cassina, Extended Product Lifecycle Management, PhD thesis at Politecnico di Milano, April 2008 J. Cassina, M. Taisch, A Vision on Extended Product Lifecycle Management, APMS – Innovation in networks, 14-17 September 2008, Espoo, Finland. Jun, H., Kiritsis, D., and Xirouchakis, P. 2007. Research issues on closed-loop PLM. Comput. Ind. 58, 8-9 (Dec. 2007), 855-868. J. Cassina, M. Taisch, D. Potter, A.K. Parlikad, Development of PROMISE Architecture and PDKM Semantic Object Model, ICE 2008, 23-25 June, Lisboa, Portugal. E. Newcomer and G. Lomow - Understanding SOA with Web Services - Addison Wesley - 2005 M. Bell - Service-Oriented Modeling (SOA): Service Analysis, Design, and Architecture - Wiley & Sons 2008 T. Erl - Service-oriented Architecture: Concepts, Technology, and Design - Upper Saddle River: Prentice Hall PTR - 2005 F.Jammes and H.Smit Service-Oriented Paradigms in Industrial Automation - IEEE Transactions on Industrial Informatics, VOL. 1, NO. 1, February 2005 F.B. Vernadat - Interoperable enterprise systems: Principles, concepts, and methods - Annual Reviews in Control, 31 (2007) 137–145 W.A. Estrem - An evaluation framework for deploying Web Services in the next generation manufacturing enterprise - Robotics and Computer Integrated Manufacturing 19 (2003) 509–519 W.Y. Zhang, J.W. Yin - Exploring Semantic Web technologies for ontology-based modeling in collaborative engineering design - Int J Adv Manuf Technol (2008) 36:833–843 SOCRADES website http://www.socrades.eu

Multimodal User Support in IPS² Business Model

R. Gegusch1, C. Geisert2, B. Hoege3, C. Stelzer2, M. Roetting3, G. Seliger1, E. Uhlmann2 Department of Machine Tools and Factory Management, Chair of Assembly Technology and Factory Management, Technische Universität Berlin, Pascalstr. 8-9, D-10587 Berlin, Germany 2 Department of Machine Tools and Factory Management, Chair for Manufacturing Technology, Technische Universität Berlin, Pascalstr. 8-9, D-10587 Berlin, Germany 3 Department of Psychology and Ergonomics, Chair of Human-Machine Systems, Technische Universität Berlin, Franklinstr. 28-29, D-10587 Berlin, Germany [email protected]; [email protected]; [email protected] 1

Abstract The use models of Industrial Product-Service Systems are based on the idea of offering functionality, availability or results. This paper proposes a concept for multimodal user support in interaction with condition monitoring and knowledge generation, whilst taking knowledge protection into consideration. A shared-vision system connects a less-qualified person with an expert for solving problems collaboratively. The user is instructed via multimodal user interfaces, which require data related to the service design, construction model as well as the current machine condition. Such data is acquired by a process accompanying information system, which deliver information relevant to the required service from sensors. Keywords: Industrial Product-Service Systems, Shared Vision, Virtual Life Cycle Unit, Knowledge Generation

1 INTRODUCTION To keep competitive in times of global economy, manufacturers of capital goods (e.g. machine tools) have to expand their business activities beyond selling physical products. Industrial customers make their day-to-day business by selling this kind of products (e.g. manufacture components) which means that they are interested in functionality over a time frame to achieve a result. Therefore, manufactures of capital goods should offer functionality, availability or results instead of selling products to meet the customer needs. Selling functionality means that the customer gets the physical products designed to its requirements. The supplier is responsible for maintenance and the risk of breakdowns. In the case of selling availability or results, the risk for breakdowns and the execution of maintenance processes goes over to the provider. To guarantee the functionality over a time frame or to achieve the result the provider has to have full control about when, where, what for, why, by whom and how the machine is being used and maintained. Unwanted or unqualified modifications or changes on the product might risk the loss of the granted availability or result. The manufacturer has to provide qualified service technicians that have to be able to support the customer on the location where the machine tool is set up. From the economical point of view the required on-site activities for maintenance and support should be carried out with minimal personnel effort of the supplier. Since teleservice, as an often stated support mechanism [1], is not suitable to solve all technical problems, one possibility for on-site support is to outsource technical service to local service providers that act as subcontractors. Another way to keep the manufacturer’s effort small is to enable the owner of the machine tool to maintain it by its own staff. Nowadays, in both cases the manufacturer has to publish required

CIRP IPS2 Conference 2009

125

information material (e.g. construction drawings, detailed service manuals). Therefore, a transfer of sensitive data and know-how is often unavoidable while the machine tool manufacturer risks loosing service contracts. The loss of know-how is a tremendous risk for first world companies producing or selling on the global market, especially regarding low cost countries [2]. So, the challenge is to enable the local machine operator regardless of the technical qualification of its staff to perform maintenance and inspection tasks for complex machines. This has to be done at the time and place where it is required and without transferring detailed knowledge. Classical use models based on selling physical products and offering product accompanying services are not able to face this challenge due to the independent design of product and service. The idea of offering functionality, availability or result instead of selling machines and additional services leads to the necessity of an integrated and mutually determined planning, provision and use of product and service shares including its immanent software components, so called Industrial Product-Service Systems (IPS²) [3]. An Industrial Product-Service Systems (IPS²) is built by a combination of a product and a service, so solutions which are offered to customers, content shares of product and service. Depending on this solution it is possible to characterize three different IPS² use models: function-, availability- or result-oriented [4]. The product is described as a tangible, which can be delivered by the IPS² provider to the customer and is the profit making component of an IPS². The service is intangible, which provide a value to the customer [5]. The IPS² provider is the only contract and contact partner of the customer. He is responsible for all activities, needed to use the IPS² as agreed in the contract.

In this paper a concept for multimodal user support in an availability-oriented IPS² will be described and demonstrated within a micro production scenario. 2 PREREQUISITES TO GUARANTEE AVAILABILITY The considered physical product has to feature special characteristics to enable the above mentioned use model of guaranteed availability without the provision of on-site service technicians. Integrated Condition Monitoring The provider of an availability-oriented IPS² needs to know when the machine is going to fail, for example due to wear of relevant components. Therefore, the conditions of all function relevant components have to be monitored by a Condition Monitoring System (CMS). This CMS uses sensor and control integrated signals to generate condition related characteristic values [6]. It detects trends of these values and, if possible estimates the remaining life-time of the respective component. Integrated Load Monitoring Machine tools are designed for a predefined set of loads that depends on the customer specific type of application, for example material to be machined and process parameters. The resulting load profile gives information about the way the machine tool is used [7]. The IPS² provider can combine this profile with the monitored condition of the machine to derive a cause-and-effect chain and to check up on the correct treatment of the machine. For example a shock sensor integrated into a lathe detects mechanical shocks on the spindle that may cause damages of the bearings. The recordings of these events indicate a reason for reduced life-time of the spindle. Tamper-resistant Design Since the IPS² provider is responsible for the availability of his production system he has to guard against manipulation. This means that all damageable parts are difficult to access by unauthorized persons. For parts where this is not possible due to functional restrictions, a monitoring system has to detect and log interventions. The design of the machine has also to ensure that no technology relevant know-how can be obtained during service operations. Communication Interface To inform the IPS² provider about machine conditions and maintenance activities needed, it has to be equipped with a communication interface. This interface is also used by the IPS² provider to authorize and support the local machine operator to execute service processes and for teleservice purposes [8]. Modern User Support A possibility to facilitate users to do services beyond their qualification without additional personnel costs is the use of modern technical human-machine interface systems (e.g. head mounted displays) in combination with product and process accompanying information systems with knowledge generation. Intelligent display methods can prevent unintentional knowledge transfer. In the proposed use model, the technical service shall be performed by the machine user. To avoid the loss of know-how, required technical information is only available during a service operation. Therefore, adapted service instructions have to be given dependent on the situation and the user on-site. The instructions should enable the machine user to perform the service operation without the need of technological know-how. To reach this goal an integrated design of relevant machine components and service processes is required.

Service Process Control and Documentation Service and product shares influence their specifications complementary and can be partially substituted against each other in the concept of IPS². Service processes have to be designed as modules in an availability-oriented IPS² to ensure planning reliability and process reliability, as well. To control an arbitrary process it has to be almost deterministic. Therefore, measurements that indicate start and end of each process step have to be defined. Further measurable quantities are needed to evaluate the service quality. The documentation of the service process provides evidence of executed service and can be used for quality management and knowledge generation to optimize the process or its model. The following sub-chapters give a more detailed overview of the above mentioned features and show exemplary concrete solutions. 2.1 Multimodal User Support The integration of multimodal user interfaces is an adequate solution to reduce the complexity of humanmachine interfaces and to reduce errors while interacting [9]. Automated, knowledge based, and adapted usermachine dialogues can provide a reliable support for qualified and less-qualified users. As described above, availability-oriented use models require short downtimes of the machine and low costs of service. To reduce service costs and to realize immediate and personal support for the machine user, one established model of service is teleservice. It was mentioned first by Kearney & Trecker, a machine tool manufacturer, for the description of telecommunication regarding to customer service, e.g. start-up, maintenance, and repairing [10]. In particular, Massberg et al. point out that teleservice leads to a close relationship of customer and provider where the human being plays a decisive role [11]. The Projects ARVIKA [12] and ARTESAS [13] introduce head worn user support devices and demonstrate that technology for industrial e-services [14] already exists. In case of support by teleservice, e.g. in repairing, two people (the technician and the remote-service expert) collaborate to solve a problem, which is called a remote collaborative physical task [15], while the process of collaboration can be called as distributed problem solving. A system which supports two persons in distributed problem solving was described by Velichkovsky, who tested the effect of gaze transfer in his experiments [16]. Gaze-position transfer in this experiment of solving a puzzle task improved the performance and changed the communication process positively. To support a machine user in a remote collaborative physical task, a mobile and stationary device is needed. This system is called a shared-vision system and enables two persons to solve a problem from separated places. The following support process is already developed and in the phase of testing by Hoege and Roetting: A user wears a head mounted system (cf. figure 1) at the place of the machine. S/he is connected by internet with a remote specialist who is sitting in front of a PC monitor. The scene camera of the user transmits the view on the machine to the remote specialist as well as the gaze coordinates which are measured by a head mounted eye tracker. The remote specialist can see the view and the gaze-position of the user on a monitor in front of her/him. For supporting the user, the specialist’s gaze coordinates are recorded by a remote eye tracker. His/her gaze data is transmitted to the user’s head mounted optical seetrough display. In this display the user can see the gazeposition of the remote specialist. Additionally, both are connected by an audio interface (headset) for verbal

126

communication. Normally, a person is looking at the object his or her attention is focused on. Therefore, the transmission of the gaze positions of the two participants will support and ease the communication between them.

scene camera

head mounted display

economical, environmental and social product and process attributes and parameters [20] e.g. location, utilization, efficiency, emissions, condition, malfunctions and failures.

head band

head mounted eye tracker

Figure 1: head worn components of a shared-vision system. One component of the shared-vision system is a head mounted display which is used to place information directly into the field of view of a user. In case of a seethrough display it is possible to augment the real world with additional information. The visualization of a person’s gaze into the view of another person is an application of augmented reality (AR). Augmented reality is part of the reality-virtuality continuum defined by Milgram and Kishino in 1994 [17]. One side represents the real world as it exists and the other side represents a virtual world computed and realized by adequate technology. Additionally, AR can be used to visualize, e.g. instructions, data by integrated sensors, or exploded drawings into the view of a worker, so s/he does not have to switch his focus of attention between a printed instruction manual or a monitor displaying the sensor data separated of the machine. Speaker-independent speech recognition and gesture recognition for interaction with the user interfaces in the HMD or the PC-based interface of the micro production machine can be easily integrated for a multimodal and hands-free interaction concept. For optimal user support, a seamless integration of collected data by sensors and for an automated knowledge generation, an implementation of a technically intelligent [18] device into the multimodal interaction concept is necessary. 2.2 Virtual Life Cycle Unit Effective and efficient adaptation can help to reduce resource consumption by e.g. extending the product’s life span and by supporting availability-oriented use models. IPS² providers are confronted with increasing demands for product and process availability, reliability and safety. Therefore, the assessment, prediction, diagnosis, monitoring, and control of product and process behavior are desirable. Adaptation is facilitated by IPS² accompanying information systems which are capable of acquiring, processing, and communicating relevant IPS² product and process data and information, whereby information stands for linked data. These data and information deliver potential sources to obtain experience, like inferences about conditions, wear or quality aspects and deliver knowledge about the behavior and usage of their products [19]. The Virtual Life Cycle Unit (VLCU) is a concept for an IPS² accompanying information system for knowledge generation and control (cf. figure 2). A VLCU acquires via sensors or ITdocuments data and information from IPS² operations and components, wherein the term operation includes products and services. This may include technical,

127

Figure 2: Modular concept of a Virtual Life Cycle Unit. The acquired data and information will be processed with the objective to generate knowledge, e.g. inferences between product or process parameters and failures. Acquired and processed data, information and knowledge are being communicated for further evaluation. With the help of condition prognoses the need of maintenance can be identified and maintenance activities can be automatically deployed. These data, information and knowledge can also be used to assist processes of the value creation chain such as development, redesign, production, recycling and disposal, and supporting processes such as control processes in a service procedure. In the IPS² operation phase the VLCU is able to acquire information about the usage behavior, resource needs, services and workers. This sets a base for the search of inferences, e.g. worker qualification, service efficiency or flowcharts regarding the user demands and requirements. This knowledge is invaluable for cost and resource reduction in service planning. The goal is to increase the use- and time-productivity of resources in IPS² by finding inferences about adaptations between different usage phases and to improve the IPS² work sequences. Decision trees are used to determine the best course of action, in situations having several possible alternatives with uncertain outcomes. The resulting chart or diagram displays the structure of a particular decision, the interrelationships and interplay between different alternatives, decisions, and possible outcomes. [21] A decision tree can be generated out of recorded process documentation data lists. There is a variety of algorithms for building decision trees that share the desirable quality of interpretability. A well known and over the years frequently used is C4.5 [22]. C4.5 builds decision trees from a set of training data, using the concept of information entropy. Information entropy is a measure of the uncertainty associated with a random variable. It quantifies, in the sense of an expected value, the information contained in a message. The decision trees generated by C4.5 can be used for classification, and for this reason, C4.5 is often referred to as a statistical classifier. Decision trees can describe the workflow of complex processes with possible variations of process steps due to external conditions. These might be corroded parts, wrong assembled components or variations in the geometry of the assembled screws. The challenge of acquiring and processing data from complex machines can be solved by concentrating on standard components. Standard components like bearings, gears, compressors, pumps, dampers, filters, hose lines or pneumatic components are integrated into

various more complex products, e.g. assembly systems, ground conveyors or industrial robots. By focusing on the assessment of data on standard components of machines, the development effort for the assessment of complex products is distributed technically and economically on many applications [20]. Also, less overall expertise is needed in order to develop a VLCU system for complex products or components, because the subsystems can be examined almost independently of each others. The solution space is significantly reduced for each VLCU designer, figure 3.

these locks. The login and access procedure can be realized by an internet connection, so that the provider is able to control who is maintaining the machine. In the DFG funded research project “Disassembly Factories”, magnetic connection modules have been developed, which open by an electronic signal, see figure 4 [23]. By measuring the magnetic force and recognizing the disconnection a VLCU is able to detect and control the access to machine components. Thus, the VLCU can support a sequential and simultaneous support in accessing machine modules during a maintenance process.

Figure 4: Arrangements of magnet and connecting element [23]. Figure 3: Products, standard components, interest groups, and business areas. Maintenance processes on those standard components are a ‘door’ for data acquiring of a service. This enables a direct connection to the standards component, failure prognoses and to detect critical system parameters, inferences between services on a component and its condition, e.g. remaining life time or failure rate. With the flexibility of an IPS² concept the influences on a customer solution is typical divided in two specific parts: the product and the service specification. 2.3 Availability-Oriented IPS² Product Specification Each IPS² use model requires special design characteristics of machine components due to the focus of the use model. For the availability-oriented IPS² the need for easy, safe and secure user interaction for service execution is given. These tasks can be performed by constructive features, e.g. access restrictions via locked access gates. Modular construction: The IPS² production system is subdivided in functional sections. Each section is realized as a module with mechanic and/or electronic interfaces. Each module represents a closed system and can be disconnected separately without disassembly of the module itself. Therefore, a modular design enables an easy interchange of units. In addition, it protects technological know-how because the user does not need to get detailed information about the overall functionality of the production system. Black-Box-components: If a module cannot be disassembled by unauthorized persons without destroying it, it is called a black box. Such a component is described only by its input and output characteristics. Its internal configuration and technology is not transparent. Therefore, it is a suitable measure to avoid knowledge transfer of key technologies used within a product system. Electromechanical locks: Electromechanical locks can be installed instead of conventional connecting elements like simple screws to ensure that maintenance processes can only be executed by authorized personnel. The user has to be authorized by the IPS² provider and logged into the system to activate

2.4 Availability-Oriented IPS² Service Specification Not only the product specification differs from conventional approaches regarding the requirements of the chosen industrial product-service system but also the design of service processes. To enable a person to execute service processes like during maintenance of machines regardless of its technical qualification, the service specification has to be adapted. While service technicians are well schooled in the branch specific technical terminology and have operating experience, machine users often do not need that kind of qualification. Therefore, service process design has to take into account that additional support is needed [24]. This can be realized by adapted instructions and a user-friendly interface [25]. 3

APPLICATION SCENARIO IN THE FIELD OF MICRO MANUFACTORING

3.1 Dynamical Adaption Assistance of Service Operations The above mentioned shared-vision device connects a less-qualified person with an expert system for solving problems collaboratively. The expert system’s functions range from automatically generated instructions to individual support by expert staff, who can give operating instructions individually and verbally. The usage of the proposed shared-vision system enables less-qualified users to execute any designed service process at a predefined quality level. The expert system provides the process models, data regarding the current machine condition and the required technical information to enable the service to be carried out. The idea of knowledge protection leads to a capsular design of functionally relevant machine components, which can be accessed only by authorized personnel logged into the system. With process relevant data out of the spindle’s usage phase the decision tree for the workflow of the maintenance process can be updated. 3.2 Design of a Micro Manufacturing Spindle for an Availability Oriented IPS² For a description of an assistance of service operation a scenario focuses on the milling spindle as the most stressed core component of the machine tool. Since the spindle is the most stressed component of the milling

128

machine, it has to be maintained due to progressing wear. The wear behavior depends mainly on the conditions of spindle operation, e.g. work piece material, and technological parameters. Therefore a periodical maintenance is not sufficient to guarantee the availability of the IPS². The IPS² use model determines the maintenance service activities to be done by the customer’s personnel regardless of its technical qualification (ref. chapter 1). The milling spindle of this scenario is constructed of four functional modules with integrated sensors for condition monitoring (cf. figure 5). A spindle with the whole construction information from SycoTec GmbH & Co. KG is used to show the relevant processes and the needed interaction between product and service. Media Module (1) In this module all supplies media are brought together. The circuit points, e. g. the electricity, compressed air, to run the spindle are combined in this module.

mechanism, sealing elements, an outside lying motor coil and the ball bearings. The balls of the bearing are manufactured of ceramic. Furthermore, the motor module includes the tool chuck. All modules are displayed in a CAD-model with single pieces, figure 6. A detailed description of all movements to exchange the individual parts is basically needed for the individual adaptation of a service operation. Aspects of user interaction corresponding with the construction of the spindle: The separation of each module can be realized due to the modular design of the spindle. Modularization reduces the amount of needed information for maintaining a single module. Furthermore, it protects knowledge about the function and the relation between all modules. Spring Pistons

Combination of cylinders Electronic locks components

Figure 5: Modules of the spindle (CAD-file) © SycoTec. The closure head is equipped with different connectors. The diameters of the circuit points are different, so that a connection mistake is eliminated. The plug diameter for the air sealing is smaller than the one for compressed air for the tool change (figure 5). A greater diameter means a higher pressure range; so that the user of the spindle can intuitional connect the spindle with the needed media. This module is sealed with a seal disk against the cylinder module (2) and is fixed with two socket screws (M4x25) on the cylinder module. Cylinder Module (2) This module includes a system of several pistons in series to press the spindle for the exchange of the milling tool. The inner functional cylinder is housed inside another cylinder (cf. figure 6). The functional cylinder is designed as a combination of the single elements of these pistons. Three springs are compressed when the pistons in the functional cylinder are pressurized. In the non charged system the springs close the tool chuck. The cylinder module is combined of parts with black-box character (e. g. the described cylinder). A sealing ring seals the cylinder module against the connector module (3). Both modules are screwed together by the bottom of the cylinder module and by the head of the connector module with three socket screws (M3x20). Connector Module (3) To merge the cylinder unit with the motor shaft a connector is needed. This connector compensates the deviation of form and position of the parts. Furthermore the axial forces to move the chuck are transmitted over this module. Motor Module (4) This module includes the motor for the rotating motion. The module is built by a central shaft for the clamping

129

Figure 6: Module (1) and (2) © SycoTec with additional electronic locks [22] (CAD-file). As described in the cylinder module (2) the pistons with the sealing elements can be integrated in a pressure chamber, so that this component is capsulated against the other components. The use of detailed design drawings is required for a special instruction to exchange for example the sealing elements of this cylinder. The media module (1) can be separated by loosening the two standard screws (figure 5). The user for a service needs a socket wrench to execute this process step. In a second design the whole rotating spindle can be secured with special mechanics combined with electronic parts against unauthorized contact (figure 6). As a lock the modules can be separated only by a login of an individual, authorized user. The media module (1) can pick up the locks to connect it with the cylinder module (2). For the exchange of for example a broken spring the locks have to be opened. The login to the specific locks can take place by a user specific transponder, which has to be placed over the media module (1). By this movement relevant user information can be logged for service documentation. 3.3 Description of a Service Process A critical part of adaptive assistance is the qualification of the employee. Due to the huge variety of qualification patterns and the problems in the assessment of them, the employee should choose the amount of information in the adaptation process by her/himself. Instead of defining instruction sets for different levels of qualification, a simple hierarchical model (level of instruction complexity, figure 7) is suggested.

Level 3: graphical visualisation of instruction Level 2: additional information

text

Level 1: simple work instruction

Remote Service

Level 4: video instruction with audio-visual material

Figure 7: hierarchical model: level of instruction complexity In this exemplary micro production scenario the VLCU detects a change in the spindle condition. On the basis of predefined rules the VLCU advices maintenance for example “Loss of clamping force, check springs!” Is the user a less-qualified technician or has never executed this maintenance before, which is registered in his/hers personal user profile, the maintenance process has to be authorized by a remote connection to the IPS² provider. The user is warned by a signal and after authorization the service process begins. In this case “Change the springs in the cylinder module” is the 1st level of the work instruction and is displayed in a defined area of the machine’s graphical user interface. If there is a lack of procedural knowledge regarding the steps of the working sequence the technician can switch to a higher level of instruction complexity. The 2nd level of instruction complexity describes the simple work instruction with additional text for each step of the procedure in a checklist. “Change the springs in the cylinder module by the following steps: 1. Loosen two screws M4 of the media module with a 3.0 socket wrench”. At least three components should be described: the operation, its position, and the tool which is needed. The user confirms the current step as soon as the action is finished and the next step of the checklist appears. If the step takes longer than necessary or the user requests more information, the 3rd level of instruction complexity is offered, a graphical visualization of the instruction. The description of operations only by text can be very difficult and abstract. An additional graphic like an exploded drawing with added symbols for actions can facilitate comprehension and provide additional information. Level 4 of instruction complexity explains each step of the instructions by also using aural and visual modalities if the user is still not capable to change the spirals. As the problem could not be solved by following the previous instruction steps, obviously support by an expert is needed. So if passing through all levels of instruction material does not lead to the fulfillment of the task, the remote service has to be contacted. The IPS² provider has different possibilities to provide this remote service, i.e. the remote assistance for the employee by an expert. One of these possibilities could be a shared-vision system. Especially in difficult and less well defined problem situations the communication through gaze information and aural instruction can lead to an efficient problem solving. Another reason for contacting the remote service could be, if the VLCU detects unforeseen events during the maintenance process. While loosing a screw in the first step of the maintenance procedure, the VLCU detects via an electronic torque wrench with a force sensor that much more power is being used than usually needed. The user

is instructed to stop the process and an expert is contacted via the shared-vision system. The expert detects a corroded screw. This might be due to high humidity in the factory hall of the customer, which was not expected by the provider. Corroded screws were not considered in the design of the maintenance workflow. The expert will give an error report with the information “humidity and corroded screws” as input, which will be processed by the VLCU. However, not always high humidity results in corroded screws. The best input for the workflow decision tree results out of the combination of applied torque force to loosen the screw, humidity, machine model, screw type etc. A statistical evaluation according to decision tree generation leads to a significant workflow tree for future processes, figure 8. Before success

force 120Nm

different  screw type



loosen screw

success use X‐screw

failed

open case

contact service by shared‐ vision system

After success

force 120Nm

humidity >80% for >100h

loosen screw

… true

screw coroded

use …

different  screw type

use X‐screw



success

open case

false

work instruction

sensors &  VLCU

failed

contact service by shared‐vision system

Figure 8: decision tree After finishing the maintenance procedure the user ends the process with a functional test. Is the test successful, work can continue and the completed task is registered to the user’s personal profile. If not, the remote service will be activated or go on, respectively. 4 SUMMARY In a maintenance scenario of a micro manufacturing spindle the user support with the use of a shared vision system, assisted by a VLCU generated decision tree has been shown. Further the VLCU enables condition prognostics for a reliable and cost effective production process. The selected use model “availability-oriented” determines design characteristics, e.g. modular construction and/or electronic locks, that differ from full user intervention possibility to a non-demountable product. The described concept enables IPS² providers to compete in the global market, to protect their knowledge, and to offer any solution from the physical product to a complex service. 5 ACKNOWLEDGMENTS We express our sincere thanks to the German Research Foundation (Deutsche Forschungsgemeinschaft) for funding this research within the Collaborative Research Project SFB/TR29 on Industrial Product-Service Systems – dynamic interdependency of products and services in the production area.

130

6 REFERENCES [1] Hudetz, W., 1998, Teleservice-state-of-the-art and future developments, in: IECON '98: Proceedings of the 24th Annual Conference of the IEEE. Vol. 4, pp: 2103-2107, ISBN: 0-7803-4503-7. [2] Abele, C., 2007, Branche kompakt - VR China Maschinenbau und Anlagenbau, Bundesagentur für Außenwirtschaft, Juli 2007. [3] Uhlmann, E., Meier, H, Stelzer, C., Geisert, C., Bochnig, H, Sadek, K., 2008, Design of PSS Based on Customer Requirements, First International Seminar on PSS, Bochum, Shaker; Auflage: 1 (Juni 2008), ISBN-10: 3832272275. [4] Tucker, A., Tischner, U. , 2005, New Business for Old Europe – Product-Service Development, Competiveness and Sustainability, Greanleaf Publishing, Sheffield, UK [5] Tan. A. R., McAloone, T. C., 2006, Characteristics of strategies in Product/Service-System Development, International Design Conference – Design 2006, Dubrovnik – Croatia, May 15-18. [6] Möller, B. et. al., 2005, FuE-Abschlussbericht des BMBF-Verbundprojektes "Intelligente Spindeleinheit" (ISPI), project coordination: Möller, B. - Paul Müller Industrie GmbH & Co. KG. [7] Uhlmann, E., Geisert, C., Hohwieler, E., 2008, Monitoring of Slowly Processing Deterioration of CNC-Machine Axes, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, Professional Engineering Publishing, vol. 222, no. 7 (2008). [8] Hohwieler, E., Geisert, C., 2004, Intelligent machines offer condition monitoring and maintenance prediction services, in: Teti, R. (editor): Intelligent Computation in Manufacturing Engineering, Proceedings of 4th CIRP International Seminar (CIRP ICME ´04), 30 June - 2 July, 2004, Sorrento, Italy. Neapel, 2004, pp. 599-604. [9] Oviatt, S.L. Multimodal interfaces. The HumanComputer Interaction Handbook: Fundamentals, Evolving Technologies and Emerging Applications. J. Jacko and A. Sears, Eds. Lawrence Erlbaum, Mahwah, NJ, 2003, pp. 286–304. [10] Jaschinski , C., Spiess, M., May, S. (1997): TeleService - Eine Einführung, Forschungszentrum für Rationalisierung an der RWTH, Aachen, Sonderdruck 13/97 [11] Massberg, W., Hermsen, M., Zuther, M., 2000, TELec – Multimedialer Teleservice: Technik – Organisation – Vermarktung – Erfahrungsberichte. Schriftenreihe des Lehrstuhls für Produktionssysteme, ISBN 3-8265-7408-7, Aachen: Shaker, 2000. [12] Arvika Konsortium: ARVIKA – Augmented Reality für Entwicklung, Produktion und Service, http://www.arvika.de/, last access: 25.08.2008. [13] Artesas Konsortium: ARTESAS – Advanced Augmented Reality Technologies for Industrial

131

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22] [23]

[24]

[25]

Service Applications. http://www.artesas.de/, last access: 25.08.2008. Stolz, M.F., 2007, Developement of intelligent, machine-oriented services for micromachining, IPAIAO-Forschung und Praxis, Vol. 448, ISBN: 3939890-01-4, pp 39-43. Fussell, S., Kraut, R., Siegel, J., 2000, Coordination of communication: effects of shared visual context on collaborative work, in: Proceedings of the 2000 ACM conference on Computer supported cooperative work, ACM New York, NY, USA, ISBN: 1-58113-222-0, pp: 21-30. Velichkovsky, B.M., 1995, Communicating attention: Gaze position transfer in cooperative problem solving. In: Pragmatics and cognition, Vol. 3(2), 199222. Milgram, P., Kishino, F., 1994, A Taxonomy of Mixed Reality Visual Displays. In: IEICE Transactions on Information Systems, E77-D (12), pp: 1321-1329. Westkämper, E., 1999, Technical Intelligence in the Manufacturing of the Future, In: Ashayeri, J., Sullivan, W.G., Ahmad, M.M., Tilburg Univ.: Flexible Automation and Intelligent Manufacturing 1999. Proceedings of the Ninth International FAIM Conference, New York: Begell House, 1999, ISBN: 1-56700-133-5, S.105-114. Buchholz, A., Middendorf, A., Ray, P., Reichl, H., Seliger, G., 2004, Increased Reliability through Assessment of Standard Components with Life Cycle Units, Electronics Goes Green 2004+, Berlin, Germany, p. 287-292. Seliger, G., Gegusch, R., Odry, D., 2007, Competitive Products and Processes, CIRP Competitive Manufacturing ‘07, Stellenbosch, SouthAfrica. Y. Yuan and M.J. Shaw, "Induction of Fuzzy Decision Trees," Fuzzy Sets and Systems, vol. 69, 1995, pp. 125–139. Quinlan, J. R., 1993, C4.5: Programs for Machine Learning. Morgan Kaufmann Publishers, 1993. Grudzien, W., 2002, Beitrag zur Steigerung der Nutzenproduktivität von Ressourcen durch eine Life Cycle Unit, Stuttgart : Fraunhofer IRB Verlag, 2002, ISBN: 3-8167-6174-7 Uhlmann, E., Meier, H. et al., 2008, Design of PSS Based on Customer Requirements, In: Transregio 29 - Product-Service Systems - Dynamic Interdependency of Products and Services in the Production Area, International Seminar on PSS, January 21-22, 2008, Bochum, Germany, ShakerVerlag, Aachen, p. 13-18. VDI/VDE (Hrsg.), 2004, User-friendly design of useware for machines - Design of dialogues for touchscreens, VDI/VDE 3850, Part 1 to Part 3.

Framework for the Integration of Service and Technology Strategies 1

2

2

3

E. Jühling , M. Torney , C. Herrmann , K. Dröder AutoUni, Volkswagen Aktiengesellschaft, Herrmann-Münch-Straße 1, D-38440 Wolfsburg, Germany 2 Institut of Machine Tools and Production Technology (IWF), Technische Universität Braunschweig, Langer Kamp 19B, D-38106 Braunschweig, Germany 3 Group Research, Volkswagen Aktiengesellschaft, Berliner Ring 2, D-38440 Wolfsburg, Germany

1

Abstract After sales service is a highly profitable business for manufacturers of technology-driven products. Due to this fact competitors want to share in high profit margins. At the same time after sales business has to deal with an increasing range of variants of products and technologies, shorter life cycles and changing customer demands. In spite of these manifold challenges, often neither after sales departments are involved in the early product development stage nor are customer demands and technical parameters considered in the service development processes entirely. Therefore an integration of service and technology strategies is necessary. This paper presents a framework for this integration that visualises the complex interdependencies and interfaces between service as well as product and motor vehicle workshop technologies. Keywords: Service Engineering, Technology Roadmapping, Automotive After Sales

Total volume p.a. Volume per variant TT Roadster

Volume / Variants

1 INTRODUCTION Organisational and technical challenges are determined by the interaction of different factors. Cross-Linked thinking is a method for the analysis of the interdependencies (Figure 1). E.g., the higher the economic success is, the higher is the technical progress. New markets mean rising investments and innovation. A problem-adequate form of organisation is a precondition for a good strategic position and economic success [1].

Source: Volkswagen AG

Beetle Dune

new markets

Bora

+

+ + investments + innovation + +

technological progress economic situation

_

job market

+ rationalization _

+ + product program

TT

Golf A3

Käfer

Touran

Octavia Caddy

Golf

Leon Ibiza

1950

+ + strategic + position

+ costs + new forms of + organization + + + imposts + + co-proprietor

1974

1985

TODAY

Figure 2: Increasing variant variety in automotive industry. Increased competition as well as increased customer demands also lead to faster implementation of new technologies in vehicles [2]. As a result, the development towards a highly diversified automotive market is accompanied by a continuous acceleration of product cycles and growing product complexity (Figure 3).

+ economic success _ equity capital +

_ involvement of employees social + + awareness claims for environmental + protection + “the more…..the more” / “the less….the less” _ “the more…the less” / “the less…the more”

Golf V Golf VI Golf IV

Figure 1: Situation analysis by cross-linked thinking [1]. The automotive industry but also other industries are embedded in a rapidly changing environment (Figure 2). Rising variant variety and/or rising individualising of vehicles on the one hand are determined by the possibility for differentiation in competitive markets and on the other hand by the customer demand for individualised products and the technological progress of producing variants or customer individual vehicles economically.

Complexity

+ customer demands + disposable income

Golf

Altea

Jetta

+

New Beetle

Golf III Golf II Golf I

1974

1983

1991

1997

2003 2008

Source: Volkswagen AG

Figure 3: Acceleration of product cycles in automotive industry.

CIRP IPS2 Conference 2009

132

The outlined development leads to radical changes in the service and after-sales markets. Today service intervals of 15,000 miles or more are usual (Figure 4). The share of electronic in vehicles has increased dramatically over the last decades [3]; [4]. But at the same time the product life span remained unchanged and even easily rose (1215 years). As consequences new workshop technologies are required and new challenges for an effective spare part management of electronic components after the end of production exist [3]. IN THE PAST

Increasing complexity (especially electronics)

• Oil change every 1,500 miles • Solely mechanical components • Sparse wiring • Visual inspection • Repair of failures

Proliferation of variants Longer service intervals

Increased competition and TODAY shorter model cycles • Intervals 15,000 miles or more • Networked electronic and mechatronic systems • Moe than 2 miles of wiring • Automobile diagnostic systems instead of visual inspection • Exchange of modules instead of repair

Figure 4: Challenges for automotive maintenance operations; adapted from [4]. As the customer expects high product availability, the increased product complexity requires an appropriate service offer. An adequate service organisation and the ability to handle different vehicle technologies efficiently are necessary. Apart from the relevance regarding brand image and customer loyalty also economic success is determined crucially by the service. Apart from the actual service achievement the sale of spare parts determines the turnover of automotive companies. On the whole, the different factors result in a strong pressure regarding innovation and costs on the companies, and the constraint to market the innovative products in a short period of time [2]. Finally, service is becoming the reservoir of challenges from a technological as well as organisational point of view. 2 TECHNOLOGY As it is described before, one of the major challenges facing service organisations like the automotive after sales is how to maximise the value of its investments in technology [5]. Despite that Farrukh et al. criticise that there is a lack of a systematic approach to manage technology. Often companies have a well-established new product development process but still come up against problems if technologies and products have to be developed simultaneously [6]. There is also the fact that demands from automotive after markets have only little influence in the early product development process. “Often a single focus (...) on e.g. design for production in order to cut down costs, e.g. by using more integrated parts, may result in increased costs for service and endof-life treatment, instead of reducing the overall cost for the product, i.e. the total life cycle cost increase” [7]. 2.1 Technology Management One answer to this problem can be a holistic Technology Management like it is defined by the European Institute for Technology & Innovation Management. “Technology management addresses the effective identification, selection, acquisition, development, exploitation and protection of technologies (product, process and infrastructural) needed to maintain a market position and

business performance in accordance with the company’s objectives" [8]. In the context of the automotive sector where especially electronics rapidly change effective technology management depends on the ability to forecast trends as well as to anticipate their potential impacts [9]. An appropriate forecast and planning method is needed, which links both technology and business objectives [10]; [11]. In case of automotive aftermarket “service operations along with the required skills as well as remanufacturing technologies and the involved failure diagnosis requirements” [4] have to be combined with vehicle segment and workshop specific objectives. 2.2 Roadmapping Basics Roadmapping is such a foresight method [12], which assists technology strategy creation and management in cases where cross-functional alignment and integration are key requirements. For that reason it has evolved as a best practice, mainly for large, global organisations [13]; [14]. After the focus of interest has always been on the end result, the roadmap itself, and not on the process in the 1990s, nowadays technology roadmapping is defined as “a needs-driven technology planning process to help identify, select, and develop technology alternatives to satisfy a set of product needs.” [15]; [16] Roadmap Formats Nevertheless many companies fail to apply roadmaps. One reason is that a wide range of roadmap formats 1 exists , which have to be customised to specific needs of the firm and its business context [17]. The most common form of technology roadmaps is a multi-layered graphical illustration of how technology and product developments link to business goals (Figure 5). An integrated time axis indicates when particular circumstances, events, objectives, products and technologies are expected to emerge [18]. Time Business Goals

Product

Technology

Figure 5: Format of a common technology roadmap. Roadmappping Process In addition to this diversity of forms, there is little practical support in implementing a roadmapping process and keeping it alive. Although there have been some efforts to share experiences companies typically have to reinvent the process. Authors who have summarised key technology roadmapping process steps are Bray and Garcia [15], EIRMA [19] and Groenveld [20] [17]. However, these processes do not include detailed 1 For further information see: Phaal, R.; Farrukh, C. J. P.; Probert, D. R., 2001, T-Plan: the fast-start to technology roadmapping: planning your route to success, Institute for Manufacturing, University of Cambridge, 2001.

133

Roadmapping Process

Roadmap

Horizon Broadening

Market Pull

Time

Market Information

Workshop Objectives

Service-Market-Analysis Periodic Review & Adaptation

Market-Technology Options Evaluation

Roadmap Creation

Workshop Technology

Roadmap Release

Vehicle Technology

Technology Assessment

Technology Push

Identification of Technology Available/Feasible

Source: partly adopted from EIRMA (1997), p. 24

Horizon Broadening

Figure 6: Technology roadmap and roadmapping process for automotive after sales; partly adapted from [19]. guidance. For that reason Phaal et al. have tried to fill this gap by the development of the “T-Plan fast-start approach” [17]. The authors pick up the best practice strategic planning process of EIRMA, which widens the roadmapping process to market pull and technology push aspects, and facilitate the process with workshops [17]. Although Wells et al. [5] are some of the few authors, who emphasise the use of roadmaps for service 2 organisations , often the product and not the service is the centre of attention. According to EIRMA the only difference is, that “industries close to the consumer are responding to targets set by the market place, while industries further from the consumer are setting their own targets as a consequence of developing scientific knowledge” [19]. This classification is not suitable for automotive after sales services. For one thing the automotive after sales is close to the consumer, for another thing it has to deal with established vehicle technologies. Consequently roadmapping for automotive after sales has to evaluate options to fill the gap between market pull and technology push as it is shown in Figure 6.

Nevertheless, a change in the product metrics in the sense of design for service must not be neglected in the long run. On the other side (market pull) customer oriented service strategies with excellent service quality and processes is the input. In case of automotive aftermarket workshops are intermediaries between the original equipment manufacturer and the customer. How this service infrastructure and the corresponding processes could be optimised and adapted for future challenges is described in the next chapter (service).

Data The next challenge is to infuse the automotive after sales roadmap with data. Data has to be global, timely, accurate and meaningful [21]. On the side of technology push, technology has to be divided by vehicle technology and workshop technology. For automotive after sales vehicle technology is established on short notice. Services could only be improved by innovative workshop technology, which has to be assessed. Data concerning future vehicle technology usually exist in the shape of research and development roadmaps. Consequently the sub layer of an automotive after sales roadmap could easily be compacted and afterwards copied and pasted.

3

2 For further information see: Wells, R.; Phaal, R.; Farrukh, C.; Probert, D., 2004, Technology Roadmapping for Service Organization, Research Technology Management, March 2004, Vol. 47, Issue 2, 46-51.

134

2.3 Outlook After the evaluation of options to fill the gap between market pull and technology push the automotive after sales roadmap can be drawn. To stay current with the information the roadmap portrays, the roadmapping process contains a periodically review and adoption phase [10]. Furthermore the horizon has to be broadened, which could be associated with budget or strategy cycles [17]. SERVICE

3.1 Service in the automotive industry Services in the automotive industry include all activities that create benefits for car customers over the car’s lifecycle. To this regard, these services are classified as product-related services which can be defined in the following way [22]: Concerning the car, services can be differentiated in technical or no-technical services. While technical services cover all activities, which preserve the mobility and quality of the car like accident repairs or inspection. Furthermore, automotive services can be differentiated in services that: 1.) obtain mobility (like car-sharing, finance or leasing) 2.) preserve mobility (like maintenance, mobility guarantee) 3.) expand mobility by offering services (like customer club, travelling support etc.)

between “physical product” and “service product” will be used in the remainder of this article. In order to develop new services in an efficient and successful way, adequate models for the planning and development of services are necessary. A new research discipline Service Engineering was founded for the systematic development of services. The notion of “Service Engineering” is based on the assumption that services can be developed as physical products. Therefore, “Service Engineering can be defined as the systematic development and design of services, using suitable models, methods, and tools” [24]; [25] Service Engineering mainly targets the improvement of service planning and service developing procedures, resulting in more professional services.

Finally, also automotive services can be divided into “presales”, “sales”, and “after-sales” services depending on their stage in the life cycle. The two former services are focusing on sales-promotional and sales-supporting activities like financing, advice for the product choice and configuration. The latter services (“after-sales”) include all activities ranging from the usage phase to the end-of-lifestage, such as maintenance, spare part (management) or recycling. As stated above, after sales services are a highly profitable business for car-manufacturers. Consequently, this paper focuses on these kinds of services whereas the technical services respectively the workshop technology is in the spotlight. 3.2 Service Development

Procedure Model for Service Engineering The general object of a model for developing services is to structure and manage the complex, multi-disciplinary service development process more efficiently. The development process for Service Engineering (in the automotive industry) with its separate stages is described in the process model visualised in Figure 7. By combining service and development methods and tools into single process steps, the development process assists service development from the initial concept down to the final implementation, allowing for a systemised development and assessment of services. The process is divided into three main stages: “service planning”, “service conception” and “service implementation”. The first phase, “service planning”, includes the situation analysis of the company and the environment analysis, in order to identify the requirements of the stakeholders (i.e. customers, manufacturers, staff, legislators, distributors). However, depending on the stakeholder group, the derived requirements can have a different and partly opposing focus. Based on this result, a target system will be deducted from the later design and assessment of the service concept. The result of the “service conception” stage is an evaluated, model-based service concept. According to the characteristics of services, the three dimensions

Service Development in the Automotive Industry Generally talking, the development and planning of services and service strategies for the automotive industry are lacking in formalised models, as they are developed in an unsystematically and spontaneous way. Simultaneously, the after sales market is influenced by a variety of changes, such as technical, economic, ecological, socio-cultural, or legal aspects (see figure 1). Examples are new and more complex car technologies, changing competition conditions affected by legislative mandates, aging in society, decreasing customer loyalty etc., which cause manifold factors that have to be taken into account. Alongside this multi-disciplinary background the main question is how to design the service development process efficient in order to ensure highquality service processes. Definitions and Goals of Service Engineering Equivalent to physical products, the development process for services needs to be systemised or standardised. Due to this fact, it is essential to have a more productorientated view on service. That means it has to be seen as a separate “product”, which requires an adequate development process. Consequently, services are not solely seen as “black boxes” but as “a designable part of the business activities” [23]. Hence, the differentiation

Service development process Service Conception

Service Implementation

Service Product Model (What?)

Service concept

Stakeholders requirements Target system

Situation Analysis company and enviroment

Service Planning

Process Model (How?)

improvements

Adaption caused by changing resources

(Whereby?)

information infrastructure

targets for adaption

Resource Model

staff

Requirements

technical equiment

Workshop technology

Figure 7: Service development process in particular regarding workshop technology.

135

Customer: all

Objectiv (Target group)

Workshop reference model Typ of

Objectiv customer (Target group) Serivce Product model Business unit A

Type of car

Durability of car

Process model

Serivce Product model

workshop format A

Business unit A

Typ of car: all Duralility of car: until 12 years

Process model Supporting Processes

Resource model Staff

Business unit B

Resource model

Business Prozesses

information

Business unit C

Supporting Processes

Staff

infrastructure

Technical equiment

Business unit D Business unit E

information Business unit B

Customer: all

Business Prozesses

Objectiv (Target group)

Typ of car: one brand, only accident repairs Duralility of car: until 20 years

infrastructure

workshop format B

Business unit C

Serivce Product model

Process model

Business unit A

Supporting Processes

Business unit B

Business Prozesses

Resource model Staff

Technical equiment information

infrastructure

Technical equiment

Figure 8: Workshop reference model. structure, process, and outcome have to be considered for a systematic design of services in every step of the development process [26]. Due to this fact, the service concept is finally described and documented by the three following models [27]. •





Service product model The service product model describes the outcome respectively the planed range of offered services, their content and characteristic i.e. the amount and type of service levels or the structure of the service packages. The service product model is mostly determined by the stakeholders and market requirements identified in the situation analysis [28]. Process model While the product service model describes what kind of services will be offered, the process model specifies how the aspired service will be available. Therefore, based on the service product model, a corresponding process model has to be derived, which documents the needed processes and interfaces to realise high-quality services. Resource model Determined by the two former models, the resource model clarifies what kind of resources is needed to fulfil them. This includes aspects such as staff and information requirements, infrastructure but also technical equipment, which includes the considered workshop technology (see chapter 2). If the planed technology for the technical equipment is not available or another (newer) technology respectively technical equipment opens up better possibilities, it has to be displaced and the resource model needs to be modified. Consequently, the process and product model have to be adapted as well.

136

Finally, the phase “service implementation” attends for testing the service concept and confirms improvements for adaptation in a feedback loop. 3.3 Workshop format The main purpose of a (automotive) workshop is to offer service in the car’s after sales stage. A workshop format describes i.e. the structure, dimension and objective of an automotive workshop based on a reference model shown in Figure 8. The reference model (Figure 8) is structured according to the discussed procedure model in chapter 3.2 in the three perspectives outcome, process and resources. On basis of this reference model specific workshop formats are configured in dependence of their objectives, which are determined by the needs of their target group. Thereby every target group consists of a “type of customer” (i.e. business people, elderly people, women etc.), a “type of car” (i.e. special brands, old-timer, utility vehicle) and the “durability of car” (i.e. till 4, 8 or 12 years). Hence, every workshop format has its own combination of product, process and workshop model. 4

FRAMEWORK

4.1 Reason for an integrative development As described in the chapters before, both disciplines roadmapping and service engineering for after sales services are dependent on each other. For one thing new workshop objectives, the upper layer of a roadmap (see Figure 6), are partly defined in the course of service engineering. For another thing the complete roadmap contains crucial data especially vehicle and workshop technologies to initiate a service engineering process. To keep roadmapping alive and adapt workshop formats to

Roadmap

Time

Workshop Objectives

Workshop Technology

Design for Service

Vehicle Technology

Adaption Workshop Format

a



z

Vehicle Technology Time Technology

internal

external

internal

external

internal

external

(1…5)

(1…5)

(1…5)

(1…5)

(1…5)

(1…5)

a

Stakeholders requirements Target system

Objectives

Situation Analysis company and enviroment

Market Drivers

Service Product Model (What?)

Service concept

Vehicle Technology/Workshop Objectives Matrix

Process Model (How?)

Resource Model (Whereby?)

… z

staff

information infrastructure

technical equiment

Workshop technology

Modified Workshop Objectives Time

Innovative Workshop Technology

Figure 9 : Framework for the integration of service and technology strategies.

4.2 Framework for the integration of service and technology strategies In Figure 4 the interfaces between roadmapping and service engineering for automotive after sales are illustrated. Because roadmapping as well as service engineering are no singular task (see chapter 2.3), but continuing processes, Figure 9 contains a cycle. Below, the cycle with its interfaces, data basis and supporting tools will be described.

Especially technologies which result in manifold negative effects should be analysed more precisely. If there are vehicle technologies, which could not be handled within a regular workshop in due consideration of market drivers more specifically workshop objectives, project management will be launched to close the gap. As a result the automotive after sales roadmap like it is illustrated in Figure 6 can be drawn. If no meaningful solutions could be found to fill the gap between vehicle technology and workshop objectives through workshop technology, a change in the product metrics in the sense of design for service will be essential.

Vehicle Technology/Workshop Objectives Matrix One of the two central elements of this framework is the “Vehicle Technology/Workshop Objectives”-Matrix. This matrix consists of the two dimensions “technology” and “market drivers”. Data basis for the dimension technology are research and development roadmaps concerning vehicle technology, which usually exist within the firm (see chapter 2.2  Data). On the other hand market drivers, for the main part workshop objectives, as one result of the situation analysis (see chapter 3.2), generate the horizontal axis. To highlight the importance of particular market drivers a prioritization by weighting the elements is possible. Additionally market drivers can be divided by internal and external aspects to meet the specific requirements of the automotive after sales, being determined by technology and at the same time near to the customer. After the axes are defined, the fields within the matrix have to be filled. In the shape of a pairwise comparison each technology has to be rated against the accomplishment oft the market drivers. The accomplishment can be positive as well as negative.

Adaption of the workshop format According to the generated automotive after sales roadmap and its changes for workshop technology (respectively for the technical equipment), the resource model of the workshop format and its service concept have to be adapted as illustrated in chapter 3.2. This needs to be done in an adequate way to prevent the specified objectives of the workshop format. If these objectives cannot be achieved with the given technology, for a short time solution the workshop objectives and therefore the workshop format have to be modified. Furthermore the market drivers have to be adapted and replaced in the “Vehicle Technology/Workshop Objectives”-Matrix. For a mid-term solution suitable, new innovative workshop technologies have to be developed. In the case of an adaptation, the process cycle of the framework has to start again with the “Vehicle Technology/Workshop Objectives”-Matrix. It has to check if the adapted objectives are henceforward achievable with the planed technology of the automotive after sales roadmap.

technology trends as well as trend reversals an integrated development framework is required.

137

Conclusion The framework shows how the disciplines roadmapping and service engineering interact and among which phases data is transferred. Especially the “Vehicle Technology/Workshop Objectives”-Matrix functions as a manageable tool to facilitate the process of exchange. 5 SUMMARY Against the background of the multiplicity challenges facing the automotive after sales, this framework offers an integrative approach to be all set. This framework could be used for sensitising decision makers of product and service departments to complex interdependencies and necessity for overall cooperation and collaborative development. Not only interfaces between market pull and technology push are considered and merged through roadmapping and service engineering, also interfaces to related disciplines like design for service are revealed. The next challenge will be the rollout of this approach in the automotive after sales. Especially the affiliation of the framework in existing management processes and the adaptation to keep it up to date will be the centre of attention. 6 REFERENCES [1] Gomez, P.; Zimmermann, T., 1992, Unternehmensorganisation. Profile, Dynamik, Methodik, Frankfurt Main u.a., Campus-Verl. [2] Herrmann, C.; Mansour, M.; Mateika, M.: Strategic and Operational Life Cycle Management – Model, Methods and Activities In: Proceedings of the 12th International CIRP Seminar on LCE 2005, Laboratoire 3S, Grenoble, France, April 3-5, 2005 [3] Herrmann, C.; Graf, R.; Luger, T.; Kuhn, V.: ReX Options in Closed-Loop Supply Chains for Spare Part Management. In: Proceedings Global Conference on Sustainable Product Development and Life Cycle Engineering Berlin, 2004, pp. 139-142. [4] Steinhilper, R.: Automotive service engineering and remanufacturing: New technologies and opportunities. In: LCE 2008, 15th CIRP International Conference on Life Cycle Engineering Applying Life Cycle Knowledge to Engineering Solutions. 17-19 March 2008, Sydney, Australia, 2008. [5] Wells, R.; Phaal, R.; Farrukh, C.; Probert, D., 2004, Technology Roadmapping for Service Organization, Research Technology Management, March 2004, Vol. 47, Issue 2, 4651. [6] Farrukh, C.; Fraser, P.; Hadjidakis, D.; Phaal, R.; Probert, D.; Trainsh, D., 2004, Developing an Integrated Technology Management Process, Research-Technology Management, Volume 47, Number 4, 1 July 2004, 39-46. [7] Lindahl, M.; Sundin, E.; Sakao, T.; Shimomura, Y., 2007, Integrated Product and Service Engineering versus Design for Environment – A Comparison and Evaluation of Advantages and Disadvantages, Advances in Life Cycle Engineering for Sustainable Manufacturing Businesses, Waseda University, Tokyo, Japan, June 11th-13th, 2007. [8] European Institute for Technology & Innovation Management, http://www-eitm.eng.cam.ac.uk/ mission.html/ mission.html

138

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

Rossini, F. A.; Porter, A. L., 1999, Technology Forecasting, in: Dorf, R. C. (ed.): The Technology Management Handbook, Berlin: Springer (1999), 5-26 - 5-31. Lizaso, F.; Reger, G., 2004, Scenario-based Roadmapping – A Conceptual View, EU - US Seminar "New Technology Foresight, Forecasting & Assessment Methods", Seville, 13 - 14 May 2004. Probert, D. R.; Farrukh, C. J. P.; Phaal, R., 2003, Technology roadmapping - developing a practical approach for linking resources to strategic goals, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture Heft Volume 217, Number 9 / 2003. Laat, B. de, 2004, Conditions for effectiveness of roadmapping, a cross-sectional analysis of 80 different exercises, EU - US Seminar "New Technology Foresight, Forecasting & Assessment Methods", Seville, 13 - 14 May 2004. Whalen, P. J., 2007, Strategic and Technology Planning on a Roadmapping Foundation, Research-Technology Management, Volume 50, Number 3, May-June 2007, 40-51. Phaal, R.; Farrukh, C.; Mitchell, R.; Probert, D., 2003, Starting-Up Roadmapping Fast, Research Technology Management, Volume 46, Number 2, 52-58. Garcia, M. L.; Bray, O. H., 1997, Fundamentals of Technology Roadmapping, http://www.sandia. gov/PHMCOE/pdf/Sandia'sFundamentalsofTech .pdf. Bray, O. H.; Garcia, M. L., 1997, Technology Roadmapping: The Integration of Strategic and Technology Planning for Competiveness, PICMET: Portland international conference on management of engineering and technology, Portland, OR (United States), Jul 1997, 27-31. Phaal, R.; Farrukh, C.; Probert, D., 2004, Technology roadmapping - A planning framework for evolution and revolution, in: Technological Forecasting and Social Change, Volume 71, Number 1, January 2004 , 5-26. Probert, D. R.; Farrukh, C. J. P.; Phaal, R., 2003, Technology roadmapping - developing a practical approach for linking resources to strategic goals, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, Volume 217, Number 9 / 2003. EIRMA, 1997, Technology roadmapping delivering business vision, Working group report, European Industrial Research Management Association, Paris, N°52. Groenveld, P., 2007, Roadmapping Integrates Business and Technology, Research Technology Management, Nov/Dec2007, Vol. 50, Issue 6, 49-58. Grossman, D. S., 2004, Putting Technology on the Road, Research-Technology Management, Volume 47, Number 2, 1 March 2004 , 41-46. Reindl, Stefan, 2005, Mobilitätsdienstleistungen in der Automobilwirtschaft, in: Diez, Willy; Reindl, Stefan; Brachat, Hannes (ed.), Grundlagen der Automobilwirtschaft, 4. Aufl. München: Auto Business Verlag, S. 423–466.

[23]

[24] [25]

[26]

[27]

[28]

Fähnrich, K.-P.; Meiren, T.; Hertweck A.; Baumeister, M.; Demuß, L.; Gaiser, B.; Zerr, K., 1999, Service Engineering, Ergebnisse einer empirischen Studie zum Stand der Dienstleistungsentwicklung in Deutschland, IRB, Stuttgart, 1999. Shostak, L., 1984, Designing Services that Deliver. Harvard Business Review 62, 73-78. Bullinger, H.-J., 1999, Entwicklung innovativer Dienstleistungen, in: Bullinger, H.J. (ed.), Dienstleistungen – Innovation für Wachstum und Beschäftigung, Wiesbaden 1999, 49-65. Bullinger, H.-J.; Schreiner, P., 2006, Service Engineering: Ein Rahmenkonzept für die systematische Entwicklung von Dienstleistungen, in: Bullinger, H.J.; Scheer, A.W. (ed.): Service Engineering: Entwicklung und Gestaltung innovativer Dienstleistungen, 2. Auflage, Berlin et al. 2006, 53-85. Scheer, A.-W.; Grieble, O.; Klein, R., 2006, Modellbasiertes Dienstleistungsmanagement, in: Bullinger, H.J.; Scheer, A.-W. (ed.), Service Engineering: Entwicklung und Gestaltung innovativer Dienstleistungen, 2. Auflage, Berlin et al. 2006, 19-52. Husen, C. van, 2007, Anforderungsanalyse für produktbegleitende Dienstleistungen, Heimsheim, Jost-Jetter Verlag.

139

Strategy Assessment and Decision based Implications for Integrated Product-Service-Suppliers 1

1

R. Schmitt , S. Hatfield 1 Department of Production Quality and Metrology, Fraunhofer Institute for Production Technology IPT, Steinbachstr. 17, Aachen, Germany [email protected], [email protected]

Abstract In order to provide effective Product-Service-Solutions, so-called Integrated Solutions, especially industrial SME face challenges regarding the selection of appropriate business models regarding their internal organisation. Contingency Theory claims that a good fit between structural, strategic and external factors is necessary for a company’s success. The servitization strategy can be organised on a continuum of Individualisation and Standardisation. It is, therefore, vital to align business processes, organisational structure and leadership styles with the corresponding strategy. An assessment concept is presented which allows strategy identification as well as implementation guidelines for the organisational development of Integrated Solution Suppliers increasingly providing Product-Service-Solutions. Keywords: Service, Model, Process

1 INTRODUCTION Many industrial companies have decided to enhance service offerings in order to complement their core products [1]. Reasons for this range from the need to establish single selling propositions by individualised offers to the offering of mass services in order to make core products profitable [2]. Industrial services recently account for a stronger growth in turnover than industrial goods do [3]. This is due to lower invest costs and the absence of warehousing. Integrated Solution Suppliers offer product-service bundles that are individual and enable the customer to solve problems or substitute lacking competencies [4]. Despite the strong will of offering services, especially for small and medium sized enterprises (SME), the definition of a business model for the providing company often poses many questions and challenges. These begin with a clear definition of a strategy and end with the successful implementation of a business model including the design of the inherent processes and leadership aspects. Well known principles from strategic and business research can be applied here. Hereby, it is crucial to align the organisational structure and processes with the defined strategy, in order to guarantee a frictionless implementation. Leadership aspects such as management style and tools play a vital role in supporting the implementation and maintaining the achieved changes [5]. In order to identify the integrated service strategy, companies require to know the linkage between corporate goals and possible service strategies. Furthermore, they need to know how much effort it takes to conduct a strategic change from their current positioning to a future desired state. For this, a strategy assessment support is welcomed. Once the firms know what strategy they want

CIRP IPS2 Conference 2009

140

to follow, possible organisational actions need to be indicated. 2

THORETICAL BACKGROUND

2.1 Contingency Theory and Resource-Advantage Theory The Contingency Theory claims that the best fit between the external environment, the internal strategy and the organisational orientation has to be achieved in order for a company to perform successfully. Mintzberg and other scientists underlined this theory by empirical studies [6]. Furthermore, concerning the organisational orientation, several further factors need to be considered, that enable a firm to establish an internally consistent organisation. This is described in the Resource-Advantage Theory of Competition, whereby the perception of a company should be such, that the customers value the resources as advantageous over other companies [7]. A superior relative customer value is achieved by resources that are difficult to imitate and accumulate. Several resources can be combinded with each other in order to create such a superior value and therfore a competitive advantage. Neu and Brown recently define internal resources as Strategy, Processes, Structure, Human Resource as well as Measurement and Rewards [8]. More concrete success factors for integrated Product-Service-Solutions are described in the following. 2.2 Concept and Success Factor for Integrated Services Especially for Customer Solutions, further success factors can be identified, which need to be considered when developing Product-Service-Solutions. Tuli et al. [4] found 4 core phases that are essential for a perceived customer solution. These phases result by means of depth interviews and imply a new process-centric view on

solutions rather than a product-centric view defining solutions as customised and integrated goods and services. The emphasis clearly lies on the interaction of companies and their long-term relationship. The first important phase is the Requirements Definition and emphasises the importance of the deep understanding of cutomers’ needs that sometimes are not even easy to articulate. Close relational ties are needed in this discovery process and form the basis for the definition of future needs. Customization and Integration involve designing, modifying and selecting products to fit into a customer’s environment. The concept of the customer’s role as a co-creator of value becomes of importance here. Next, Deployment includes the delivery of products and their installation into a customer’s environment. Additional modifications might be needed at this stage. The staff competencies regarding the direct interaction with customers are determining factors of the perceived solution quality. Lastly, Postdeployment Support encompasses deploying new products in response to evolving customer requirements but also the ongoing relationship and providing solutions for emerging problems with the obtained products are central. All four phases must be performed well by the supplier in order for the customer to perceive the solution as high quality. In addition to this concept, success factors are identified, which concentrate on the company’s organisation: contingent hierarchies, documentation emphasis, incentive externality, customer-interactor stability and process articulation. 2.3 The Continuum of Individualisation and Standardisation for Integrated Services Porter [9] described two main strategies for achieving competitive advantages: cost-leadership and qualityleadership. The first mentioned aims at achieving economies of scale and therefore cost-reduction by mass production. Here lies close relation to Standardisation activities in order to render the underlying processes more efficient (see figure 1). The second strategy aims at a high degree of differentiation against competitors in form of single-selling-propositions. Especially product accompanying services propose promising potentials as they can provide individual problem solutions. This is referred to here as Individualisation.

Competition Strategy CostLeadership

Differentiation

Standardisation

Individualisation

effectiveness [10]. These theories are not new but companies following the novel trend of offering ProductService-Solutions need indicators in order to decide which Solution Strategy to follow. 3

REQUIREMENTS FOR STRATEGY IDENTIFICATION AND ORGANISATIONAL IMPLICATION Considering the Contingency Theory as well as the Outpacing approach, processes, organisational structures and leadership methods should differ between Individualisation and Standardisation in order to be consistent with the currently followed strategy. This means, if the Service Strategy implies Standardisation, then consistent orientations and standardisation methods should be implied on all organisational levels, Individualisation respectively (see figure 2). For this purpose, the implications of Standardisation and Individualisation Strategies are explained in the following regarding the needs of Product-Service-Suppliers.

Solution Strategy Organisational Structure Process Design Leadership Style Figure 2: Organisational design aspects depending on focussed Solution Strategy. 3.1 Indicators for the Continuum of Standardisation and Individualisation Individualisation and Standardisation can be seen as two endpoints of a continuum. This means, that companies can be located anywhere on this continuum between the two extremes. If a Product-Service-Supplier has elaborated solutions which can easily be produced in larger quantities for which there is a market, then he will tend to the direction of Standardisation. A ProductService-Supplier might also want to develop specific problem solutions for individual customers for very complex conditions in order to raise customer satisfaction and loyalty. This supplier will tend to the direction of Individualisation. These strategies are similar to those articulated by Porter, but they focus less on the market than on the desired type of solution. Expert Interviews with three SME from the research project ‘HyPro’ generated the main goals for these two strategy types. These are listed as extracted criteria in table 1 should be the basis of a strategy assessment.

Solution Strategy Figure 1: Solution-Strategy derived from ‘traditional’ Strategy. Porter [9] claims that the decision for one of these strategies must be made to avoid being ‘stuck in the middle’. Hybrid competition strategies imply switching from one strategy to the next at the right moment by which the dilemma is erased and competitors overtaken. This strategy, also known as ‘Outpacing’, requires a high sensitivity towards competition, product value and cost

Standardisation solution elaboration

Individualisation specificity/individuality of solution

quantity of production

complexity of customer situation

market opportunity

need for high customer satisfaction and loyalty

profitability of core products (non-service)

competition for core products (non-service)

Table 1: Indicators for different Solution Strategies. Furthermore, a differentiation between the current and a future strategic situation should be mapped, in order to estimate the effort of further strategy implementation.

141

The following implications are based on literature research on organisational, process and leadership issues. 3.2 Implications for the Organisational Structure Following the idea of Standardisation and Individualisation of solutions, organisational structures need to reflect these strategies by consistent design. As Product-Service-Suppliers, so-called Solution Suppliers, are considered, the focus shall lie on the organisation of service units as they need to complement the existing manufacturing units in order to develop integrated Product-Service-Solutions. Important preconditions for Standardisation are efficient structures [11]. These are established by more or less independent business units. Regarding the development of services, service departments and service organisations offer high transparency of costs and profits as well as the opportunity to use incentive or working hour models. These are important fundamentals in order to achieve a higher degree of professionalism and also more rational processes [12]. With a high degree of Standardisation, also service offerings for external products can be made. An important precondition for Individualisation are effective structures that guarantee flexibility in order to be able to react to specific customer requirements [12]. Such flexibility is achieved by project organisations and the flexible integration of experts from various departments across the organisation. These structures can also be established as secondary structures accompanying primary business units. The high involvement of experts corresponds to the concept of contingent hierarchies [8], meaning that customers can rely on the fact that the expert for their problem will also be the one to give advice and decide on measures to be taken. The basis for a good project organisation is a mutual interest of all involved actors, so that splinter groups are avoided and a collaborative working environment is realised [12]. If these conditions are fulfilled, then a strong integration of know-how across business units, necessary for individualised solutions, is achieved. Therefore, the Product-Service-Supplier’s need for the criteria in table 2 should be the basis for the selection of organisational structures to support the chosen Solution Strategy. Standardisation

Individualisation

cost and benefit transparency

know-how integration across units

rational processes

mutual interest and collaboration

incentive and working hours models

cross-selling potentials

service offering for external products

one-face-to-thecustomer policy

failure avoidance / minimal risk-taking

flexibility

Table 2: Indicators for different organisational structures. 3.3 Implications for Core Processes Concerning Standardisation efforts within processes of Product-Service-Suppliers, the greatest potentials lie within a well structured and transparent order processing. Therefore, means of process visualisation, such as Service Blueprints, are of great importance and have to indicate configuration points and sub-processes when needed. Also, key performance indicators that are able to monitor costs and lead times provide vital information for controlling mechanisms. A structured solutions portfolio is the basis for this as it allows a cost-based pricing of the final solutions.

142

Individualisation efforts focus on the collaborative development of the desired solution [13]. The collaboration lies not only between certain internal experts and units but also with the customer as a cocreator of value. Within the solution development, the transparency of the involved actors with their authorisations and function is more important than the detailed description of single steps within the process. Know-how exchange becomes a necessity and can therefore also be organised by key-accounts. Nonetheless, the definition of interfaces between all parties must be described along the four phases of integrated solutions, as described in section 2.2. The basis of individualised solutions here can also be provided by pre-arranged modular solutions. Indicators, that need to be considered here are listed in table 3 below. Standardisation emphasis on order processing

Individualisation emphasis on solution development

availability of structured portfolio

intensity of interaction between experts

availability of clear process descriptions

intensity of interaction with customers

availability of key performance indicators

project focus outweighs product focus

Table 3: Indicators for different core process structures. 3.4 Implications for Leadership Approaches Following the arguments of the former two sub-sections, two leadership styles can be indicated. one leadership style is needed for the monitoring and controlling of standardised processes as well as the use of incentives. This is given by the transactional leadership approach [14]. Individualisation requires flexible actions, a high degree of self-organisation of staff and visions able to commit employees to an overall vision and enhance joint efforts. This is what the transformative approach is aimed at. The first approach, is based on transactions between leaders and followers. A desired set of actions and behaviour is negotiated and rewarded by the management. Incentives and rules are therefore the most common management techniques applied in the case of Standardisation. Deviations from the rules are followed by corrective actions. The transformative approach relies on strong ideals, intellectual stimulation, individual consideration and inspirational motivation. Leaders able to provide all these aspects are often described as charismatic and act as role models for their employees. They do not so much tell their staff what and how to do something but coach them to set the right priorities, see issues from different aspects and make the right decisions for the group or project goals. Resulting indicators for adequate leadership styles are listed in table 4. Standardisation clear goals and transparent requirements

Individualisation ideals and visions to identify staff with

monitoring of deviations and corrective actions

consideration of individual challenges and goals

transparent incentives to motivate and commit them to the tasks

inspiration and stimulation by coaching for new ideas

Table 4: Indicators for different leadership styles.

3.5 Appraisal of Solution Strategy and Implications Within this section, the end-points of the continuum of Individualisation and Standardisation have been considered to make clear where emphasis can lie with respect to a chosen Solution Strategy, i.e. ProductService-Strategy. Nevertheless, is must be evident that nuances and mixtures lay between the two end points of this continuum and need to be considered with care. Furthermore, regarding a chosen strategy, implication guidelines can be articulated but with respect to the heterogeneity of organisations, they can not be seen as total measures of action but rather points of advice and reflection. Nonetheless, an assessment tool is aimed at, which is able to support in identifying Solution-Strategy and which consequently has a function to feed back corresponding and consistent advisory comments. 4

OUTLOOK ON DESIGN OF SOLUTION STRATEGY ASSESSMENT TOOL In the following, the basic requirements for the Solution Strategy Assessment Tool are defined and an outlook on its design will be given. This tool represents a decision support model which allows an indication on which organisational measures are appropriate for the chosen Solution Strategy. The requirements follow the contents of section three and are complemented with requirements for the resulting data analysis. From these, specifications for the design are derived. 4.1 Assessment Tool Requirements First, as mentioned in section 3.1, the current position of Solution-Supplier on the strategy continuum should be assessable (strategic status quo). Second, a future set of goals should be assessable and allocated to a future position on the strategy continuum (absolute strategic goal). Third, the difference from current and future position on the strategy continuum should be determined to indicate the need of action (relative strategic goal). Fourth, all indicators in section three should be included in either the current or the future state analysis by the formulation of questions/items (content validity). Fifth, an algorithm on absolute and relative strategic goal should allow the systematic selection of strategic organisational guidelines (declarative validity) - without numerical values resulting on the middle of the continuum. Sixth, the advisory guidelines should include under which circumstances they pose possible risks. For example, an independent service department poses too high initial and coordination efforts, if the percentage of employees does not exceed 10 %. Seventh, the feedback of possible advisory actions should be in immediate timely relation to the completion of items in the assessment. 4.2 Resulting Assessment Tool Specifications From the above listed requirements, concrete specifications can be derived to fulfil them. First, the formulated items and their appraisal, by rating of importance or degree of consent, must be allocated to an underlying scale depicting the continuum of Individualisation and Standardisation. Second, the scale should prohibit the achievement of numerical values in the middle of the continuum but pose a forced choice appraisal. A 6-point scale, e.g., would be suitable and understandable.

Third, resulting from second, arithmetic means with a .5 ending should be avoided in order to clearly select appropriate strategic guidelines and measures. Fourth, a condition based feedback should be enabled, so that possible risks for certain measures can be indicated. 5 OUTLOOK ON FURTHER USE OF THE MODEL The categories for the organisational implementation of the desired strategy can also be used for a direct reconciliation with customer requirements. According to Porter, real competitive advantages can only be achieved if the customer perceives an offered solution as superior to one offered by a competitor [9]. So far, the methodology has been based on internal strategic deliberations. In order to validate the derived strategic implications for organisational development, customer requirements ought to be reflected. Next, criteria for organisational performance of Integrated-Solution-Suppliers needs to be determined according to the findings of section 3. Focussing on the solution strategy of Individualisation and Tuli’s success factors [4], following criteria can exemplarily be derived: •

know-how availability across business especially service and production units

units,



degree of interaction across business especially service and production units

units,



role transparency and authorities of all parties, especially for customer experts



degree of documentation for solution development, especially process visualisation



degree of documentation of effective and ineffective experiences, esp. lessons learnt



degree of incentives across business units, especially sales, engineering and production



duration of customer especially with key-accounts



degree of goal commitment achieved by charismatic leadership

interaction/relationship,

• degree of leaders’ support for self-organisation. Alternatively, for the strategy of Standardisation, the criteria for organisational performance can be exemplarily defined respective the findings in section: •

amount of errors within order processing,



lead-time for order processing,



formalisation of order processing,



quality of processed order,



transparency of job descriptions,



transparency of organisation chart,



transparency of interfaces between divisions,



transparency of target agreements,

transparency of performance indicators and respective controlling. The perceived quality criteria can then be prioritised and weighted according to the importance of the requirements for the customer. The corporate skills of Integrated Solution Suppliers are also assessed regarding their degree of realisation in corporate activities. Here, following the chosen strategy, supported by the assessment tool, the criteria for individualised or standardised solutions are selected. Then, in expert workshops, with key-account managers e.g., the impact of corporate skills on the fulfilment of customer requirements in the sense of perceived service quality



143

can be assessed by correlations, similar to the Quality Function Deployment method [15]. From the resulting impact, visualised in a scatter plot with a portfolio, the experts can see which requirements are satisfied with solid competences and therefore pose a competitive advantage. This occurs when solutions of Integrated Suppliers are mature and also perceived as superior by the customer. Unstable customer advantages are achieved by requirements perceived as fulfilled by the customer but where the competences in the organisation have not reached a maturity to always guarantee this fulfilment. The solution here would be to establish needed competences by further organisational development and change management. Corporate advantages are achieved when competences are implemented in corporate activities, though are not of much help, if they are not seen as such in the eyes of the customer. Customer communication and marketing are needed in order to enhance these benefits for customer solutions. The least favourable case occurs when neither competences are up to scratch nor customers requirements are fulfilled and therefore mean a competitive disadvantage. In this case companies have two options: either neglect of this business field or massive build-up of the lacking corporate skills requiring professional change management. 6 SUMMARY In the introduction, the importance of enhanced product accompanying services was noted. In order to render Product-Service-Solutions successful, organisations need strategy conform organisational measures for the achievement of resource and competitive advantages. This is described by the Contingency and ResourceAdvantage Theory. The emphasis of this paper lies on the consistent orientation of organisational structure, processes and leadership. The determining factor for this is the identification of the Solution Strategy on the continuum between Standardisation and Individualisation, which is derived from Porters Competition Theory and adapted to the Customer Solution Process by Tuli. It must be noted, that Tuli’s approach lies closer to the Individualisation strategy and is closer to the common definition of Integrated-Product-Service-Suppliers as mentioned in section 1. In section 3, the main indicators for Individualisation and Standardisation are listed as well as the corresponding implications for structure, process and leadership style. In accordance with the Outpacing theory, these implications are not ‘black-and-white’ rules but are seen as organisational advice and companies can oscillate between them regarding which strategic orientation they currently need to follow. In section 4, requirements and specifications are derived in order to use the insights of the former sections for the conception and design of an assessment tool that enables ProductService-Suppliers to base their strategic decisions on appraisals and minimise risk of failure by indication of appropriate measures and the risks again therein. Further research will be conducted on the validation of the assessment tool to be developed. Also, the use of the developed criteria within a competitive advantage assessment including service quality dimensions remains to be implemented and validated.

144

7 ACKNOWLEDGEMENTS The support of the German Federal Ministry of Education and Research (BMBF) as well as the Project Management Agency – part of the German Aerospace Center (PT-DLR) is greatly acknowledged. We extend our thanks to our colleagues at Mercator School of Management, Department of Management & Marketing under Prof. Schmitz. 8 REFERENCES [1] Oliva, R., Kallenberg, R., 2003, Managing the transition from products to services, International Journal of Service Industry Management, Vol. 14, No.2, p. 160-172. [2] Wise, R., Baugartner, P., 1999, Go downstream: the new imperative in manufacturing, Harvard Business Review, Vol. 77, No. 5, p. 133-141. [3] Schmitt, R. et al., 2006, Service zweiter Klasse?, Qualität und Zuverlässigkeit, Vol. 51, No. 8. [4] Tuli, K. R. et al., 2007, Rethinking Customer Solutions: From Product Bundle to Relational Processes, Journal of Marketing, Vol. 71, No. 7, p. 1-17. [5] Galbraith, J. R., 1973, Designing Complex Organisations, Reading, MA: Addison Wesley. [6] Mintzberg, H., 1979, The Structuring of Organizations: A Synthesis of the Research, Englewood Cliffs, NJ: Prentice Hill [7] Hunt, S. D., Morgan, R. M., 1995, The Comparative Advantage Theory of Competition, Journal of Marketing, Vol. 59, No. 4, p. 1-15. [8] Neu, W. A., Brown, S., W., 2005, Forming Successful Business-to-Business Services in Goods-Dominant Firms, Journal of Service Research, Vol. 8, No. 1, p. 3-17. [9] Porter, M. E., 1981, The contributions of industrial organization to strategic management, Academy of Management Review, Vol. 6, No. 4, p. 609-620. [10] Gilbert, X., Strebel, P., 1987, Strategies to outpace the competition, Journal of Business Strategy, Vol. 8, No. 1, p. 28-36. [11] Galbraith, J. R., 2002, Organizing to Deliver Solutions, Organizational Dynamics, Vol. 31, No. 2, p. 194–207. [12] Rainfurth, C., Lay, G., 2005, Organisatorische Verankerung produktbegleitender Dienstleistungen in Industriefirmen, Zeitschrift für Führung und Organisation, Vol. 74, No. 2, p. 71–77. [13] Haque, B., Pawar, K. S., Barson, R. J., 2003, The application of business process modelling to organisational analysis of concurrent engineering environments, Technovation, No. 23, p. 147-162. [14] Judge, T., Piccolo, R. F., 2004, Transformational and Transactional leadership: A Meta-Analytic Test of Their Relative Validity, Journal of Applied psychology, American Psychological Association, pp, 755-768. [15] Schmitt, R., Hatfield, S., 2008, Strategic Servicification - A Quality based approach beyond Service-Engineering, Manufacturing Systems and st Technologies for the New Frontier, The 41 CIRP Conference on Manufacturing Systems, p. 511-514.

A Framework for Cross Disciplinary Efforts in Services Research P.J. Wild, 1, 2 P. J. Clarkson, 2 & D.C. McFarlane 1 1 Institute for Manufacturing, 2 Engineering Design Centre Department of Engineering, University of Cambridge, Cambridge [email protected] Abstract Interdisciplinary Services research programmes commonly entail an integrative goal, that is, to integrate theory & findings from the multiple disciplines involved. Services research has been frequently described as existing in silos, but little has been put forwards towards remedying this. This paper presents a framework for systematically relating different kinds of Services research by offering a set of sensitising concepts. Working from the view that services are consistently defined as activities, rather than objects or artefacts the concepts of the framework are drawn from Activity Modelling approaches, such as Task Analysis, Domain & Process modelling, & Soft Systems Methodology. Keywords: Services, Activities, Interdisciplinary / Integrative Research

1 INTRODUCTION Interdisciplinary Services research programmes commonly entail an integrative goal, − that is − to systematically integrate theory & findings from the multiple disciplines such as Management (e.g. Services Operations & Marketing, Organisational structures & transitions); Design, Manufacturing, Arts & Computing (User-Centred Design, Software as Service, & ServicesOriented Architectures) based disciplines. Services research has been frequently described as existing in silos [e.g., 1-3], but little has been put forward towards remedying this, beyond exhortations to work together. This paper presents a framework for systematically relating different kinds of Services research by offering a set of concepts & answers a frequent call within services discussion [e.g., 2, 3]. Working from the view that services are consistently defined as activities − rather than objects or artefacts − the concepts of the framework are drawn from Activity Modelling approaches, such as Task Analysis [4], Domain [5] & Process modelling [6], & Soft Systems Methodology [SSM, 7]. Ironically despite multiple assertions that services are activities, very little work has taken such an activity perspective on services [for exceptions see 8, 9]. By embracing elements such as: domain, value, values, actants, activities, goals, & environment; the elements of the framework can relate to a number of pertinent streams of Services research. This includes the IHIP debate in Marketing, the Service Dominant Logic (SDL), PSS, Functional Products, Design ‘Philosophies’ of Services, & technologies such as HUMS & SOA. This provides a set of concepts with which to relate systematically research themes & disciplines within interdisciplinary projects. This should help to enable building common ground between disciplines & researchers.

CIRP Design Conference 2009

145

1.1 Paper Overview Section 2 discusses pockets of services research. Section 3 discusses the nature of Integrative research. Section 4 presents an overview of the Activity Based Framework for Services (ABFS). Section 5 relates several strands of Services research to the framework’s elements, we include The IHIP characteristics; Service Blueprinting; the Service-Dominant Logic; & Product-Service Systems. Section 6 concludes the paper. 2 POCKETS OF SERVICE RESEARCH It is difficult to trace the growth in services because of differences in the ways that they are defined & reported over time & between nations. However, a figure that is often cited is that services account for 70-80% of Western economic activity. During recent years, a number of monikers have been put forwards for a shift to service as the focus of economic activity: Functional Economy; Servitization; Service systems; & the Service-Dominant Logic. The term Services embraces a number of different forms & contexts; from services that are intangible, heterogeneous, perishable, with a production process that is inseparable from use or consumption; via services on people such as medicine & education; through to maintenance procedures on hardware. A ‘radical’ position has emerged in the Service-Dominant Logic [10], where all marketable offerings, whether classed as goods or services, provide an element of service. Although this has been heavily contested & debated. For many years Services Marketing, taking its lead from earlier work in Economics, held that a number of ‘characteristics’ were enough to distinguish clearly products from services, namely Intangibility, Heterogeneity, Inseparability, & Perishability (IHIP). The IHIP characteristics partially enabled services marketing to ‘break away’ from mainstream marketing & fuelled a number of research streams. These included the refinement & debate of these ‘service defining characteristics;’ as well as other issues concerning the

effective marketing & execution of services (e.g. service quality, customer relationship management, measures of service quality, & human aspects of service execution). Several papers [11-13] have questioned IHIP as a foundation for Services Marketing. Some suggest refinements [14, 15] others alternative paradigms / logics such as Nonownership [12] or the Service-Dominant logic [10]. Despite this quite deep questioning papers are citing the IHIP characteristics as the key elements of service [e.g., 16-18], as opposed to four among many facets of relevance to Services? Design, Engineering & Manufacturing have also taken notice of the shift to a service economy. In the UK research programs such as IPAS, KIM, & S4T focus around some aspect of Service. A number of approaches have emerged that explicitly tackle the Design of services, or the Co-Design of products & services [19-21]. The most prominent approaches are Product Service Systems (PSS) [21] & Functional Products [20]. There are also number of ‘design philosophies’ for innovative & effective services design processes [e.g., 19, 22]. In computing, the emergence of Service-Oriented Architectures [23] & the crossover of ideas between Services & User Centred Design [19, 24, 25] provide insights into: how to support modularity / reconfigurability in processes & technologies; & how to approach Design of Services in a meaningful, repeatable & user-centred manner. We observe two relevant issues in services research: firstly, their activity based nature & secondly, the dispersion of relevant work in different fields. The first observation is that the activity-based nature of services implies a difference to the ‘tangible’ nature of products. Many authors emphasise activities in their definitions. For example, Hill [26] notes that “some change is brought about in the condition of some person or good, with the agreement of the person concerned or economic unit owning the good [p.318].” Despite this, outside of Service Blueprinting, & Tomiyama’s work [e.g., 9], the activity-based nature of Services has been neglected. However, since Service Blueprinting’s introduction by Shostack [8], few links are made with the activity-oriented research undertaken in various communities (e.g., Task Analysis [4], Process Modelling [6], Soft Systems Methodology [7]). Service Blueprinting has no explicit notion of goal, actor, & role, & has a relatively simplistic approach to temporal issues in service processes. Tomiyama [9] asserts “we still don’t know, for instance, how to describe service goals, how to decompose the total goal into subgoals, how to find a service mechanism to achieve a subgoal [p. 10],” seemingly unaware of work that considers these issues in depth [e.g., 4, 5, 27]. Secondly, Services research is dispersed amongst different communities. Chesbrough & Spohrer characterised the situation as “subfields … emerging in siloed academic areas such as management, engineering, & computer science schools, but precious few attempts have been made to integrate them [1, p. 36].” Disciplines involved in Services research appear to have their own dogma & mythology, & all too often end up with minimal interaction with other relevant areas. When they do, they take on work such as the IHIP characteristics that have been substantially critiqued. Laudable efforts such as the Emergence conference in 2006, & the IfM & IBM’s Service Science Management & Engineering efforts [3], have still not engaged with key communities.

We need of course to consider what integration & disciplines are, & we briefly consider these issues in section 3. 3 BEYOND DISCIPLINES: INTEGRATIVE RESEARCH Tress et al. [28] suggest three forms of cross-disciplinary research multi-, inter-, & trans-disciplinary. All three integrate across disciplines. Multi-disciplinary work uses a thematic umbrella to provide impetus to integration efforts; whilst inter- & trans-disciplinary programs make use of a common research goal. Transdisciplinary research is further distinguished from interdisciplinary work by the presence of non-academic participants in knowledge creation. Szostak [29] notes that the closest synonym for Integration is synthesis. This needs to be more than appropriation of methods or knowledge from another discipline. Integration involves critical reflection about: the strengths & weaknesses of different disciplinary insights; & biases inherent either in disciplinary practice or more general academic practice. Overall, Integration involves finding common ground among the different disciplinary insights. Following Tress et al. [28] & Szostak’s [29] definitions there are in effect, three kinds of synthesis around a theme, around a goal, & around a theme, goal & integration of knowledge for & from non-academic community members. If integrative research aims to produce new knowledge by combining knowledge & experience from different disciplines, we need to consider what a discipline is. Long & Dowell [30] note that “most definitions assume three primary characteristics of disciplines: knowledge; practice & a general problem [p. 11] going onto to formally define a discipline as “the use of knowledge to support practices seeking solutions to a general problem having a particular scope [30, p. 12, emp. added].” In Figure 1 we have two disciplines, each with their own general problem, scope knowledge, practices & solutions. Through combination, a new area can occur (e.g., Economic Psychology) & number of positive & negative outcomes can occur. These range from Appropriation, Cooperation (complementary goals), Collaboration (shared goals), through to negative interactions based around Lip Service, Misappropriation or Greedy reductionism [31]. Greedy reductionism is any form of misplaced reductionism. Whether this means boiling all things down to an inappropriate level, or assuming that all phenomena are instances of some other discipline (e.g., all sociology is psychology, all marketing is economics). Other issues are the varying disciplinary views on Pragmatism; & the distinction between producing exploitable research & being involved in the exploitation process [e.g., 7, 32].

Figure 1: Illustration of the nature & some of pitfalls of Integration When we consider how Integration relates to research projects & programmes, there is some prescriptive advice

146

for how to go about Integrative research [33, 34 35, 36]. These are generally manifest as a number of activities such as “Resolving disciplinary conflicts by working toward a common vocabulary [36]” or “Integrating the individual pieces to determine a pattern of mutual relatedness & relevancy [34].” One recurring piece of advice is to develop some form of framework to integrate different strands of the research. Such a framework anchors other integrative research activities, such as communication & glossary development; through to deeper consideration of ontological (what exists in the domain) & epistemological (how we know what we know about the domain) issues. The nature of such a framework is our next concern. 3.1 What is meant by a Framework? A framework can be seen to be a general set of concepts for understanding a research area. It is not tightly organised enough to be a predictive theory. It aims to sketch out the general concepts of a field of enquiry & the possible relationships between them. In contrast, a theory is an intellectual structure that embodies enough detail to make a prediction about a domain. A model is the application of a theory to specific phenomena. An aim of theory is to be general across many phenomena. A theoretical approach would aim to generate a number of models to account for different instances of phenomenon. The framework is aimed at general understanding & communication rather than being an unequivocal verified or validated description of the world. Rather than stating that Services are activities, we state that we view services as activities. At this stage in its development, the Framework is put forwards to think about the Services domain, not to be a full & complete ontology of Service Systems. We recognise that the boundaries between the entities of the framework are negotiated & open to change. Rather than being an absolute classification of elements of a Service System. 4 AN ACTIVITY FRAMEWORK FOR SERVICES In this section, we sketch the Activity Based Framework for Services (ABFS) for the consideration of Services research across disciplines. As noted, the ABFS developed from the observation that Services are defined as activities. We use the term activity to refer to a number of approaches that take their concern with action, for example: Task Analysis [4], Process Modelling [6], Domain Modelling [5] & SSM [7]. The ABFS’s elements are a composite from different activity-oriented approaches. There is no unified method whose modelling concepts match the ABFS’s elements. Each activity modelling approach embodies differing firstclass modelling concepts. In the introduction, we noted that services are activities not objects or artefacts. Most activity approaches consider some aspect of objects or artefacts. In the former case, the objects are acted upon or reacted to; in the latter as entities that support both mundane (e.g., pen & paper) & novel tasks (collaborative web based writing). In addition, as a service system, there is interdependence between different service elements. Goals, for example, are scoped to what can be done in the service domain activities achieve these goals. Concepts within activity modelling such as task, roles, & domain-object represent much of interest & relevance to those wishing to relate different strands of Services. For example, the debate on the IHIP characteristics [12] & Intangibles [14] relate to domain concepts; the works on types & models of service activities relate to activity descriptions & goal models.

147

4.1 Overview of the ABFS The ABFS is represented schematically in Figure 2. Service activities are assumed to be carried out within a Service System. The service system embraces the objects (both informational & physical); the goals & values held by various individual & collective Actants. Service activities are carried out by people & artefacts to affect the objects in a domain. A service system can be considered to have a variety of effectiveness measures, depending on the value (i.e., benefit) sought, & this is evaluated when the quality goals are balanced against the resource costs. A service system has an environment, which has social, cultural, political physical dimensions. Borrowing from Dowell & Long [5], we suggest that actants & artefacts have Structures & Behaviours. The costs of setting up & maintaining these structures & behaviours are evaluated in Service Effectiveness assessments. 4.2 The main components of the Framework The following lists the main components of the Framework. Service Domain Is the ‘world’ whose possibilities & constraints are organised in relation to specific goals. Objects are the elemental constituents of a domain, such as heads, & hairstyles or engines & aeroplanes. Objects can be abstract / intangible (i.e. informational) or concrete / tangible (i.e. physical) & can emerge at different levels of analysis. Objects can also exist within actants & artefacts. In some contexts, a person’s knowledge can be considered the domain, as can a collective agent’s shared understanding. Domains such as the internet & 3D graphics exist almost solely within artefacts, though, on occasion, a physical representation can be made to paper & other such media. Domains ‘overlap’ between different Service Systems, & could be conceptualised as subdomains or overlapping with other domains. Therefore, an Aeroengine can be a domain in itself or can be a subdomain for aeroplanes & pilots; which in turn are sub domains of air-traffic control activities / services.

Figure 2: Schematic of an Activity Based Framework for Services Service Goals Are the specification of needed changes to service domain objects [5] & shape both the activities carried out & reflect value, values, & effectiveness measures. They are carried out by actants, & sophisticated artefacts but can only be ‘held’ by actants. Classically, they are represented hierarchically [4, 5], each goal mapping to a finer description of change or maintenance of a domain object. As service objects can be either concrete or

abstract, goals also vary in their concreteness or abstractness. Other work suggests that goals can be heterarchical [4], being embedded in a complex of higher& lower-level goals [27]. Higher-level goals can reflect qualities such as those relating to user experience; or a general goal that applies to all activities, reflecting general values held towards the world (e.g., cost efficiency or minimisation of ecological-impact). There is a relationship between these higher-level goals & the value & expected values that actants have. Costs can be accounted for through financial measures, as well as others such as fatigue, social discord, or depletion of natural resources. Services Activities The sequence & type of actions (physical / non physical) carried out in order to achieve goals. Activities are concerned with changing (education, surgery, technology, upgrades) or maintaining (preventative healthcare, hardware maintenance) the states of service domain object attributes [5], when they are carried out they should achieve all, or part of, a service goal. There are multiple classifications of service activities focussing around organisational division [37]; Economic Relationships [14, 26, 38]; the relationship to a product [39] or a more ‘general’ consideration of their nature [40, 41]. Service Actants (Individual & Collective) Actants [see 42] are those entities capable of carrying out activities & the term covers people, & groups of people. Work within services marketing has often stressed that service providers & consumers are inseparable [e.g., 43] leading some to argue that services are always cocreated between service producer & consumer [10, 44]. However, this says little about the nature of those roles or the other actors acting as stakeholders in service performance. Nor does it allow us to consider the strength & weakness of co-creation on owned service domains, rather than personal domains. Typically, personal maintenance services (e.g., haircuts & surgery) need the co-location of the consumer & producer. Services on owned property often have less need for co-location, but require ‘set up’ activities in order to hand over the product to be serviced. Co-creation also says little about the role of artefacts & technologies. Whilst education has traditionally been co-located, technologies such as books, distance learning & teleconferencing can break the need for co-location, allowing development of knowledge in a non co-located asynchronous manner. The service domain remains the same (the learner’s mind, knowledge, & skills); the means of changing it differ. Another aspect of actants is their ownership relationship with other aspects of a Service System. Work within Services marketing has stated that services do not result in any change of ownership of an entity [45, 46]. This has been generalised in approaches such as the ServiceDominant logic & Nonownership paradigms. However, many services facilitate the movement & exchange of artefacts, there is danger in over generalising. Service Artefacts & Technologies Are the artefacts, tools, & technologies used to carry out services. Artefacts is a term that suggests both artificiality & making [47]. Artefacts are created entities that serve some purpose in activities; they rely on some form of technology for their creation. Because they are created for a purpose, rather than by random or natural processes, it is tempting to classify them according to a single purpose. Example purposes would include informational (e.g., books & databases); physical (e.g., tools such as spades & bulldozers); aesthetic (art); recreational (e.g., sports

equipment); financial (bank accounts & credit cards). However, artefacts are multifaceted having a number of purposes. These purposes can be aligned or in competition with each other. They can also change over time. Archaeological artefacts may have had an original purpose, but after years in the ground or under the sea, serve to give insight into ancient life & culture. Even art works, viewed by many as serving a pure & aesthetic purpose can serve financial (investment), informational (propaganda), or nationalistic purposes. Service Effectiveness At this level of description, more traditional quality & effectiveness concerns would be expressed. Dowell & Long [5] see effectiveness as a function of task quality (the quality of a ‘output’ created by a Service System) & resource costs (the costs to participants of establishing structures & producing behaviours). Desired effectiveness is set by specifying desired task quality & the desired resource costs. Similarly, actual performance can be measured as a function of the actual task quality & the actual resource costs [5 p. 139]. Long & Dowell considered resources as being cognitive, conative, or affective, however, the notion generalises to other processes & entities such as organisational structure & responsiveness. Service Values Are the criteria with which judgements are made about other entities [7]. As such, they affect how actants view the effectiveness of a service system element (e.g., goal satisfaction, effectiveness). There is a tendency for some individuals & groups to see values as true & objective, those who do not share your values are classed as having none. Witness the view by many in environmental movement that short-term non-resource renewing profit driven enterprises have no values. Such enterprises have values, but they are focussed on judging things by profitability, capital liquidity, rather than sustainability. Values are key aspect of scoping the other elements of a Service System. So the form of the Service System can reflect the Values held by its actants. Whilst the goal of restaurants is to provide food & generate profit; the values held by the owners, staff & patrons can drive radically different manifestations of eating location, menu & experience. Comparing a high-class joint with a roadside catering outlet without reference to Value systems would be meaningless. Service Environment The environment often becomes a catch-all in activitybased approaches, covering all the things that are not first class entities in their modelling worldview. The reduction of impact on the environment has been a motivating factor in some Services research [21], & discussed as a side effect in others. It is important to formally represent not just the immediate environment of interest to services (i.e. the service domain), but the wider environment within which activities are carried out. Here we depart from the main theme of the paper, & find that activity approaches can benefit from environmental concepts within Services Research [e.g., 48]. Service Structures & Behaviours Structures provide capabilities (e.g., knowledge, skills, information) in reference to a domain, whilst 'behaviours' are the activation of these structures to perform tasks. In Dowell & Long’s [5] work, Structures & Behaviours applied to humans & computers. Human behaviours are considered purpose-driven; in contrast, artefacts are created to serve these purposes. Structures are physical (e.g., electronic, neural, bio-mechanical & physiological)

148

or abstract (e.g., software or cognitive or social schemes & processes). Behaviours may be physical, such as printing to paper or selecting a menu, or abstract, such as deciding which document to open, or problem solving. However, the concept can generalise to other aspects of a service system including the environment & thus we include social, cultural & physical Structures & Behaviours [48-50]. Other aspects of the ABFS Relationships Are numerous & not yet well enough explicated to make predictions. However, as general guide, key relationships are those between: •

goals & the actants who hold them;



values & the actants who hold them;



goals & domain objects;



activities & the actants / artefacts that carry them out;



effectiveness measures;



quality measures & values;



actants & roles;



actants & grouping of actants;



objects & the activities that maintain & change them;

&

the

resource

costs

&

quality



values & the goals / activity structures that achieve or are influenced by them; Boundaries Are socio-technical [51], sometimes reflecting the physical composition of the world, at others times, decisions about an entities’ scope. An example of the latter is boundary between the domain & artefacts & technologies. In some cases, the artefact may contain − in whole or part − the domain in question (e.g., 3D graphics or email). However, because a domain is enabled by an artefact does not mean it can be ‘reduced’ to the artefact. Film & television were in some way enabled by camera, television & projector technologies, but can now be viewed through computers, & the sophisticated nature of both home & cinema based systems blurs the distinctions between things. The notion of centrality of purpose may enable activity modellers to scope the roles of the domain & artefacts. Therefore, Engine Health monitoring technologies may help to support tasks beyond the main function of thrust, power & lift for an engine; even though they are embedded & considered in other perspectives as part of the engine. Information & Knowledge Information & Knowledge are explicitly represented through the specification of intangible domain objects [see 14] & implicitly specified through Structures & Behaviours [5]. In the case of the former, information & knowledge can be originals [14], that is additions to knowledge; or intangible products, that is, information-based products. In the latter case, information & knowledge about the operation of a domain can be formally represented as part of the domain, for example the message of a communication, or informally as part of the structures & behaviours of Service System actants. 5 ILLUSTRATION OF THE ABFS’S COVERAGE & USE Within this section, we recount & compare the elements of the ABFS number of approaches with services research. Because of space restrictions, we will restrict our consideration to The IHIP paradigm; Service Blueprinting; the Service-Dominant Logic; & Product-Service Systems.

149

One concern is to see whether the approaches can be classed as a framework for all Services Research, by considering the overall scope of their components against the ABFS. 5.1 The Service Dominant Logic Vargo & Lusch [10] are concerned with providing a service-based worldview on Marketing & Economics. In this revised worldview, goods & services are subsumed in relation to the Service-Dominant Logic (SDL). Vargo & Lusch define service as “the application of specialized competences (knowledge & skills), through deeds, processes, & performances for the benefit of another entity or the entity itself [10, p.2]. Within the SDL value, rather than being embedded in goods / products by manufacturers, is co-created between supplier & consumer during service activities; although value is never explicitly defined in their work. The SDL builds on two types of resources. Operand resources: are resources upon which an operation or act is performed to produce an effect, & in Vargo & Lusch’s view, have been dominant in human history, that is the goods based view. Operant resources: are employed to act upon operand resources, & concern issues such as knowledge & skills. Noting that they are “likely to be dynamic & infinite & not static … they enable humans to multiply the value of natural resources & to create additional operant resources [p. 3].” Operand & Operant mirror a common distinction between concrete & abstract entities & activities [5, 14, 40]. Operant resources are the province of the physical concrete aspects of the domain & environment. Whilst Operand map to the structures & behaviours or actants, or programmed / embodied in artefacts. Activities

--

Goals

--

Actants

Considered as co-creators of value

Artefacts & Technologies

--

Domain Structures & Behaviours

Operand & Operant resources.

Effectiveness

--

Value

Undefined, despite using the term 122 times in their paper [10].

Values

--

Environment

--

Table 1:- Comparing the SDL against the elements of the ABFS Compared against the ABFS, we can see that the SDL is limited in its scope as an organising framework for services research. 5.2 Service Blueprinting In Services Marketing, the most refined approach to modelling service activities is Service Blueprinting [8] an approach that has evolved considerably since its introduction by Shostack. It has also become widely known outside of Services Marketing, having been applied in Design, Engineering, Healthcare, & Tourism. It would be fair to say that other approaches such as Servicescapes, Service Encounters, & Moments-of Truth owe a lot to Shostack’s original work. Table 2 compares Service Blueprinting against the ABFS.

Activities

Explicitly represented

Goals

Implicit

Actants

Implied by lines of visibility

Artefacts & Technologies

Can be partially covered by tangibles

Domain

Can be partially covered by tangible

Structures & Behaviours

--

Effectiveness

Time-oriented

Value

--

Values

(Speed, cost)

Environment

--

Table 2:- Comparing Blueprinting against the elements of the ABFS Shostack’s work was motivated to identify & represent service functions; benefits; as well as standards & tolerances. One of the original strengths of Service Blueprinting was its focus on service as activities. Service Blueprinting has evolved since its introduction, embracing more lines of visbility, & an iterative cycle of development. However, the approach misses other elements of activity models, such as those outlined in section 3. 5.3 The IHIP Characteristics (and Debate) Rathmell’s [45] paper made several observations were made about services three of which were that Services are Imperishable, Intangible & having Imprecise standards. In combination with Inseparability, the observations evolved to be the four IHIP characteristics, that is, services are Intangible, Heterogeneous, Inseparable, & Perishable. These have been defined as “Intangibility− lacking the palpable or tactile quality of goods. Heterogeneity − the relative inability to standardize the output of services in comparison to goods. Inseparability of production & consumption − the simultaneous nature of service production & consumption compared with the sequential nature of production, purchase, & consumption that characterizes physical products. Perishability − the relative inability to inventory services as compared to goods [13, p. 326]. A review by Zeithaml et al. [43] helped to solidify the perception that these four characteristics, define services & fully distinguish them from goods / products. Since then the IHIP characteristics have become widely cited as the key characteristics of services. Lovelock & Gummesson [12, p. 21-2] noted a textbook consensus about the IHIP characteristics arguing that they can be described as a paradigm for Services Marketing. The IHIP characteristics are often uncritically accepted outside services marketing as the defining features of Service [e.g., 16-18]. Hill’s [14] concerns remain highly relevant. Hill [14] argued for a retention of a distinction between Services & Goods. However, the dyad needed refining into a triad, with Intangible goods being added to the Tangible goods. Hill notes that intangibles are ‘Originals’, “additions to knowledge & new information of all kinds, & also new creations of an artistic or literary nature [14, p.438].” Examples of Originals include books, music compositions, films, processes, plans, blueprints & computer programs. Intangible goods generally need a physical manifestation. In an Economic context, they can be broken into Producer & Consumer durables. Some artefacts, such as software, act as one or the other in organisational or personal contexts. Combined with Lovelock’s observation that

Services can be focused, we gain a fourfold division, Intangible & Intangible goods, & tangibly & intangibly focussed services & activities. This demonstrates one set of relationships between Intangible & Intangible goods, & Tangible & Intangible focussed services & activities. Rather than misclassifying intangibles as services, a refined understanding is gained. Table 3 represents a 2 by 2 matrix of products & services that vary in their tangibility. Tangible Product Intangible product Cleaning & Creation of (S) MP3 upgrading memory Music recording (TS) MP3 player (Original (TP) represented as IP) Intangible Download (IS) of Support for finding Service MP3 file to MP3 & purchasing (IS) player (TP) music (IP) Table 3:- Matrix of Intangible products & services The notion of a Service Domain relates most readily to goods, just as domains can be composed of a range of concrete (tangible) & abstract (intangible) products vary in their tangibility. By considering objects (and activities) as varying in concreteness or abstractness, we gain valuable insight into the debates on the properties of products & service. Table 4 compares the refined understanding of the IHIP characteristics derived from synthesising the works of Hill [14], Lovelock [40] & others [43, 45, 52]. Tangible Service

Activities

Explicitly represented by tangible & intangible activities [40].

Goals

--

Actants

Implied by inseparability [43]

Artefacts & Technologies

--

Domain

Explicitly represented as tangible & intangible domain objects [14]

Structures & Behaviours

--

Effectiveness

--

Value

--

Values

--

Environment

--

Table 4:- Comparing IHIP characteristics against the elements of the ABFS 5.4 Product-Service Systems (PSS) The term PSS was coined by Goedkoop et al. [53] who defined a PSS as “a marketable set of products & services capable of jointly fulfilling a user’s need. The PS system is provided by either a single company or by an alliance of companies. It can enclose products (or just one) plus additional services. It can enclose a service plus an additional product. & product & service can be equally important for the function fulfilment [p.18].” A number of papers have been keen to stress that PSS are combinations of numerous entities such as people, computers organisations, not just products & services. A key aspect of the approach is that a focus on development & exchange of Service − rather than products − is means of achieving greater sustainability. Table 5 compares the elements of the ABFS against PSS.

150

Activities

Some consideration of activity types [e.g. 39]

Goals

--

Actants

--

Artefacts & Technologies

Some

Domain

Implicitly through consideration of products

Structures & Behaviours

--

Effectiveness

--

Value

--

Values

--

Environment

A consistent theme within the literature on PSS [21], but the environment is rarely conceptualised or described in any depth.

Table 5:- Comparing PSS against the ABFS PSS, despite its purported systemic nature, has given little in depth consideration of wider elements of a service system & how they might interact. “How is the environment conceptualised?” “How could a focus of service benefit & / or reduce the impact on the environment?” are relevant & pressing questions, but something deeper is needed. We can find this in other work on the Functional Economy [48] Natural Capital [54] & Ecotoxicology [55]. 6 SUMMARY & CONCLUSIONS This paper has had two aims; the first is to introduce the ABFS as a candidate integrative framework for Services research projects. The second has been to relate this framework to a number of Services research strands across disciplines. As a sensitising framework, it is not meant to have ontological status; rather it aims to systematically relate strands of Services Research together. We have related the framework to four strands of Services Research, namely Service Blueprinting; the Service-Dominant Logic; the IHIP characteristics & debate; & Product-Service Systems. Figure 3 illustrates at a gross level of how each aspect of the framework relates to several strands of service research. The scope is rated as being low, medium, or high. It is clear that the profile of each approach is different. For example, the SDL is heavily focussed around Value & Actants, as its concern is with the cocreation of value. The headings refer to: the IHIP debate [11-14, 26]’; Service Blueprinting [e.g., 8]; Nelson’s consideration of service values [17, 22]; the Nonownership paradigm [12]; the Service Dominant Logic [10]; PSS [e.g., 21]; Functional Products [20]; & the Functional Economy [48]. No current approach considers all the elements of a service system. Thus, the framework has shown where there are gaps in disciplines & perspectives. We can conclude that no single discipline involved in services research has a pre-existing framework suitable for integration of the knowledge, practices, solutions, general problems & scopes, across the many disciplines involved in Services Research.

151

Figure 3: Scope of Various Service Research Strands against the framework 7 ACKNOWLEDGMENTS This work has been supported in part from the TSB / IPAS project; the EPSRC EDC block grant; & the EPSRC/BAES S4T programme. 8 REFERENCES [1.] Chesbrough, H. & J. Spohrer, A research manifesto for services science. CACM, 2006. 49(7): p. 35-40. [2.] Fisk, R., Service Science, the Elephant & the Blind Men, Who's Who? 2008, University of Strathclyde Glasgow. [3.] IfM & IBM, Succeeding through Service Innovation. 2008, Cambridge: IfM, University of Cambridge [4.] Diaper, D., Understanding Task Analysis for HumanComputer Interaction, in The Handbook of Task Analysis for HCI, D. Diaper & N.A. Stanton, Editors. 2004, LEA: Mahwah. p. 5-47. [5.] Dowell, J. & J.B. Long, Conception of the Cognitive Engineering Design Problem. Ergonomics, 1998. 41(2): p. 126-139. [6.] Melão, N. & M. Pidd, A conceptual framework for understanding business processes & business process modelling. Information Systems Journal, 2000. 10(2): p. 105-129. [7.] Checkland, P.B. & J. Poulter, Learning for Action. 2006, Chichester Wiley. [8.] Shostack, G.L., How to Design a Service. European Journal of Marketing, 1982. 16(1). [9.] Tomiyama, T. A Design Methodology of Services. in 15th ICED 2005. Melbourne. [10.] Vargo, S.L. & R.F. Lusch, Evolving to a new dominant logic for marketing. Journal of Marketing, 2004. 68(1): p. 1-17. [11.] Wyckham, R.G., P.T. Fitzroy, & G.D. Mandry, Marketing of Services. European Journal of Marketing, 1975. 9(1): p. 59. [12.] Lovelock, C.H. & E. Gummesson, Whither Services Marketing? Journal of Service Research, 2004. 7(1): p. 20-41. [13.] Vargo, S.L. & R.F. Lusch, The four service marketing myths. Journal of Service Research, 2004. 6(4): p. 324-35. [14.] Hill, P., Tangibles, intangibles & services. Canadian Journal of Economics, 1999. 32(2): p. 426. [15.] Wild, P.J., et al. Towards a Framework for Profiling of Products & Services. in 5th ICMR. 2007. Leicester: Inderscience.

[16.] Candi, M., The role of design in the development of technology-based services. Design Studies, 2007. 28(6): p. 559-583. [17.] Kundu, S., et al., Implications for Engineering Information Systems Design in the Product-service Paradigm, in Advances in Life Cycle Engineering for Sustainable Manufacturing Businesses, S. Takata & Y. Umeda, Editors. 2007, Springer London. [18.] Baltacioglu, T., et al., A New Framework for Service Supply Chains. The Service Industries Journal, 2007. 27(2): p. 105-124. [19.] Parker, S. & J. Heapy, The Journey to the Interface. 2006, DEMOS: London. [20.] Alonso-Rasgado, M.T. & G. Thompson, A Rapid Design Process for Total Care Product Creation. Journal of Engineering Design, 2007. 17(6): p. 509531. [21.] Baines, T., et al., State-of-the-art in Product ServiceSystems. Proceedings of the I MECH E Part B Journal of Engineering Manufacture, 2007. 221(10): p. 1543-1552. [22.] Nelson, H.G., Systems Science in Service to Humanity. Systems Research & Behavioral Science, 2002. 19(5): p. 407-416. [23.] Papazoglou, M. & W.-J. van den Heuvel, Service oriented architectures. The VLDB Journal, 2007. 16(3): p. 389-415. [24.] Wild, P.J., HCI & the Analysis, Design, & Evaluation of Services, in HCI 2008, O. Abuelmaatti & D. England, Editors. 2008: Liverpool p. 207-208. [25.] van Dijk, G., S. Minocha, & A. Laing, Consumers, channels & communication: Online & offline communication in service consumption. Interacting with Computers, 2007. 19(1): p. 7-19. [26.] Hill, P., On Goods & Services. Review of Income & Wealth, 1977. 23(4): p. 315-338. [27.] Wild, P.J., P. Johnson, & H. Johnson. Towards a Composite Model for Multitasking. in TAMODIA 2004. 2004. Prague: ACM Press. [28.] Tress, G., B. Tress, & G. Fry, Clarifying Integrative Research Concepts in Landscape Ecology. Landscape Ecology, 2005. 20(4): p. 479-493. [29.] Szostak, R., How & Why to Teach Interdisciplinary Research Practice. Journal of Research Practice, 2007. 3(2). [30.] Long, J.B. & J. Dowell. Conceptions of the Discipline of HCI. in HCI'89. 1989. Nottingham: CUP. [31.] Dennett, D., Darwin's Dangerous Idea: Evolution & the Meaning of Life. 1995, London: Penguin. [32.] Røyrvik, J. & E. Bjørnsen. Experiencing Engineering Design. in Design 2006: 9th International Design Conference. 2006. Dubrovnik. [33.] Szostak, R., How to do interdisciplinarity. Integrative Studies, 2002. 20(103-122). [34.] Klein, J., Interdisciplinarity: History, theory, & practice. 1990, Detroit: Wayne State University Press. [35.] Tress, B., G. Tress, & G. Fry, Ten steps to success in integrative research projects, in From Landscape

[36.] [37.]

[38.]

[39.]

[40.]

[41.]

[42.]

[43.]

[44.]

[45.] [46.] [47.] [48.]

[49.] [50.]

[51.] [52.]

[53.]

[54.]

[55.]

152

Research to Landscape Planning, B. Tress, et al., Editors. 2005, Springer-Verlag: Dordrecht. p. 241257. Newell, W.H., A Theory of Interdisciplinary Studies. Integrative Studies, 2001. 19(1-25). Shostack, G.L., Breaking Free from Product Marketing. Journal of Marketing, 1977. 41(2): p. 7380. Hill, P. & R. Youngman, The Measurement of Intangibles in Macroeconomic Statistics, in PRISM. 2002, City University Business School: London. Tukker, A., Eight types of product-service system: eight ways to sustainability? Experiences from SusProNet. Business Strategy & the Environment, 2004. 13(4): p. 246-260. Lovelock, C.H., Classifying Services to Gain Strategic Marketing Insights. Journal of Marketing, 1983. 7(Summer): p. 9-20. Berry, L.L. & S.K. Lampo, Teaching an Old Service New Tricks: The Promise of Service Redesign. Journal of Service Research, 2000. 2(3): p. 265-275. Latour, B., Science in Action: How to Follow Scientists & Engineers. 1987, Cambridge, MA.: Harvard University Press. Zeithaml, V.A., A. Parasuraman, & L.L. Berry, Problems & Strategies in Services Marketing. Journal of Marketing, 1985. 49(2): p. 33-46. Spohrer, J., P.P. Maglio, J. Bailey, & D. Gruhl, Steps Toward a Science of Service Systems. Computer, 2007. 40(1): p. 71-77. Rathmell, J.M., What Is Meant by Services? Journal of Marketing, 1966. 30(4): p. 32-36. Judd, R.C., The Case for Redefining Services. Journal of Marketing, 1964. 28(1): p. 58-59. Crowe, M., R. Beeby, & J. Gammack, Constructing Systems & Information. 1996, London: McGraw-Hill. Stahel, W.R., The Functional Economy: Cultural & Organizational Change. Science & Public Policy, 1986. 13(4): p. 121-130. Hall, E.T., The Silent Language. 1959, New York: Doubleday. Elster, J., Explaining Social Behavior: More Nuts & Bolts for the Social Sciences. 2007, Cambridge: Cambridge University Press. Mingers, J., Realising Systems Thinking. 2006, New York: Springer-Verlag. Fisk, R.P., S.W. Brown, & M.J. Bitner, Tracking the Evolution of the Services Marketing Literature. Journal of Retailing, 1993. 69(1): p. 61-103. Goedkoop, M., C. van Halen, H. te Riele, & P. Rommens, Product Service Systems, Ecological & Economic Basics. 1999, PRé Consultants: Amersfoort · The Netherlands. p. 132. Costanza, R., et al., The value of the world's ecosystem services & natural capital. Nature, 1997. 387(6630): p. 253-260. Cairns, J., Ecotoxicology & sustainable use of the planet. Toxicology & Industrial Health, 2002. 18(4): p. 162-170.

New Models For Sustainable Fashion Industry System: A Case Study About Fashion Net Factories P. Ranzo, M. A. Sbordone, R. Veneziano Department IDEAS Industrial Design Environment and History, Second University of Naples Italy, Faculty of Architecture, Monastero di S. Lorenzo ad Septimum - via S. Lorenzo - 81031 Aversa (CE) Italy [email protected], [email protected], [email protected]

Abstract The paper aims to describe a design methodology in design strategy and design services sectors, it highlights the importance of designing new models of production and distribution in the fashion driven sector. In complex industrial systems you have passed from a “product oriented” concept to a “costumer oriented” one. This strategy focuses on customization of the production and distribution mechanisms that increase globally competitiveness. In the "network society" defined as a social network that spreads through the network logic and which is powered by information technology, it emerges a new form of communication based on horizontal networks of communication. Keywords: Sustainable productive system, Network society, Virtual and creative communities.

1 INTRODUCTION The topic of this paper relates to remarks upon the impact that the new economy has on production systems which are linked to the fashion system, especially to the creation of use and exchange values. The fashion system is mainly based on creativity supported by continuously renewing contents. Peter Drücker defined these products as “knowledge based products”. He outlined a production system characterized by a “cultural” industrial configuration. According to Drücker, production linked to fashion can be defined as ”highly cultural”. As a matter of fact, its main value depends on design and creative support. [1] In the old economy, the industry’s value lied in the manufacturing industry, leaving out the contribution from a different kind of knowledge that comes from other fields. Very few resources were dedicated to research, experimentation, design and creativity. Moreover, knowledge and manufacturing industry were strictly separated. Nowadays, the challenge lies in acknowledging the economic value of creativity and services in order to boost their development by integrating them in the production processes. Referring to the “design oriented” Italian industry, a metaindustry is often mentioned. It is an industry in which the productivity is directly proportional to the ability of elaborating knowledge and turning it into Capital. In order to make our factories competitive, it is necessary to consider the ways in which they can incorporate culture in our products and integrate as much as possible the competences in the creative process to produce a real innovation. You must reconsider the concept of innovation by connecting non technological elements of the process to it. The creation of the immaterial value takes place in “a complex combination in which the technological innovation plays a key role, but its representation, as well as

CIRP IPS2 Conference 2009

153

information sharing, ideas, senses and symbols are important too”. The relevance of design and research in creating worth and value in the economy of Made in Italy products, as well as the integration of knowledge in the product and the connection between services, manufacturing industry and the entire creative laboratory, are at the core of the whole process. In the industrial model of the new economy, an integration takes place. In it “the manufacturing basis still plays a key role, but the creation of the value and the inner nature of the products depend on the immaterial knowledge offered by the specialised services”. [2] Some aspects of consumer dynamics, like customerization, need special co-design methodologies that deal with customers. Generally speaking, the co-construction of productive sectors which relates to fashion requires considering the professional figures and the opportunity that young creativity is given. In the current industrial configuration, goods production is carried out where services and industry meet. For example, the production branches connected to made in Italy products are based on a system in which leader factories are mainly set in Northern and Central Italy, where there already is an appropriate system of infrastructures and services. In Southern Italy, instead, small factories work especially as sub-suppliers and represent the weak link of a system that produces in the global market. The automatic work without any kind of creativity (wellfinished manufacturing) that includes the final packaging of the product (mostly based on labour rather than design) is set outside Italy. Nevertheless, there is some kind of countertrend, especially in fashion industry, where production excellence is necessary to a luxury oriented system. Italian factories that operates in this system, find offering totally made in Italy products (whose productive process is traceable and certified) strategically convenient.

There are more and more initiatives in this field, which include also the problems of certifying the origin of products. Traditional productive organizations like industrial districts, that nonetheless represent one of the main characteristics of Italian productive system, are not the solution to this problem. Industrial districts evolved into Factories nets which are the best way of replying to the globalized economy. By using information and communication technologies, the competitiveness of Italian territorial systems grows. Factories nets, especially those with “multiple gravitational centres” – large factories and lots of small ad medium ones – may, depending on the market, merge in “centres” that act as the engine of the whole system. The main problem of competitiveness is coordinating all factories “constellations”. Generally in vertical factories nets, this task is carried out by the main factory, which acts a catalyst. In more complex situations, when you need to act strategically in relation to the system Country, you need a recognised excellence centre which can guide the innovation and transformation progress of the whole system and gather all figures in a shared vision. This excellence centre can acts as a place where creativity, research, excellence of production and new technologies can serve the real innovation process of the factories. Innovation should not only consist of transferring knowhows to the factories, but also sharing new ideas to keep up to the external changes. 2

THE EVOLUTION OF THE PRODUCTIVE FASHION INDUSTRY

2.1 The evolution of the industrial districts and the role of Design in the Fashion industry In the Italian scientific debate industrial districts are the subject of remarks upon the nature of the productive model and its ability of adapting to the ever-changing market. The likely decline of the districts, the continuous rise of new industrial groups, the establishment of new characters call for new innovative development strategies. The rise of competitive pressure (relating to placement in global markets) must be fought by a better integration of “knowledge” and “actors” of the scientific and economic world. If the industry can be compared with a “connective texture of cells and technology and innovation provide it with filling”, the productive activities based on technological innovation, which can not be easily reproduced, can revitalize the economic scene. It is worth thinking of the evolution of the Italian production system in the last years, over which the district models have played a key role. [3] After a growing phase, an assessment one followed to which the districts reacted in different ways according to their own characteristics. Today there’s a more complex condition. As a result, the birth of so called “technological districts”, founded on the development of specific branches, requires political strategies which have already been tested in international contexts. It is necessary to use competitive strategies and act in a highly innovative way to help develop a new cooperation between prominent figures of the world of research and production.

In this context, a remark on design and creativity as competitive tools and opportunities to boost industrial districts is strictly necessary. The associative model of the industrial district has led to higher competitive advantages compared to the ones achieved by the single units which it is made of, and today the districts system is the most vital part of the Italian productive system. The close link with the territory, the territorial concentration of the productions, the competences derived from an “excellence know how” start the engine of a machine which takes advantage of the complex system of connections between productive units. The competitive advantage of this model depends on both the nature of the organization model, which has technically highly qualified competences, and the versatility and adaptation that enable it to reply effectively to the requests and changes of the market. This model has been further developing over the last years, moving from a first generation industrial district to a second generation one or meta-district. It has been trying to overcome the problems derived from the so called “short nets”, a system of relations based only on local areas. The establishment of the second generation of industrial district model characterises the passage from “the local dimension to the global one”. [4] It widens the global idea and fosters the interaction with the outside. The current direction of the development leads towards the birth of new forms of diversified specializations. They aim to resettle a balance between competitiveness actions, territorial synergies and new cooperative forms in highly technological branches. The industrial districts, which do not substitute or overlap the classic ones, open up to specialized areas and global relations. The industrial district, defined as a “socio-territorial entity, characterised by the active presence, in a environmentally and historically defined territorial area, of a group of people and industrial enterprises”, has an explorative and design aptitude which enables very complex innovation processes to take place. [5] The model is founded on systems of connections between competitive small and medium enterprises in a narrow territorial sphere. It can boost new synergies between independent and widespread productive units. The district is an open and flexible form of collaboration, where the units, which exchange explicit (codified) and implicit (contextual) knowledge, are play an active part in a circuit in which the local and global cognitive spheres integrates. Several studies on district economy show that the most effective districts in terms of innovation move towards three guide lines: the promotion of high quality products in order to boost international visibility; the application of institutional forms of organization of nets of connection; the experimentation of wider and more open outlines. On the one hand, there’s the need to realize design strategies that boost quality levels of productions and to set up stronger connections between research institutions and productive realities. On the other hand, the likely reduction of strategic value units, which determines the district by crossing the limits to a global level, emerges. In the last case, the scenario that has been developing lately leads to virtual districts which, supported by Information and Communication Technologies (ICT), boost global associations and integrated systems. This is possible on condition that every single enterprise can run a complex net and transfer and share its know-

154

how. As a consequence, design plays a key role in the productive system and works in several decisional and design areas. Design is defined as an “incessantly starter of innovation”. It is included in the interaction process between the economic system and the productive system, through the setting of enterprise strategies and the planning of products and services. It is the new innovative resource which fosters the competitiveness of the enterprises and guides them toward new scenarios and new association methodologies. Design has very unstable action limits. The overlapping and integration of fields helps build a complex planning course. Through design you can put together management, environment, product and services system. 2.2 Planning methodology to innovate fashion oriented production. The composition of fashion oriented productive branches is a competitive advantage (flexibility and productive excellence, connection to traditionally productive handmade branches, close links to the territory…). The dimension of the factories itself, though, stops the economic growth. Small and medium factories which characterize this field cannot afford the investments necessary for the research and the creative and design processes. As a result, innovation is out of reach for most of the “actors”of the system, who aim to the excellence of some productions and to the ability of acting as a “die”. According to Richard Florida, the University of Stanford and MIT were of paramount importance in the development of Silicon Valley and of the even more productive area in Boston. Every important innovation is based on university and young creativity. An important evaluation in order to plan strategic guidelines of development in the areas in which Universities work is the brain drain/gain index . It corresponds to the capability of keeping the talents in the area in which the university is based. The key to the economic development of new industrial systems is planning even more effective ways of transferring university researchers and human wealth to the territory in which it works. In the current productive system, in which the winning enterprise is the one which can put as much know how as possible in the products, the strategy which should be used at a “country” level, lies in the creation of an ecosystem in which the main figures, according to the role they play, contribute to the development of really competitive strategies. The core of this idea is to realize a series of strategies in order to create an engine for creativity and innovation, which can gather all phases - creativity, research, services, transfer of know how to the enterprises, production. Research centres, main enterprises, sub-suppliers factories, young researcher and creative talents must all be taken into consideration in this phase. The core of this project will be the realisation of a centre dedicated to research, the transfer of the achievements of the enterprises, and the development of creative ecosystems. The centre will be founded on both a university and an enterprise basis. The university will supply it with education and research while the enterprises will put together university resources, development and innovation. As a result, there will be a district dedicated to research, innovation, education in the design and fashion driven field. 3

CASE STUDIES

3.1 Design Polling Lab and Fashion Net Factories In order to define strategic actions and new creative models of interaction between productive areas,

155

innovation and design, you can refer to areas of interactive innovation or Living Labs, whose experience is highly innovative. The Living Lab model, created by a group of Scandinavian factories, was planned with the purpose of involving several users in the development of new design ideas (especially the ones relating to ICT), in urban workplaces and areas. In time, this experimentation turned into a network that involves many European cities. It also represents strategies that relate to cooperation plans which are based on the reproduction, trough systems of shared creativity, of the dynamics of fertilization of the territory. Considering this experience as a model, you can give birth to a creative lab that supports the productive field and guarantees a more effective and structured cooperation between university and productive branches. The Design Polling lab is an itinerant lab that relates to all figures that take part in the process, from designers to entrepreneurs to consumers. [6] Moving from the evolution of the idea of co-design, which lives in the one to one relation between entrepreneur and designer and materializes in the product (as a sum of planning choices, actual realization and consumer needs), the DP Lab experiments a new planning methodology by widening the number of figures which take part in the project. This new model of design environment consists of resources and methods that make interactions between designer, producers and users more effective. In the current scenario, there is an even bigger need to reduce the time that passes between the design and the commercialisation of the product. As a result, needs analysis and the direct or indirect involvement of the users make a product successful in the market. Users who buy an experience, actually pay for spending some time enjoying a series of unforgettable events that the enterprise organises in order to keep them “directly” busy. Consequently, at the origin of products/services or system of services design, new ways of interaction with users must be planned. In this scene, the quick evolvement of the demand and the consequent change of the offer, shed more light on the role of the user in the chain of value. The boosting of relations between client and supplier (which have a biunique nature and that realize in the offer of services and instruments of socio-economical mediation) contributes to the realization of knowledge intensive goods. These very intense relations widen the spheres and the competences that help define the design choices and support the birth of a collective collaboration. Therefore, the concept of product changes completely. In addition to its actual dimension, there is a more complex dimension which is linked to the other and sometimes overshadows it. This dimension includes cultural and social aspects of the place the product belongs to. In this direction the DPLab aims both to define cooperative solutions among different figures which take part in the same design process and to build an efficient system which can involve entrepreneurs, institutions, designers and consumers since the very first phases of the planning of products/services. It is an area of research and innovation where you can experiment new comparison and innovation between university and productive design oriented branches. It is a free space, characterised by the direct exchange of knowledge, in which you can design and test new products for innovative scenarios. The wider and wider involvement of figures is even more effective if it works, simultaneously, through different actions. So, the more various the actions in the Lab is (workshops, talks, seminars) the wider the community involved in it is. This will lead to a wider range of ideas connected to the Lab. Consequently, on the one hand this will create a bigger amount of solutions of a specific problem, on the other hand it will enhance the project that will be closer to the needs of the users. This

cooperative model includes the expectations of the consumers. It also deals with the innovation projects related to the final product. The DPLab, which is by nature itinerant, has a flexible structure and fosters the connection of the different figures to the territory that play a role in the production and consumption process of a product, a service or a system of services. Each and every experience in the Lab relates to a different group of stakeholders, university departments, public agencies of development, consumers. They all aim to the realization of an innovative product, which can be created in the short term. The creative lab community, which aims to common targets, is based on the symbiotic integration. It highlights the characteristics of every single components according to their own knowledge. [7] Consequently, an actual application of the DPLab model is represented by the experimentation of a virtual fashion city. It is a virtual city that favours communication and information which, from the inside, involves the units of the district. The FNF enterprise consists of a whole that is made up of many equivalent parts, put together by the will of creating a model in order to use not just the fashion product, but the whole system. The virtual district is like an agora made up of places that are like pictures of the identity of the brand they belong to. They are places where the relationship between clients and enterprises can be customized and the communicative identity realizes itself. They are like imaginary spaces where the real places are reproduced in a virtual way. Body features are linked to relation ones, so relationships define the place-meeting among users. The city derives its shape from the pictures of every brand while users became active part of the process. By sharing information a learning relationship takes place. During it the producer obtains information about the client. The main advantage of the new one to one relationship is founded both on customized communication and on location of goods and services. You can reach a high level customisation of services, which leads to costumer fidelization (one of the most important dynamics of the e-commerce). Distances between offers are deleted and competition stays on the same level. Individual communication with a single client overcomes mass communication, goods become a unique individual experience, so the clients approaches and relates to the brand in an exclusive way. Internet favours the combination of different types of communication linked

to nature, use, dimension and quality of the interface, internal and external environment of the organizations. Consumers and factories share the same space. By stimulating the biggest amount of contacts with users, you boost the chances to offer targeted solutions . Involvement, emotional participation, critics and the creation of a virtual community shed light on the clients’ attention and the fidelization process. Through econsumers the use of goods and services is actually a free choice, justified by personal taste and interests. In this experimentation, the virtual promotion of the activities of the districts deals with a specific planning methodology. In addition to this, it defines the methodology’s requirements, the way in which communication and distribution take place, the content of the interface in the virtual environment and the consequent way of interacting with users community. 4 REFERENCES [1] Drucker P. F., Management Challenges for the 21st Century, New York: HarperCollins, 1999. [2] Belussi, F., Pilotti, L., “The development of an explorative analytical model of knowledge creation, learning and innovation within the Italian industrial districts”, Geografiska Annaler, n. 84, 2002. [3] Belassi, F., Gottardi, G., Rullani, E., The Technological Evolution of Industrial Districts, Kluwer, Boston, 2003. [4] Belassi, F. , Sammarra, A., Industrial districts, relocation, and the governance of global value chain, Cleup, Padova, 2005. [5] Becattini, G., From Industrial Districts to Local Development, Elgar, Cheltenham, 2003. [6] Prahalad, C. K., Venkatram Ramaswamy, The Future of Competition: Co-Creating Unique Value with Customers, Harvard Business School Press, 2004. [7] Sawhney, N., Cooperative Innovation in the Commons: Rethinking Distributed Collaboration and Intellectual Property for Sustainable Design Innovation, M.S. Media Arts and Sciences, Massachusetts Institute of Technology, 2003, http://web.media.mit.edu/~nitin/thesis/

156

Product service value analysis: two complementary points of view T. Alix, Y. Ducq, B. Vallespir IMS, UMR CNRS 5218, ENSEIRB, University Bordeaux 1 351, cours de la liberation, 33405 Talence, FRANCE [email protected]

Abstract Product Service System is an innovation strategy, used by manufacturers to increase their competitiveness, while satisfying specific customer needs. Here the focus is on the additional services associated to products and precisely on their value to determine their position in firms’ portfolio and by extension their legitimacy to reach the implied objective of profitability. The study rests on the use of the value analysis methodology and compares the value of the offer for the manufacturer taking account of its expected benefits and the value of the same offer for customers defined by the way of expected quality criteria. Keywords: Product-service, value analysis, product service oriented PSS, customer loyalty, innovation strategy

1 INTRODUCTION Whatever their sectors of activity, firms are studying the opportunities to create, develop and propose product service system (PSS) to fulfill specific consumer demands and enhance firm competitiveness. For manufacturers, this problem is complex because the ownership of the tangible product can change (from the customer to the service provider) and also because the focus is on the utility value of products in relation to the client’s activities. In this article we consider the case of product oriented PSS defined as a PSS where ownership of the tangible product is transferred to the consumer, but where additional services are provided. The problematic concerns the definition of the price of such an offer as additional services are most of the time perceived by customers as tools for differentiation and consequently supposed to be free of charge. Practically only few firms’ sell such service although they cost money. The main reasons come from: •

The difficulty for manufacturers to clearly define the cost of services because of their IHIP characteristics (Intangibility, Heterogeneity, Inseparability and Perishability).

The difficulty to associate a standard of price to an offer linked to a mercantile strategy. The challenge is then to define a PSS that would be profitable for both new manufacturer service providers and customers i.e. of a so high value that these latest would pay to own the additional services. To reach this objective, firms’ have to manage:





Customers’ satisfaction through the fulfillment of service needs and through the respect of quality criteria expected by during the whole service life cycle.



The value of each customer regarding the profitability that is generated. This profitability runs on the short, medium and long term, is not only monetary and necessitates comparing the costs that rest on the offer definition and delivery to the set of advantages that reflect firms’ benefits.

CIRP IPS2 Conference 2009

157

The idea behind is to compare the value of an additional service for the customer to the value of the same offer for the provider to determine its position in the portfolio of the firm and consequently its legitimacy to be proposed and enhance firm competitiveness and profitability. The study rests on the use of a smoothly adapted version of the value analysis principles proposed by Lawrence Miles in 1946 [1]. The position of the offer in the portfolio is determined using a matrix inspired by the BCG matrix well known in the management science [2]. Indeed, the results of the value analysis are gathered in a matrix that can be used as a strategic tool to analyze the relevance of an additional service offer integrating its costs and the mercantile strategy. Outlines consist in analyzing the functions expected by the customer and by the firms’ managers to the cost of the offer and to compare them, once the additional service is determined. For this concern, in the remaining section of this paper, we first present the context and the motivations of this study and define the offer we consider and its characteristics. Second, after a reminder of the value analysis method and of its limits, the cost function matrix of the firm is presented based on the benefits expected from a product service offer and based on the inherent costs. The cost function matrix of the customer is detailed based on its expectations in a fifth part and finally compared to the previous one using a matrix before giving limits to our works and before concluding. 2

STUDY CONTEXT AND JUSTIFICATIONS

2.1 Product-service offer Currently the revenues of many firms are becoming dominated by the sales of services rather than products, or by the sales of products together with services to gain competitive differentiation in markets marked by increasing product commoditization [3]. Service offers have different denomination in the literature depending on which character is considered regarding the tangible offer: complementary [4], [5], combined [6], or dependent [7], [8]. Here we focus on product-service as defined by Furrer in 1997: “Product-service are services supplied in addition to a product and increasing its value for the

customers” [9]. Technology-based firms can propose services such as maintenance contracts for product upgrades, technical support, strategic consulting, etc, because of the complexity of their products and to technological progress. Other examples of services proposed by manufacturing firms are listed in [10] and [11]. As we can see these services are mainly a matter for the tertiary sector of activity.

or customer high value or firm and customer low value? These questions fall into the problematic of new productservice design in a win-win strategy. De facto, we will use the value analysis method to study this sub-problematic.

2.2 Profitability background Revenues released by such services can become larger than product revenues. Software product companies are a good example of the success of service sales. Basically, big companies have become customary with product-service sales; they generate profit and can even sell services independently from products. SME in the manufacturing area are less hardened to this routine and only propose to sell a service jointly to the sale of a product. Profit for these kinds of enterprises is questioned; practically, a study performed by Baglin showed that whatever the type of service only 31% of SMEs sells them. The reasons that explain this loss of profit and this brake to the development of service offers is twofold [12]:

3.1 Value analysis definition and outcomes Value Analysis (VA) is “A systematic approach used to analyse functional requirements of products or services for the purpose of achieving their essential functions at the lowest total cost”. It defines a "basic function" or “main functions” as anything that makes the product work or sell and defines “secondary functions” or "supporting functions" as functions describing the manner in which the basic function(s) are implemented. Main functions cannot be cancelled (their fulfillment is essential), while secondary functions can be modified or eliminated to reduce product cost. Basically, objectives of VA are to optimize product design and to increase the difference between the cost and the value of a product via the application of a function analysis to the component parts of a product (standard EN 12973). The concept of value is difficult to determine as it is a subjective concept. Its common characteristic is a high level of performance, capability, emotional appeal, style, etc. relative to its cost. A review of the literature led Zeithmal to identify four common uses of the term [16]:



Service costing which can be difficult to evaluate because of service specificities [13], [14].



Price fixing that can be modified depending on whether the service is interpreted: tools for competition demarcation or real added value for the customer.

2.3 Specificities of product-service Products / services distinctions, mainly proposed by marketing and other disciplines, stem from “idiosyncratic qualities or the development of service packages according to market segment” [15]. Historically, services 1 were defined using IHIP characteristics and the duality outcome-versus-process. Conclusions being that services contrast with goods and are what these latest are not. Recently, a debate that challenges these definitions has become clear considering that they only serve the definition and use of specific managements models, methods and tools. As a result products and services can encompass the same properties according to the scale of time and place considered and can be more or less IHIP. Product-service can be tangible or not, but what is sure is that (i) they serve firm differentiation and the underlying objective of customer loyalty and, (ii) they cost money to the firm which provide them. Providers have then to master the costs of a productservice and to enhance its value so that it is worth in the eyes of the consumer. In this case, customers might become loyal and even pay for the service. The value of the offer might also be examined from the provider point of view to determine if the costs generated by the production and delivery of the service are not too important regarding the performance of the expected benefits especially if customers do not participate. The comparison of the two values could help responding to two questions: •

which product-service providers should propose that can be of a so high value that customers will pay to own it, of a so high value that it is profitable for firms and does not encroach upon the margin released by the core product?



which position does my product-service occupy in term of value: firm and customer high value, only firm

1

3

VALUE ANALYSIS: A SIMPLISTIC RECALL



value as a price;



value as what I get for what I give;



value as the trade-off between price and quality;

• value as an overall assessment of subjective worth. In the following, we use this latter definition as it is the widest one and encompasses all attributes of price, quality, and satisfaction... in the subjective worth concept. In 1999, Goyhenetche has expressed the value concept by a ratio [17].

value =

appreciation of offered services trust * appreciation of resources expenditures risks

value =

performance of functions costs

(1)

where the cost is the amount that is incurred in the production and delivery of the product. It might refer to the total life cycle costs: over the whole life span of the product (costs of the material, the manufacturing and assembly as well as all planning, investment and arrangement costs). In practice, in VA costs refer to cost components.

3.2 Value analysis method and process From a product design point of view, products of high value first address the basic function's performance and stress the achievement of all of the performance attributes. Secondary functions used to attract customers are then developed. A way to increase value consists in increasing the performance of the product functions or in reducing the costs. The elimination or combination of secondary functions can for example reduce costs without detracting from the worth of the product. In all cases, as performance and costs can be neither reduced to zero nor infinite, VA allows to compare the costs of a product to the value as perceived by the

Intangibility, heterogeneity, inseparability and perishability.

158

customer on the basis of the expected service specificities covering. The VA procedure contains different steps that can be summarized as follows: once the problem and its scope are defined, basic and secondary functions are listed. A cost function matrix is then built and allows to allocate shares of the cost to the individual functions. Product functions with a high cost-function ratio are identified as opportunities for further investigations and improvement that are brainstormed, analyzed and selected.

3.3 Value analysis improvements Strengths concern:

strengths

and

points



the method recognition all over the world,



the method applicability to all sectors of activity,

the interdisciplinary, systematic approach to problem solving. Points for improvements:



and

for

organized



the use, limited to the design of tangibles,



The cost determining that is often oversimplified. Indeed, when referring to consumer products, the costs taken into account are mainly and directly associated with product manufacturing: component costs. In the case of expensive capital equipment, costs refer to manufacturing, installation, maintenance and decommissioning costs. Most of the time, no indirect costs are taken into account.

The scarcity of points of view confrontation. The VE method states that the repartition of the products costs is coherent with the benefits expected by the customer. The method is never used as a strategic tool that can help a business operator to choose which business to develop, which product or service to sell or what customer to target regarding its expected benefits, the cost of the offer and the customer value. The following sections deal with all these points, trying to apply value analysis to new product-service design of high value for the different protagonists.



4

PRODUCT-SERVICE VALUE ANALYSIS: FROM THE PROVIDER POINT OF VIEW

4.1 Problem and scope The works procedures of VA lead us firstly to define the problem and its scope. Obviously, here we focus on new product-service design. This one can be either tangible or intangible which means that there can be components or not. This point is important to notice because value analysis deals with a cost function matrix based on the sharing out of component costs. Adaptations should be made to address the intangible product-service part. The service is not free of charge, costs are determined hereafter. 4.2 Value analysis matrix construction Description of functions: from the provider point of view We assume that the functions which participate to the definition of the value for the provider concern the expected benefits of a product-service proposal. A study performed by Malleret on this subject showed that they relate to four major themes [11]: •

The construction of a customer loyalty by the building of dependency relationships between a consumer and a provider that can lead toward profitability.



The search for differentiation that allows to retain existing consumers and to attract new ones.

159



The increase and stabilizing of firms’ turnover due to the possibility to generate regular income and to have cash flow disposal.

The corporate image reinforcement in fields like technological advanced, product quality... A review of the literature in management led us to list other benefits that can be expected, among which:





The occupation of an existing or new market to participate to market share division.



The possibility to create alliance with service providers and to share risks.



The possibility to increase the quickness of a design or production process using product-service based on information and communication technologies.

The possibility to shorten sales delay or negotiation phase using financial services. • The search for a PSS that is designed to have “a lower environmental impact than traditional business models" [18]. It is to note that this specific benefit is not taken into account in the rest of our study as it is not very relevant regarding the part of the PSS we consider. Each of these functions can be classed as expected performances that stem from a strategy and have priorities one to the other: classification in main or secondary functions and/or classification in percent importance of each one using cross analysis. Quantifiable criteria can be associated to each function whose level also stem from the strategy. The level really measured, that reflect the performance of the function, compared to the global cost of the service could allow determining the value of the service for the firm.



Definition of costs Costs to take into account can be divided in direct and indirect costs. Usually, direct costs are entirely tied to a product or service while the other charges common to several ones that belong to a same type or to different ones are indirect costs. In the case where costs are not only to be set for one function, they are to be added proportionately. Cost emphasis can be performed by associating a weight to the participation of the component to a function. Regarding product-service characteristics, several costs can be addressed that depends: •

on its degree of tangibility,



on the degree of interaction that is necessary between the firm contact personnel and the customer to deliver it,

on the degree of standardization of the productservice delivery process. Then costs can encompass component costs, cost of labour, and overheads. Cost of component concern: (i) the cost of consumables that are used to make the product-service tangible (raw materials, paper for documentation, ink for printers, CD for filing…), and (ii) cost of physical support necessary for its realization (eg. Manufacturing resources, specific software, computers, and vending machines). Cost of labour depends on which management function is implied in the product-service delivery system. Strategic and management studies on immaterial delivery have shown that three main functions are concerned: the marketing one, the human resource management one and the commercial one. This seems to be coherent with the concept of service activity rather based on individuals’ interactions. In material production, the



design and manufacturing costs are the most important ones. Some of these costs of labour are direct ones while others are indirect ones. For the immaterial part of product-service delivery, a service life cycle stemming from the cost functional analysis process [19] led us to insert the labour of the marketers in the indirect costs while the labour of commercial people can be considered as a direct cost. Overheads concern fixed costs (investment, rent, writing off), facility costs (taxes), and indirect labour (marketing, finance, cleaners…). Regarding the method, elements of costing are to be determined and gathered by categories taking account of the product-service specificities. Then, an estimation of the value of each one is to be done.

Function cost appreciation The description of the functions from the provider point of view and the consciousness of the product-service costs allow to build a first value analysis matrix (see Figure 1 at the end of the paper). The value analysis matrix displays the list of elements of the product service that might be taken into account, and their cost, along the left vertical side of the graph. There are as many lines as elements and several lines per element can be noted. The top horizontal legend contains the functions expected by the provider. To determine the value: •

Overheads and indirect cost labour must be shared between all of the product-services of this kind delivered to customers. This can be done on the basis of marketing studies which identify the potential amount of customers as well as the minimal amount of customers to ensure firm profitability.

Component cost must be shared between the functions according to their participation to the functions fulfilment as well as costs of direct labour. The valuation of this cost is done by comparing the duration of the interacting process to the charged cost of labour per time unit of the commercial people in the case where interactions are necessary. Once costs are shared out, it is possible to calculate the cost of each function and evaluate the relative cost of all functions. The total cost and percent contribution of the functions of the product-service under study will guide the firm deciders in selecting which functions to address for value improvement analysis regarding its percent importance.

5.1 Complexity of functions and costs determining Product service technical functionalities Initially, the service orientation that encourage manufacturing firm to accompany product with service referred to all activities suppliers can undertake to help purchasers in choosing, acquiring and using a product. Regarding Furrer’s definition, added value linked to product-service concern product functionalities valorization as well as facilities to obtain and correctly use it. Then, each one has a “raison d’être” regarding the core product, characterised by technical functionalities, herself characterised by performance levels. Moreover, product-service can also have specific functions that can be determined by analysing its life cycle and/or by analysing its environment. Interactions between the contact personnel, the users, the means necessary to realize the service, the partners, the environmental and legislative constraints …can lead to identify secondary functions. By the same way, realization constraints can give rise to additional functions. The performance recorded for each function of the product-service, once it is delivered to the customer compared to its global cost might, as previously, allow determining its value. In practise, value is much more complex to determine because of the difficulty to determine the complete list of functions expected by the customer and also because of the difficulty for him to define an objective cost. According to Hermel & Louyat [20], the customer challenges the overall value to the complete cost. The overall value refers to the different advantages obtained, supported by the provider brand image. The costs are composed of the monetary, functional and psychological cost, as well as the costs linked to the time spent to evaluate, acquire, use and eventually abandon the offer.



customer value =

whole appreciation of offered services costs

This way of doing renders the value concept very subjective as it depends on customers’ frame of mind. Moreover, advantages should not be restricted to benefits expected on technical functionalities as lots of studies have shown that the customer is waiting for something else from the exchange with the contact personnel: empathy, reactivity, availability... These latest elements can be associated to implicit functions whose fulfillment can lead to customer loyalty and to value increase for the customer and for the provider.

4.3 Investigate improvement Now that the cost contribution to the functions is established, providers are in a position to determine if high cost parts can be removed, identify high cost/low value and low cost/high value parts; ask basic questions concerning the elimination, reduction, simplification, modification or standardisation of functions, or even more of the service by itself.

Implicit performance expected by customers Customer loyalty comes under customers’ satisfaction. In 1991, Zeithaml, Berry and Parasuraman have developed a method that allows evaluating the quality of a standard service [21]. This model called servqual is based on ten criteria gathered in 5 dimensions regarding their correlations: •

tangibility: appearance of physical facilities, equipment, personnel and communication materials,

5



reliability: ability to perform the promised service dependably and accurately,



responsiveness: willingness to help customers and provide prompt service,



assurance: competence, courtesy, credibility and feel secure,

PRODUCT-SERVICE VALUE ANALYSIS: FROM THE CUSTOMER POINT OF VIEW The problem and its scope are obviously the same. It is to note here that in customer understanding, a productservice is most of the time a service that must be tangible for a part and that requires an exchange phase with a contact personnel of the firm.

empathy: accessibility, good communications and customer comprehension. As we can see, quality criteria refer to the reliability that is similar to the function “to satisfy the technical •

160

functionalities” and to other aspects that refer to service characteristics (tangibility) and to service delivery system (terms characterizing the exchange with the contact personnel). This comes from the difficulty to separate the process from its outcome in service production. As the customer can be involved, most of the time he mixes its feelings. In the following, we assume that only the criteria that concern the service by itself have to be taken into account in the list of customers’ functions while the others have to be taken into account by the firms as they refer to the way they could make customers loyal (explanations are given in section 7).

Product-service list of function from the customer point of view The list of functions of a product-service expected by customers, expressed as a verb and a noun consists of: •

the product-service raison d’être: help choosing, acquiring or using the main product,



the secondary functions linked to its environment: due to interactions with the contact personnel, the users, the means necessary to realize the service, the partners, the environmental and legislative constraints and realization constraints as mentioned previously…



the implicit functions coming from quality criteria discharged from the functions that refers to the delivery process: to obtain a tangible service.

5.2 Customer value investigation To increase the customer value appreciation and its satisfaction degree, providers have two solutions (see equation 2): •

to diminish the cost of the product-service,



to minimize the gap between the quality criteria extended to several functionalities of the productservice that the customers is waiting for (WQ) and the quality he perceives once the service is delivered (PQ), while maintaining costs (C), (2).

customer value = function{PQ,WQ, Costs}

Whole appreciation of service increase for customer value increase Increase of the appreciation can be done by managing service quality while reviewing cost ventilation on customer quality criteria extended to several functionalities to avoid cost due to a bad repartition or due to the covering of unnecessary functions. This analysis can be done building the value analysis matrix from the customer point of view, see Figure 2. The same elements and costs than in the value analysis matrix from the firm point of view are mentioned while the list of functions has changed accordingly to the appreciation criteria. 6

COMPARISON BETWEEN THE CUSTOMER AND THE FIRM VALUE Regarding the two value analysis matrix, it is possible to determine the cost contribution of the functions and to propose improvement for better repartition if necessary. Then using aggregation operator, it is possible to deduce the whole value of the product-service proposed by a firm. This one can have two positions: high or low. The value of a product-service can be analysed regarding two dimensions: the customer dimension and the firm dimension, see Figure 3. Then four cases can occur: •

The value is high for the customer and for the firm. In this case, the product service is profitable for the firm and satisfies customers. It must systematically accompany the product.



The value is low for the customer and for the firm. The abandon of the product-service or not will depend on the cash that is necessary to provide it or on the delivering difficulties.



The value is high for the customer and low for the firm. In this case, to make the customer loyal, firm should propose the service but might found solution to increase its value perhaps by increasing the price if the service is worth in the eye of the consumer, adding others services located in the high/high value. Advantages will be to dilute the costs.



The value is low for the customer and high for the firm. This position is good for the firm as it is synonymous of high profitability and customer loyalty if the customer participates or if the cost of the offer is not too important. Otherwise, as the value is low for him, he won’t be interested by this list of sales point and will surely look for other offers with high value from the firm or from its concurrent.

(2)

Cost diminishing for customer value increase Even if the customer participates to the service delivery process, costs are not easy to determine for him as they are composed of providers’ internal accounting data. Regarding the previous statement how is it possible to determine the costs linked to the time spent to evaluate or realize a service? Does any customer time the duration for service evaluation and compare it with his cost per hours? And how can be defined this latest? Going further, has any customer an idea about the salary of the contact personnel? Basically, a customer will define a cost by comparison of the prices applied on a market (acting that the margin is equal for all competitors) and/or by experience of a price applied (supposing that the margin is constant). This implies that a non lining up of the price to the concurrence or a price modification will have to be deeply justified. Most of the time, the justification comes from the explanation of the difference between a service defined as a standard and the proposed service. The price that a sample of potential customer agrees to pay an additional satisfaction increases the price of the standard and is used as a reference on the market. Conversely, a price reduction will be associated by the customer to a satisfaction diminution. As margin is low in service delivery, the price can be compared to the cost. Then a smooth modification of costs can be envisaged but is not the solution to increase customer value perception.

161

Firm value

Customer value

High

Low

High

Profitable and interesting

No profit but interesting

Low

Profitable but No interest

No profit, no interest

Figure 3: Value matrix

.

Table 1: mapping providers priorities for quality assurance to customer quality criteria The analysis of each product-service associated to a specific product, in one category, gives a synthetic view of the equilibrium of the global offer. To ensure firm global profitability, the portfolio of product-services have to be shared out to all categories.

7 LIMITS OF VALUE ANALYSIS The value analysis process underestimates most of the time the dependencies of the functions. When studying firms expected benefits, it becomes apparent that functions do not operate in a random or independent fashion. Functions as well as elements form dependencies link with other elements to make the system works. The quality criteria developed by Zeithmal can help defining sub-functions and dependencies between firm expected benefits. From quality criteria to firm’s management priorities As mentioned in section 5.1, on the ten criteria defined to evaluate the quality of a standard service, only two were integrated in the list of functions expected by the customer. What about the others? The remainders refer to the service delivery process and characterize the interactions between the contact personnel and the customer. The qualities of the exchanges are to be supported by the service provider to satisfy these latest. Lots of works done recently in the domain of service quality management have been proposed. Thirteen priorities that firms have to complete to ensure quality have been identified [21]; see the right side of Table 1. All these priorities are obviously customer oriented in order to manage an economical context of hyper competition in a continuous evolution, filled with enterprise networks and immaterial assets to acquire new customers, retain existing one by the way of an high value perceived. The link between the quality criteria and some of these priorities gives key to support customer loyalty; see Table 1.

process to study function links while exploring opportunities to develop improved systems [26]. The diagram is built upon the HOW-WHY dimensions to structure the logic of the system’s functions. Regarding the two types of FAST diagrams, the priorities could help understanding the technical aspects to set up an effective service. Then, to answer the question “How to make customer loyal”, sub functions are: to adapt the service provided in time, to activate human resources, to manage dissatisfaction, to reinforce proximity, to personalize to make customer loyal, to optimize organization, to compose and master the global offer. As we can see there exist a link between that latest subfunction of the function called FC1 in the value analysis matrix and the function FC4 of the same matrix. Then costs have to be carefully shared in order to avoid redundancies or cost bad repartition.

7.1

7.2 One step to function analysis system technique The Function Analysis System Technique (FAST) represents functions dependencies and allows creating a

8 CONCLUSION The adjunction of additional services to the main product manufacturing firm provide to their customers also called product oriented service-system is now a current way of doing even if problems concerning the cost, the price and the value of such a proposal are not solved. Our contribution is based on the use of the value analysis matrix to determine the value of an additional service offer from two points of view. The first one relates to the provider and takes its expected benefits into account. The second one relates to the customer and takes expected functionalities as well as quality criteria expected during the whole service life cycle into account. The results are gathered in a matrix that can be used as a strategic tool to analyze the relevance of a productservice offer that integrates the real costs of design, production and delivery as well as the mercantile strategy. The main limit of our contribution concerning the underestimation of functions dependency let foresee lots of potential improvements. First, dependencies should be analysed to reinforce benefits expected by the use of a value analysis matrix (i.e. improve product-service design and lower cost, focusing on essential functions to fulfil customer requirements while being profitable. Second

162

the link with the QFD matrix might be better analysed to develop and analyse improvements in the offer. Third, the list of priorities might also be better analysed to see if others functions or sub-functions become clear for the provider. Here we have only paid attention to the function “to make the customer loyal” as it correspond to the “raison d’être” of a product-service. Third, the validation of the structure of the two proposed matrix is to be done quickly, on concrete examples to verify their legitimacy.

9 [1] [2]

[3]

[4]

[5]

[6]

[7]

[8]

[9] [10]

[11]

[12]

[13]

[14]

[15]

[16]

REFERENCES Miles, L. D., 1963, Definitions: Lawrence D. Miles value Engineering Reference Center: Wendt Library Jonhson G., Scholes K., Whittington R., Frery F., 2008, Stratégique, Exploring Corporate Strategy, Pearson Education, Paris, 8e éd., 720 p., Cusumano, M., Kahl, S, Suarez, F, 2006, Product, Process, and Service: A New Industry Lifecycle Model” available online at: http://www.sbs.ox.ac.uk/NR/rdonlyres/99F135D4E982-4580-9BF08515C7B1D40B/1710/GCSServicesvsProducts_Cus umano_Kahl_Suarez.pdf Witt, R., Salomon, M., 1991, Value added services, a case study: US Electronic Components Distribution” in managing Services Across Borders, Eurolmog Presse: 149-161. Gadrey J., 1988, Des facteurs de croissance de services aux rapports sociaux de service, Revue d’Economie Industrielle, 43 : 34-48. Van Looy, B., Van Dierdonck R., Gemmen P., 1998, Service management – An integrated Approach, Prentice all. Frambach, R.T., Wels-Lips, I. and GuÈndlach, A. 1997, Proactive product service strategies - an application in the European health market, Industrial Marketing Management, 26: 341-52. Kotler, P. 1994, Marketing Management: Analysis, Planning, Implementation and Control, 8th ed., Prentice-hall, Englewood Cliffs. Furrer,O., 1997, Le rôle stratégique des services autour des produits, Revue Française de Gestion, mars-avril-mai : 98-108. Alix T. Vallespir B., 2006, Product and complementary service: looking for the right pair, The SSSM conference, October. Malleret V., 2005, La rentabilité des services dans les entreprises industrielle : enquête sur un postulat, Cahier de recherche du groupe HEC. Baglin G. Malleret V., 2005, Le developpement d’offres de services dans les PMI, Cahier de recherché du groupe HEC. Berry L.L., Yadav M.S., 1996, Capture and communicate value in the pricing of services, Sloan Management review: pp.41-51. Edvardsson, B., Gustafsson, A., Roos, I., 2005, Service portraits in service, research: a critical review, International Journal of Service Industry Management, 16(1): 107-121. Araujo, L., Spring, M., 2006, Services, products, and the institutional structure of production, Industrial Marketing Management, 35(7): 797-805. Zeithmal, V.A., 1988, Consumer perception of Price, Quality and Value: a means-End Model and Synthesis of the Evidence, Journal of Marketing, 52: 2-22.

163

[17] Goyhenetche, M. 1999. Le marketing de la valeur, Créer de la valeur pour le client, INSEP. [17] Oksana, M., 2002, Clarifying the Concept of ProductService System", Journal of Cleaner Production 10 (3): 237–245. [18] Ucalgary, 2004, Functional Cost Analysis, online at http://www.ucalgary.ca/~design/engg251/First%20Y ear%20Files/funct_cost_analysis.pdf. [19] Hermel, L.& Louyat, G., 2005, La qualité de service, AFNOR Saint Denis la Plaine, France. [20] Zeithmal, A. Parasuraman, A. Berry, L. 1991, Delivering quality service, The free Press. [21] Monin, J.M., 2001, La certification qualité dans les services, AFNOR, Paris la defense, France.

Total cost

%Cost Make Customer loyal FC1

Search For differentiation FC2

Increase Firm turnover FC3

Function (FC) Reinforce Occupy Corporate New market image FC4 FC5

Share Risks

Shorten Delays

FC6

Increase Process quickness FC7

Cost of FC6 FC6 percentage of importance

Cost of FC7 FC7 percentage of importance

Cost of FC8 FC8 percentage of importance

Satisfy function concerning interactions with partners

Satisfy Environment or legislative constraints

Satisfy Realisation constraints

FC’6

FC’7

FC’8

Cost of FC’6 FC’6 percentage of importance

Cost of FC’7 FC’7 percentage of importance

Cost of FC’8 FC’8 percentage of importance

FC8

Elements Consumables Physical support Direct cost labour Indirect cost labour Overheads % cost

Total cost

Cost of FC1 Cost of FC2 Cost of FC3 Cost of FC4 Cost of FC5 FC1 FC2 FC3 FC4 FC5 percentage of percentage of percentage of percentage of percentage of importance importance importance importance importance Figure 1: Value analysis matrix from the product-service provider point of view

%Cost Help Choosing, acquiring or using the main product FC’1

Make Service tangible

FC’2

Satisfy function concerning interactions with contact personnel FC’3

Function (FC) Satisfy Satisfy function function concerning concerning interactions interactions with users with means FC’4

FC’5

Elements Consumables Physical support Direct cost labour Indirect cost labour Overheads % cost

Cost of FC’1 FC’1 percentage of importance

Cost of FC’2 Cost of FC’3 Cost of FC’4 Cost of FC’5 FC’2 FC’3 FC’4 FC’5 percentage of percentage of percentage of percentage of importance importance importance importance Figure 2: Value analysis matrix from the customer point of view

164

Business Implications of Integrated Product and Service Offerings M. Lindahl, T. Sakao & A. Öhrwall Rönnbäck Department of Management and Engineering, Linköping University, 581 83 Linköping, Sweden [email protected]

Abstract This paper explores the business implications of Integrated Product and Service Offerings (IPSOs). The objective is to show examples of the business implications of IPSOs from a supplier’s perspective, and to suggest specifications for supporting methods needed for such an industrial company. The paper is largely based on empirical case studies of 120 Swedish manufacturing companies of all sizes. Results from the case studies show that both small and large companies that conduct the transition towards IPSOs face several important strategic challenges, some of them associated with high risk.

Keywords: Product Service Systems (PSS), IPS², methods, Integrated Product Service Engineering

1

INTRODUCTION

1.1 Challenges of integrating product and service In the search for strategies toward higher competitiveness, a trend among manufacturing companies is to create value for their customers by offering complex system solutions consisting of combinations of hardware, software and services, tailored for the specific needs of the customer (e.g. Oliva and Kallenberg [1], Vargo and Lusch [2, 3]). These offerings are often referred to as Product Service Systems (PSS) (e.g. [4]), Industrial Product Service Systems (IPS²), or, as in this paper, Integrated Product and Service Offerings (IPSOs) [5]. First of all, the challenge in this field, namely the variety of business models and the strategies behind them, should be recognized (see Section 2). Tomiyama et al. [6] present this collection of varied types and discuss it in the context of design methodology, taking the example of washing clothes, which is broken down into utilizing/renting a washing machine, coin-operated laundry, laundry services, renting clothes, etc. Allmendinger and Lombreglia [7], on the other hand, classify the strategies of companies providing PSS into four types - a) embedded innovator, b) solutionist, c) aggregator, and d) synergist - through observing practices in various industries. This classification depends on the degrees of being product-centric, reliance on business partners and so on. From an engineering viewpoint, one question in the industrial sectors concerned with this growing type of offer is how companies can find fruitful combinations of traditional product offerings with services that customers want and are prepared to pay for, that at the same time can be efficiently provided by the supplier, with sane risktaking and long-term profit. Another important question is what kinds of parameters should be addressed to do so. Previous research has attempted to answer these questions. For example, Morelli [8] argues that the design

CIRP IPS2 Conference 2009

165

of a PSS falls in a different domain than that of a traditional product. In addition, he states that the design discipline has no methodologies to operate in such domains, and proposes the use of envisioning with a scenario. Sakao and Shimomura [9], in the context of sustainable production and consumption, argue that a much larger framework than product design is needed, because the business model is ultimately changed; they suggest the use of disciplines such as engineering, marketing, and management. In other research [10], it has been argued that PSS clearly should involve several participants and new actors in its development process, and must consider company strategies (positions), organisational structures, and economic consequences, something that is further elaborated on and emphasized by Isaksson and Larsson et al. [11]. Sakao and ÖlundhSandström et al. [10], as well as Isaksson and Larsson et al. [11], argue that further research is needed for these questions considering both economic, engineering design, and environmental consequences. One important argument is that the shift in business models towards IPSOs implies a new mind-set and organising framework at the industrial company [1]. Vargo and Lusch [3] present the service-dominant logic, as opposed to the goods-dominant logic: “The process of providing service for (and in conjunction with) another party in order to obtain reciprocal service, is the purpose of economic exchange”. Goods are seen as carriers of function Instead of being the primary base for business. They [2] call for a natural shift from marketing theory and practice influenced by classical economics to a servicecentred model influenced by “marketing as a social process”, much in line with relationship marketing researchers Normann and Ramírez [12], Normann [13], Gummesson [14], and Grönroos [15]. Obviously, such a fundamental change of perspective opens new theoretical views on business models.

Although quite a lot of research has been carried out on PSS in the service marketing and environmental (ecodesign) fields (e.g. [2, 3, 16]), it appears that economic and business implications, i.e. regarding offerings, IPSOs, and business models, have been less elaborated on so far. 1.2 Objective The objective of this paper is to show some examples of the business implications of IPSOs from a supplier’s perspective, and to suggest specifications for supporting methods needed for such an industrial company. 1.3 Method The paper is mainly based on empirical case studies [17, 18] at Swedish manufacturing companies of different sizes. In total, 120 company case studies (58 small (10-50 employees), 3 medium (51-250 employees) and 59 large (>250 employees)) were conducted (size according to the European Union’s definition [19]). Of the 120 companies, some have participated in more than one case study, and are therefore counted more than once. Some complementary studies have also been made at banks (especially for financial solutions) and pure service companies (for comparison and refinement of the issues to study). The most-used data collection method in these studies has been qualitative research interviews [20]. The main purpose was to obtain a deeper understanding of the manufacturing company’s current business and product development activities regarding IPSOs vs. traditional product sales. A second purpose was to investigate potential needs for methodological support for development of IPSOs [5, 21-23]. Some companies were studied in-depth during a long time period with several interviews carried out on site; others have been studied through interviews with company managers during meetings and workshops with several companies participating. Respondents were e.g. product and service developers, CEOs, and customers. In most cases, face-to-face interviews were recorded. Question areas in the semi-structured interviews (for the purpose of this paper) were; Number of IPSOs compared to total sales volume; Profitability for IPSOs vs. traditional sales; Customer’s perceived value of IPSOs; Customer involvement in IPSO development; Contract forms; Supplier experiences of IPSOs (pros and cons); Uncertainty associated with IPSOs. Besides these question areas, other areas beyond the scope for this paper were touched upon, depending on the focus of each project. In parallel to the studies above, surveys were also conducted in Sweden, Japan, Italy and Germany [24]. This study covers 34 companies, some of which are Swedish companies already included in the Swedish studies above. 2 KEY STRATEGIC ISSUES The orientation towards IPSOs implies several important strategic implications for a supplier company, depending on its size and market position, customer demands and market characteristics. In this section, some of the most important issues found in the case studies will be scrutinized. 2.1 The role of company size and flexibility The most commonly used company size classification [19] defines four company sizes - micro (250 employees). The scope of this study did not include micro companies. However, based on the

analysis of the study’s result, 50 employees were chosen as a relevant and appropriate landmark to discriminate how businesses based on IPSOs are achieved. For example, medium-sized companies have more in common with large companies regarding this issue than with small companies; therefore, companies with 50 employees or more henceforth referred to as “large companies”. One observation, seen both in small and large companies, was that customers in the same market moved at a different pace regarding IPSOs. While some customers demanded IPSOs from their suppliers, i.e. tailored “total care” solutions with customized delivered value and often with important buyer-supplier relationships, others preferred traditional standardized (catalogue) products. These results were also supported in studies by Day [25], who argues that not all customers are willing to have the kind of close relationship to a supplier that is usually required for IPSOs and who concludes “both a product and a service-centred logic will co-exist in most markets” [25]. This can have different consequences for small and large companies, as will be addressed in the following sections. Our previous articles [24, 26] report that there are differences in providing IPSOs depending on the providers’ sizes. Large or medium-sized companies relatively often regard the customer demands, increased competition, and gaining larger product profit as the driving forces, whilst small ones rarely do. More large and medium-sized companies include operators, maintenance, repairs, and take-back responsibility in the IPSO, while fewer small ones do. Also, the ownership of the physical products often belongs to small-sized providers, as opposed to the situation found in large and medium-sized companies. In addition, small companies more often design products specifically for the IPSO than do large and medium-sized ones. Large companies Large supplier companies are organised with different functional units (departments), and thus tend to be less flexible and efficient when they begin a new type of business due to more rigid structures and higher overhead costs. Empirical results have shown that large companies often tend to create new organisational units in charge of activities linked to IPSOs. This approach both addresses internal organisation and the different customer demands on traditional products vs. IPSOs. From the characteristics of large companies described at the beginning of this subsection, being large has both advantages and disadvantages for IPSOs. A major advantage is the ease to consolidate different elements to form a package of solutions, as they can more likely be supplied from their own company. Disadvantages include the difficulty (connected to organisational rigidity) to change the development processes and ownership styles established for traditional product sales business. In addition, it takes much effort to change the skills and mindset of employees, where e.g. some used to work only in the technical world. These problems originate from the organisational subdivisions of a company. Many companies, especially larger ones, are divided into different departments, each with its own budget to follow, and hence ownership of resources, tasks and results. In addition, the functions of employees are more specifically allocated. Today’s companies are in general organised for the traditional business logic, selling products. It is quite common especially in large companies that departments, e.g. product development, production, spare parts, and after-sales, act like independent companies within the

166

company. Each department is an independent profit centre with yield requirements - and with managers whose bonuses and reputations are often related to their results of fulfilling those yield requirements. However, the traditional organisation has difficulties when switching to IPSOs. The new business logic as stated earlier implies that the focus moves from selling many products, spare parts, services etc. to instead providing an offering that reduces the need for products, spare parts, service etc. We have seen several examples of companies whose organisational structures conflict with their work with IPSOs. An example, seen in many companies, is related to spare parts. Spare parts are in traditional product sales often an important cash cow with high margins. However, when delivering IPSOs, the company itself becomes the major buyer of spare parts, and if the price models for spare parts remains unchanged (e.g. the internal spare part price is equal to that which customers pay), the cost for spare parts can jeopardize the entire company’s IPSOs concept. Furthermore, for a spare part department, IPSOs often imply that their turnover and profit will decrease - something not often popular among the staff. To conclude, when switching to IPSOs, it is important to evaluate the whole organisation in order to identify potential organisational obstacles that can jeopardize the IPSOs business, and thereafter adjust the organisation. In many cases, there are bonus systems connected to parameters related to the budget. Another challenge is how to convince customers [22]. So far, a contract in the form of “profit sharing”, where the company revenue is determined depending on the machine performance, does not seem so successful. The reasons include the accuracy of measurement and customer psychology. Here, the issue of “open books” is a key factor in the case of large companies as well. Small companies Compared to the conditions in large companies, the two most significant differences are that the small company’s transition into IPSOs is both easier, owing to more flexible organisation with fewer people involved, and done at higher risk, due to fewer financial and other resources. Many of the studied small industry companies have seen the potential, but, due to different reasons, have not fully realized it. In studies from 2002-2003, it was observed that some small companies had bundled traditional products with services into packaged ”total solutions” that would solve the customers’ problems; however, since the customers and the market were not prepared for this, many of them had to withdraw their offerings. The empirical results illustrated early on the difficulty for a small company to make the transition. The advice for these companies was therefore not to risk their traditional business while moving into the direction of IPSOs. Other empirical evidence found that small companies, during a short period of time, reoriented their complete business towards system solutions instead of traditionally selling single units, this for the benefit of their customers, reduction of waste, and also to increase profit for the supplying small company. Profitability also varied in the studied IPSOs as compared to traditional offerings; it seemed to be connected with initial cost for the small company turning into an IPSOs supplier (see also [27]), but also with the contracts and the possibility to communicate delivered value. It was common in the cases studied that service was given away for free from the supplier, and interviewed managers said that this was “comme-il-faut” in their business and taken for granted by their customers. Yet others, who had started to charge customers for service, did this with little or no fuss at all

167

from the customer side. On the contrary, this clarified business conditions between the parties, and strengthened the supplier’s position. The results of the difference in customer demands for traditional products and IPSOs [25] were also a common answer from CEOs, both in the early and in the later case studies (2003-2008), although more and more suppliers look at IPSOs today, since more and more customers ask for them. However, for the small company it can be a tricky challenge to manage both a part of their market that has realized the advantage of total solutions, and other parts that have not. Not all customers are willing to have the kind of close relationship to a supplier that IPSOs usually require. This implies that even the small company needs to be prepared to offer both traditional products and IPSOs (in different customer-specific variants). The small company’s flexibility to turn the whole organisation into ISPO also makes it difficult to uphold both lines of business simultaneously; this is the other side of the coin. Often among the studied small companies, hesitation to change into IPSOs is due to risk-taking. To fulfil larger, often more complex ISPOs including e.g. financial, product development, analysis, training, support and maintenance services, it often requires a network of several companies. If the small company is the point of contract, it needs to manage a business network consisting of different parties for each customer offering. Moreover, as e.g. Gummesson [14] points out, these parties tend to become partners. When a small company takes on large and complex IPSOs, the risk for both buyer and supplier can be higher. If the supplier fails, it always comes down to the customer; if delivering an important part of customer operations, this can be costly. However, the opportunity to find skilled specialists in small companies for specific customer-tailored solutions at a low total cost can be attractive. On the other hand, choosing a small supplier increases possibilities to work closely, thereby limiting the risk-taking. For example, in the case of failure to deliver from the small supplier or its ISPO business network, such skilled specialists can be employed by the customer company or the supplier, or parts of its activities can be incorporated into the larger company. Nevertheless, these are tricky operations. To manage risk when small companies take on large integrated product-service contracts requires careful and quite open negotiation about price and risk between the buyer and the supplier. It does not necessarily mean “open books”, i.e. full visibility of the supplier’s cost and price calculations, since this can make it difficult for the supplier to achieve profit margins. However, an open discussion of each party’s specific contribution and strengths for the IPSO fulfilment is required. The customer should also be attentive to the network of partners that the small company is depends on. If the supplier fails due to partner companies, the loser in the end is always the customer. 2.2 Managing the risks and opportunities of IPSOs The research shows that when offering an IPSO, a new mindset for how to secure the economic growth of companies needs to be developed [1-3, 28]. In the Goods-Dominant logic [2, 3], the supplier’s income is strongly connected to the sale of the physical product and from spare parts, incidental material, service and maintenance needed to keep it in serviceable condition. The distribution of income differs, from 100% on the initial product to more or less 100% on spare parts. The case studies show that the key to new IPSO business models lie in whether the supplier is able to control the physical products during the pre-use, use and post-use phases. The take-back of used products

especially has an important impact on the business model [29], with possibilities for remanufacturing. Supplier control can be exercised e.g. through kept ownership, certified user training, or via service contracts. When the ownership stays with the supplier during the use phase, a third party, e.g. a leasing company, can be the legal owner in order to finance the physical parts. Increased control over the physical artefacts during the use phase means both opportunities and risks. Below are listed some of the identified examples where the supplier has opportunities to increase control over the physical part of IPSOs. Customer’s perceived risk of ownership – Many customers would prefer to not own non-core physical products for their business. Ownership means assets on the balance sheet, and lower total capital increases the profit/total capital ratio, which is an important key performance index. Another reason is that the customer needs change, and ownership reduces flexibility to change (depending on the form of leasing contract, which may be expensive to break in advance). Since the supplier takes over the risk, this can be a way to more rapidly introduce new technology; if the technology does not work, it is the supplier’s problem and not the customer’s. If the customer needs change, the supplier can have a clause that for a fee enables the customer to change or adjust the offering. A supplier has in general an easier time than a customer to find a new user for the non-needed products, and this is an advantage for the supplier. Toyota Material Handling Group, one of the world’s leading forklift truck producers, benefits from this when they take back and trade their own or competitor’s old forklift trucks. Toyota Material Handling Group also focuses on customer-perceived risk of ownership, and uses this actively in their marketing when offering IPSOs. In their advertising they state: “Don’t buy trucks! It makes sense to use a fleet of trucks to facilitate efficient materials handling – but you don’t have to own it to use it. Think about why you buy things. You might think about buying your home - it’s an investment - and it will probably increase in value over time. You might want to own something that’s extremely special, very rare, or has sentimental value. Or something you want to keep for yourself. But why would you want to own a forklift truck? • What if your requirements change? • What else could you do with the capital? • What about peaks and troughs? • What happens when it breaks down? • What about disposal and environmental issues? • What about knowing your costs in advance? The risks of ownership – We manage them for you. (We can do it better and cheaper, it’s our business). The fact is that trucks consume capital, lose value over time, can breakdown, need maintenance, and have to be disposed of one day. The ones you need today are probably not the ones you will need tomorrow because your requirements will change.” Supplier access to physical products during the usephase – In the case of IPSOs, access to the physical product during the use-phase of the product is important for the supplier. There are several reasons for this. 1. Measure the use. This can be done in order to: a. get a base of payment; b. obtain information concerning need of maintenance; c. gain control, if correctly used; and d. improve the use (without upgrading), e.g. through better working procedures.

2. Perform maintenance. 3. Upgrade software or electronic hardware. 4. Replacement, e.g. to newer equipment. Obviously, supplier access to the physical product can be delicate during the use phase for many reasons. Use measurements may be perceived as user surveillance, threatening personal integrity. Stops for maintenance, upgrading or replacement need to be avoided if the product is used in key activities, where customer revenue depends on run-time. From the customer perspective, any operator training due to such changes also must be avoided for cost and convenience reasons. The customer may want to use the help of service providers other than the IPSO supplier in case of e.g. malfunction, especially when the IPSO is a part of a larger system. Altogether, this sets tough requirements on the supplier to be innovative both when it comes to technical solutions, e.g. for remote maintenance, knowledge about customer operations in order to minimize trouble in the use phase, and attractive maintenance contracts. Maintenance personnel or operators employed by the supplier at the customer site may be a solution. The argument is stronger if the supplier owns the product. Another sometimes applicable solution that often forces the user to accept the supplier’s access to physical products during use is through design and technical solutions, e.g. by making it complicated for the individual user to do their own maintenance. This can be done e.g. by requiring special service and maintenance equipment and tools, spare parts and consumables that only can be handled and offered by the supplier. Control over spare parts, service and maintenance – OEM spare parts, service, and maintenance are crucial for the IPSO supplier. Otherwise, other suppliers’ solutions may cause harm and danger to the use of the product. However, the difficulty lies in avoiding pirate copies (managed by patents, Intellectual Property Rights (IPRs), and laws), solutions close to the OEM suppliers from low-cost suppliers or local suppliers, which due to convenience or cost reasons may be the preferred choice of the customer. Building up the service organisation required to control the market may be costly; another solution is, again, innovative solutions that are difficult to copy, contracts, and customer relationships that foster the OEM supplier as the customer’s first choice. Rank Xerox is one of the more famous examples of IPSO suppliers that use this strategy. They have built up a system that provides control over spare parts, service and maintenance, and even have systems that can remote control customers’ printers and conduct remote maintenance. Rank Xerox can also control and in advance send out service staff before any malfunction occurs. Furthermore, they have also built up a phone support system where the customer, in the simpler cases, acts as a remote service staff member instructed by the phone support centre staff. Customer dialogue – The supplier’s need for product control during the use phase both requires and increases the opportunity for the supplier to have a constant dialogue with the customer. However, this necessitates that the customer perceives the supplier as a strategic, key supplier, since the trend has been to reduce the total number of supplier contacts [25]. Generally, the supplier must be regarded as a solution provider, expressing customer gain in terms of increased customer profit or decreased cost [30], and if so, the supplier needs to pay careful attention to any signal of customer dissatisfaction. The alert supplier will probably be one of the first to know if something is wrong, or if the customer hesitates to abandon the IPSO, which provides an opportunity to

168

improve the existing offering or to offer something new. Furthermore, if the service and maintenance is done by the supplier, it enables a closer contact and dialogue with the users of the IPSO, i.e. several points of contact at the customer side. This can be an important source of information, e.g. from the operators, about how to improve the design, or how to further customize the IPSO in order to improve customer value. According to several companies, during the use phase it is often easier to get a deeper understanding about the real needs and what the customer considers as real value. The customer’s purchaser is not always fully initiated in all needs; this may imply that user of the IPSO does not get all their needs fulfilled. Furthermore, as stated by several companies, it is preferable that this dialogue is distinguished from traditional selling situations (i.e. no purchaser staff involved). This is because customers sometimes are more cautious e.g. to fully reveal their needs, since this may be a way for the supplier to increase the price. Control over the second-hand market – When considering control over the second-hand market, it is important to remember that the focus in the IPSO concept is on providing functionality and customer value, regardless if this is performed by new or second-hand products. However, since the second-hand market exists for many products, several IPSO suppliers have experienced the crucial need to control or prevent their own second-hand market, but also often other suppliers’ products. There are several reasons for this, the most frequently mentioned of which are presented below. One reason often mentioned by suppliers is competition from companies specialized in trading with remanufactured or refurbished second-hand products (often the supplier’s own products). Those remanufactured or refurbished second-hand products are often a potential and attractive alternative for the customer, and are often manifested in a significant pressure on prices. To prevent this, companies, e.g. Toyota Material Handling in Sweden, keep the ownership of the IPSO’s ingoing products (no products go directly from customer to the second-hand market) and when providing IPSOs they often redeem the customers’ old forklift trucks (even other brands). A second reason, related to that above, is that the product’s inherent value is often quite high, even though it is replaced by newer products. However, this value depends on the owner, the owner’s knowledge about the product and the owner’s ability to take advantage of this knowledge. For example, a Toyota forklift truck has a lower value for a warehouse owner than for Toyota Material Handling in Sweden. This is because the forklift truck manufacturer, which has knowledge about e.g. remanufacturing and refurbishment can find a new customer for the used truck or use it for spare parts. At the same time, if the supplier doesn’t keep control over the secondhand market, companies may enter the market that built up the knowledge and have the ability to capitalize on this knowledge. Freebie” suppliers such as ink printer OEMs have experience with other companies which collect refills and resell the suppliers’ own ink cartridges. A third reason for control is that faulty remanufactured or refurbished products have negative influence on the customers’ apprehension of the brand. Case studies have e.g. shown that the OEM logotype and identification information is not removed when other suppliers remanufacture their products and sell them on the second-hand market. In fact, this is beyond the OEMs control and risks serious bad-will. Material and component supply – Related to previous discussion, it also implies that the supplier can build up

169

their own system for remanufacturing and refurbishment of used ingoing products and material. This can be an effective and efficient way to secure part of a company’s own component and material supply. One example is a company that has developed a patent composite (recycled thermoplastic and wood fibers) that is recyclable, and which produces various products that they can take back and reduce the need for future materials . Another example of an offering is end plugs for paper bobbins. In short, the first time they sell a plug, they need to produce the composite; in following times, they can reuse the material (only adding some new composite). This implies that their cost for material is significantly reduced. Win-win opportunities when the supplier takes over ownership – If the customer only pays for the accessibility and use of the physical product, and the supplier maintains the control, ownership or responsibility of the equipment during its use, income from spare parts, incidental materials, service and maintenance become internalized at the supplier. In other words, instead of contributing to the profit, as a cost for the supplier it burdens it. This needs to be specified in contracts, so that the supplier has access to the physical equipment, and also can control how it is being used, e.g. certified training programs for customer service, maintenance, and operator personnel. This has several important implications for the IPSO: • Focus on reducing the need for spare parts, incidental materials, service and maintenance e.g. through changed design to reduce Service and maintenance requirements, and if needed, design for ease, effectiveness and efficiency • Focus on product life-cycle issues: a desire to use ingoing material, hardware, software, and other components, spare parts and incidental materials as effective and efficient as possible. 2.3 The role of the market position This research has identified aspects to consider related to the market position, further described below. Risk of lower sales volume or quantity – Market share measures are usually based on the number of produced and sold products, or the company’s turnover as a share of total market size. Several IPSO suppliers have highlighted that this can be a problem when focusing on changes from producing and selling a large number of goods to instead delivering more customer value with less ingoing physical products, spare parts, incidental materials, etc. Fewer produced and sold products can, in the short term, decrease sales volume, since the supplier can deliver the same or more value at a lower initial price for the customer. When the customer chooses an IPSO, according to the service-dominant logic [2, 3], payment is instead distributed along the use phase. Important key performance indices may be affected, and may change the company’s market position. Risk of lower profit margin – Another problem is when the provider of an IPSO manages a network of several suppliers contributing to the complete offering [3, 16, 31], or when the IPSO includes taking over the responsibility for parts of customers’ operations [1], e.g. of a warehouse, and requires co-ordination with other suppliers. If the customer prefers one-point-of contract, this increases the supplier’s turnover. On the other hand, if the margin is low on these subcontractor contributions, this implies a lower profit margin for the IPSO supplier. Effects on company brand – IPSOs imply for many companies a major change, and the business logic is quite different from that of traditional sales. Several companies have mentioned problems explaining their IPSOs to their customers. However, it seems that market

2.4 Contract types Developing IPSOs results in a form of contracts. We have seen in practice different forms of contracts than seen in traditional product sales, which include the following types. Note that these are neither mutually exclusive nor collectively exhaustive. In addition, it should be noted that these types are categories for contracts, and different from the well-known classification of services, i.e. productoriented services, use-oriented services, and resultoriented services [32]. • Insurance contracts: The customer pays a fee to the supplier for obtaining the function that the supplier provides. This insurance guarantees access to the function instead of the specific equipment (a unique, identified object), which is usually the case in rental or financial leasing. • Rental contract: The most common rental contract is for apartments. The customer pays a fee to the supplier for using the equipment. Ownership always stays with the supplier. In the case of TMHG Sweden, the rental contract is presented in three different offerings, with more or less service included. • Financial leasing: A financial service where the customer pays an interest rate. At the end of a leasing contract period the customer usually owns the product, or has the right to buy it at very low price (terminal value). • Lease and take back: The contract stipulates that the supplier owns the hardware equipment at the end of the leasing period. Until then, the equipment is handled as if it was owned by the customer. • Pay-per-use: Customers are charged according to an amount in a predefined scale in this type of contract. This form usually requires an electronic-counting device that records the use of the hardware equipment for the supplier or a manual control (e.g. car rental, taxi). • Pay-per-hardware unit: This contract type is similar to traditional hardware selling, but the difference is that service is included in the price of the hardware unit. • Pay-per-service unit: This contract type is similar to traditional service selling (per hour or per specified service), but the difference is that hardware is included in the price of the service. • Performance-based contract: This contract type includes payment for the supplier based on the performance of the offerings at the customer. It can be associated with “profit sharing”. ESCOs (Energy Service Companies) close this type of contract particularly targeting the energy efficiency of customers. • Demonstration contract: This is a combination of a performance-based contract and a before-hand demonstration period. In the period of demonstration,

the company provides “trial” service, where the supplier takes some risk. This type is meaningful unless customers agree to the contents in a contract before their own experience. 2.5 Being solutionist vs. less risky The supplier’s fulfilment of complex offerings often requires several sub-suppliers. In such a situation it was observed, in line with previous research on business networks [7, 14, 25, 31], that the customer prefers one contract. This means that the sales volume for the IPSO supplier increases, but not necessarily the profit margin – in fact it lies in the customer’s interest that these transaction costs are low. On the other hand, the supplier takes a risk if the guaranteed function is provided through a network of partners and subcontractors, which should be rewarded through some kind of risk fee. Figure 1 is a schematic illustration of the vertical business relationship between the customer, the supplier (IPSO provider) and its partners for the specific offering in a business network. There may be several sub-tiers, i.e. suppliers to the supplier’s suppliers and partners. In the figure, the customer has one point of contract with one of the suppliers, referred to as the IPSO provider. The companies in the network provide key subsystems or key services to the IPSO, and are referred to as partners, terminology indicating the importance of a close relationship for the IPSO provider with these companies nd [14], while the 2 to nth tier suppliers are more of the component type of suppliers. Contractual business relations are indicated with arrows, i.e. delivery of service or goods from supplier to buyer and remuneration (payment) in the other direction. Customer Business relation IPSO Network

IPSO provider Business relations IPSO partner

IPSO Supply chain 2-tier suppliers

IPSO partner

IPSO partner

Business relations 2-tier suppliers

2-tier suppliers

IPSO Value Constellation

leaders or companies with leading brands can easier introduce new ISPO concepts to their customers. One important factor is most likely the credibility that comes with the leading position and the brand image that gives. A comment on this is given in [2], where the term “brand equity” is suggested to be replaced by “customer equity”, i.e. if the Service-Dominant Logic super-ordinates the Goods-Dominant Logic, customer relationships will be more important assets to a company than its brand. Access to information about competitors’ products – Remanufacturing, or merely return of used products, gives the IPSO company possibilities to gain knowledge about competitor products if any used equipment is accepted, regardless of manufacturer. This should be a large threat to manufacturers not offering then return of used products [29].

Business relations n-tier suppliers

n-tier suppliers

n-tier suppliers

Figure 1.Schematic illustration of an IPSO business network. It should be noted that integrating sub-suppliers as one package and focusing on their core competence can be contradictory. Thus, the two recent industrial trends, i.e. becoming solutionist [7] and focusing on the core competence [31], rationally lead companies to seek an optimized space. By having such optimization, a provider can stay with its own core competence and, at the same time, a customer appreciates the necessity to “assemble” supplied products and services by themselves. 3

SPECIFICATIONS FOR SUPPORT METHODS NEEDED The results, especially those described in Section 2.2, show that a set of different methods for IPSOs are needed. Methods are needed to be used by various functions within the IPSO supply chain in order to cover

170

various tasks (note that this does not mean that every single method should cover all the tasks). This is in line with existing literature, which argues that methods must be developed so that companies are supported upon designing IPSOs (e.g. [10]). This section discusses what kinds of features such methods should have, and the relative novelty of such methods.

Such methods would be useful for SMEs as well as large companies. The difference may exist in the focal parameters. SMEs might be more careful about the supplier’s risk due to the relatively smaller risk that they can bear, while large companies might be conscious about the turnover in case they pursue leadership in the market.

3.1 Goals of the methods As shown in this paper, IPSOs have a large impact on business aspects. Therefore, one type of such methods should support companies to develop their business models. Here, business development means identifying a set of values/costs over time, providers (suppliers), and properties of products/services. Those three elements exist in the level of what, who and how, respectively, if the four elements of service (what, who, why, and how) [33] are borrowed. Customers as a part of who and why are supposed to be given here, as companies do not regard identifying customers and the grounds for value/cost as a process that they strongly wish to be supported. Note that the properties of products/services are further utilized to design the physical products and service activities so that the entire body of information forms the how. How should include the information of e.g. spare parts (incl. supplied material), maintenance provided and expendable supplies.

3.5 New features of methods The set of methods to be developed as a whole have new features, as they explicitly address some key parameters described in Section 2.2., for example:

3.2 When to use methods The methods should support companies in two types of situations. Firstly, it should be helpful when companies investigate the feasibility of new types of business. Secondly, it should be powerful for companies to simulate quantitatively an offering to be proposed to a concerned customer. This requirement is not contradictory to the goals described above; i.e., the methods with those goals can be utilized in these two types of situations. 3.3 How to represent value/cost How should the value/cost, i.e. one element of the three described above, be represented? It should include profit, turnover, cost, and risk for the provider over time. Large companies might wish to know this economic information according to divisions in their company. That for the customer, on the other hand, should include qualitative representation of customer value, cost, and risk along time. Ownership of a physical product should be included in the customer value, since it is a key factor as explained previously. Furthermore, the degree of a provider’s solution for the customer, which could be calculated from the customer value and why information, may be helpful. Uncertainty causing the risk should include that of condition of physical products, quality of services, delivery of sub-suppliers, and customer requirements. Let us assume that economy (including discount rate) and future laws are to be predicted due to the current focus. 3.4 Who should use the methods IPSOs have influence on organisational structures of companies and may require a new structure. Thus, it is meaningless to discuss which one of the current sections in a company should use the methods. Therefore, this section discusses which functions of employees should be supported by the methods. One answer will be that the responsibility of using the methods should be attached to the marketing or sales function. Product development and after-sales (or service) sections should be a part of the users (i.e. contribute to utilize the methods). The reasons are that the information of products/services is addressed at a high level. Thus, the methods should work for internal communication.

171

• Feasibility of IPSO business. • Access to physical products during the use phase. • Control over spare parts. On the other hand, there are various existing methods/tools that cover some of the needed features (e.g. [33-36]). These will potentially be a part of the set of methods. 4 CONCLUDING DISCUSSION This paper attempted to illustrate business implications of IPSOs from supplier perspectives based on empirical experiences and previous research. It also highlighted several key strategic issues such as company flexibility, risks and opportunities of IPSOs, market positions, and contract types. Then, based on key issues derived from the discussions, a set of methods was suggested, to be constructed to help companies with developing their IPSOs. The next step is further deployment of those features into more detailed descriptions, and development of the set of methods. Results from the case studies show that both small and large companies that conduct the transition towards IPSOs face several important strategic challenges, some of them associated with high risk. There are also, as expected, apparent differences due to company size, both pros and cons for the larger and the smaller company. The results from the supplier studies thus also highlight important aspects for a customer-selecting provider of IPSOs. 5 ACKNOWLEDGMENTS The authors of this paper would like to thank all participating companies for their support. This research was partially supported by the Swedish Governmental Agency for Innovation Systems (VINNOVA), the Swedish Association of Graduate Engineers (Sveriges Ingenjörer) and the Swedish Environmental Protection Agency. 6 REFERENCES [1] Oliva, R. and R. Kallenberg (2003) Managing the transition from products to services. International Journal of Service Industry Management, Vol. 14(No. 2): 160-172. [2] Vargo, S.L. and R.F. Lusch (2004) Evolving to a New Dominant Logic for Marketing. Journal of Marketing, Vol. 68(No. 1): 1-17. [3] Vargo, S.L. and R.F. Lusch (2008) From Goods to Service(s): Divergences and Convergences of Logics. Industrial Marketing Management, 37: 254259. [4] Goedkoop, M.J., C.J.G. Van Halen, H.R.M. te Riele, and P.J.M. Rommens (1999) Product Service Systems, Ecological and Economic Basics. VROM: Hague, the Netherlands.

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

Lindahl, M., G. Ölundh Sandström, E. Sundin, A. Öhrwall Rönnbäck, and J. Östlin (2008) Learning networks: a method for Integrated Product and Service Engineering – experience from the IPSE project. 41st CIRP Conference on Manufacturing Systems – Manufacturing Systems and Technologies for the New Frontier. The University of Tokyo, Tokyo, Japan. Tomiyama, T., Y. Shimomura, and K. Wantanabe (2004) A Note on Service Design Methodology, in ASME 2004 Design Engineering Technical Conferences and Computers and Information in Engineering Conference: Salt Lake City. Allmendinger, G. and R. Lombreglia (2005) Four Strategies for the Age of Smart Services. Harvard Business Review, 83(10): 131-145. Morelli, N. (2006) Developing new product service systems (PSS): methodologies and operational tools. Journal of Cleaner Production, 14(17): 14951501. Sakao, T. and Y. Shimomura (2007) Service Engineering: a novel engineering discipline for producers to increase value combing service and product. Journal of Cleaner Production, (No. 15): 590-604. Sakao, T., G. Ölundh-Sandström, and D. Matzen (Accepted) Framing Design Research for Servicification through Extending PSS-Design Research. Journal of Engineering Design (Special Issue about Product Service Systems). Isaksson, O., T. Larsson, and A. Öhrwall Rönnbäck (Submitted july 2008) Development of ProductService Systems: Challenges and Opportunities for the Manufacturing Firm. Journal of Engineering Design. Normann, R. and R. Ramirez (1993) From value chain to value constellation: Designing interactive strategy. Harvard Business Review, 71(4): 65-78. Normann, R. (2001) When the Map Changes the Landscape. West Sussex, England: John Wiley and Sons Ltd. Gummesson, E. (2002) Total relationship marketing – marketing management, relationship strategy and CRM approaches for the network economy (2:nd editon). Cornwall.: Butterworth Heinemann. Grönroos, C. (2000) Service Management and Marketing: A Customer Relationship Management Approach. Chichester: Wiley. Tukker, A. and U. Tischner (2006) Product-services as a research field: past, present and future. Reflections from a decade of research. Journal of Cleaner Production, 14: 1552-1556. Eisenhardt, K.M. (1989) Building Theories from Case Study Research. Academy of Management review, 14(4): 532-550. Yin, R.K. (1994) Case Study Research - Design and Methods. 2nd ed. Thousand Oaks, California: SAGE Publications. European Union (2003) Commission recommendation of 6 May 2003 concerning the definition of micro, small and medium-sized enterprises. Official Journal of the European Union, (L 124/36). Kvale, S. (1983) The qualitative research interview a phenomenological and a hermeneutical mode of understanding. Journal of Phenomenological Psychology, 14: 171-196.

[21] Lagerholm, B. and A. Öhrwall Rönnbäck (2003) IKT för funktionsförsäljning i nya värdeskapande nätverk: Nya möjligheter med informationsoch kommunikationsteknik (IKT). En studie av affärsstrategier för ökad lönsamhet och livskraft i svensk verkstadsindustrislutrapport (in Swedish) Final project report VINNOVA (ICT for functional sales in value-added networks). [22] Sakao, T. and E. Sundin (2009) Analysis of Integrated Product and Service Offerings from current perspectives of providers and customers, in CIRP IPS2 Conference 2009: Cranfield. [23] Windahl, C., P. Andersson, C. Berggren, and C. Nehler (2004) Manufacturing Firms and Integrated Solutions - Characteristics and Implications. European Journal of Innovation Management, Vol. 7(No. 3): 218-228. [24] Lindahl, M., T. Sakao, E. Sundin, and Y. Shimomura (2009) Product/Service Systems Experiences – an International Survey of Swedish, Japanese, Italian and German Manufacturing Companies, in CIRP IPS² Conference 2009: Cranfield. [25] Day, G. (2004) Achieving Advantage with a New Dominant Logic. Invited Commentaries on “Evolving to a New Dominan Logic for Marketing”, Bolton R. N. (ed), Journal of Marketing, Vol. 68: 18-19. [26] Sakao, T., N. Napoletano, M. Tronci, E. Sundin, and M. Lindahl (Accepted) How Are Product-Service Combined Offers Provided in Germany and Italy? – Analysis with Company Sizes and Countries. Journal of Systems Science and Systems Engineering. [27] Fundin, A. (2005) Dynamics of quality attributes over life cycles of goods and services. Chalmers University of Technology. Gothenburg. [28] Ölundh, G. and S. Ritzén (2003) How do functional sales affect product development and environmental performance? International Conference on Engineering Design, ICED 03. Stockholm. [29] Östlin, J., E. Sundin, and M. Björkman (2008) Importance of closed-loop supply chain relationships. International Journal of Production Economics. [30] Anderson, H., D. Andersson, P. Brehmer, M. Huge, J. Lilliecreutz, and A. Öhrvall Rönnbäck (2004) Suppliers’ Articulation of Value Using the Internet, in Conference Proceedings, International Marketing and Purchasing (IMP) Conference: Budapest. [31] Prahalad, C.K. (2004) The Co-creation of Value. Invited Commentaries on “Evolving to a New Dominan Logic for Marketing”, Bolton R. N. (ed), Journal of Marketing, Vol. 68: 23. [32] Tukker, A. and U. Tischner (2006) New Business for Old Europe. Sheffield: Greenleaf Publishing. [33] Sakao, T., Y. Shimomura, E. Sundin, and M. Comstock (Accepted) Modeling Design Objects in CAD system for Service/Product Engineering. Computer-Aided Design. [34] Alonso-Rasgado, T. and G. Thompson (2006) A rapid design process for Total Care Product creation. Journal of Engineering Design, Vol. 17(No. 6): 509-531. [35] Arai, T. and Y. Shimomura (2005) Service CAD System - Evaluation and Quantification. Annals of the CIRP, Vol. 54/1, (ISSN 1660-2773): 463-466. [36] Meier, H. (2004) Lifecycle-based Service Design for innovative business models. Annals of the CIRP, 53(1): 393-396.

172

Innovative Service-Based Business Concepts for the Machine Tool Building Industry 1

S. Biege1, G.Copani2, G. Lay1, S. Marvulli2, M. Schröter1 Fraunhofer Institute for Systems and Innovation Research ISI, Breslauer Strasse 48, 76139 Karlsruhe, Germany 2 ITIA-CNR – Institute of Industrial Technologies and Automation, Via Bassini 15, Milan, 20133, Italy [email protected]

Abstract During the last decade, machine tool building companies have been forced to put innovative offers on the market. Due to the technical features of their products and the prevailing organizational structures in this sector, especially product-service systems are a promising way of creating a unique selling point. In this paper, potential new business concepts for machine tool builders will be presented which aim at fulfilling basic customer needs like the increase in quality, flexibility, productivity and the reduction of lead times, costs and risks. For the implementation of these product-service systems, practical examples are given. Keywords: Machine Tool Building Companies, Service-Based Business Concepts, Product-Service Systems

1 INTRODUCTION During the last decade, machine tool building companies have been facing a turbulent economic environment. Especially challenges arising from the growing competitiveness of Asian business rivals [1] force machine tool builders to launch innovative offers on the market with the purpose of strengthening customer relationships and generating long-term customer loyalty. Due to the technical features of the products and the organizational structures in this sector, especially product-service 2 systems (PSS, IPS ) are a promising way of creating a unique selling point. A range of generic business concepts based on productservice systems has been developed for these conditions, comprising high quality goods and product-related services. In this paper, six potential new business concepts for machine tool builders are presented which aim at fulfilling basic customer needs. Furthermore, it will be shown how based on promising value propositions winwin situations can be generated by new concepts of configuring the sharing of tasks between suppliers and customers and the performance of high class service delivery. For the implementation of the depicted productservice systems, practical examples from machine tool building companies from the European industry are given. The objective of this paper is to give a comprehensive overview about potential service-based business concepts for machine tool building companies. The basis for offering business concepts in the machine tool building industry are product-service systems adapted to the customers’ requirements [2]. Product-service systems are a widely accepted concept, in the field of research as well as in practice. They can be defined as a mixture of “tangible products and intangible services designed and combined so that they jointly are capable of fulfilling specific customer needs” [3] and are regarded as a powerful source of innovation in the sector of machine tool building companies [4, 5, 6]. This mature sector is of high importance for Europe and the European machine tool production and accounted for more than 19,000 million euros turnover in 2006 [7]. Europe

CIRP IPS2 Conference 2009

173

continues to be world leader in the machine tool industry, but after the collapse of the economical-political barriers and the globalization process triggered by the diffusion of new technologies, in the last decades new competitors from emerging economies have started approaching the market offering cheaper machines. As the automotive industry, in which a fundamental competition driver is cost, is the major customer of machine tool builders, European companies have lost significant market shares. Europe’s difficulties in this sector are clearly outlined by the constantly increasing gap between European production of machine tools and global production figures. Furthermore, although exports to developing countries are increasing, the ones to mature economies are on the decrease. Low cost producers from emerging markets are gaining more and more market shares [7]. Whilst for countering the market entry moment of emerging competitors, European companies could rely on superior quality as well as technology of their products, the technological and managerial gaps separating them from e.g. Asian companies have been reduced and no longer constitute a real protection for the future of this industry in Europe. The development of innovative product-service strategies will thus be a strategic response to the global market challenges. Especially in the manufacturing sector, companies increasingly have to adapt their business concept to the newly arising challenges depicted above and hence increase the service content of their offers. Productservice systems in the machine tool building sector have several advantages for machine tool builders and their customers. While manufacturers can stand out from the competition, particularly against their low cost rivals in business, they can achieve customer loyalty by developing solutions in a customer-oriented way and hence create a greater value for their customers [2]. By means of the new business concept customers can benefit from customized solutions, increase quality, flexibility and productivity and reduce risks, costs and lead times at the same time. A widely used categorization scheme [8, 9, 10] for product-service systems was compiled by Tukker [11]. In

this approach, they are divided into three subcategories sorted by decreasing product content and increasing service content (figure 1): •





Product-oriented services are strongly connected with the product. The property rights of the physical assets are transferred to the customer while the machine tool builder adds extra value to the offer by arranging services around this product. Especially technical services have a strong product relationship [12], yet the interdependencies between the product and the related service are generally weak [8].

the actors who are involved in the business transaction as well as their roles and their contribution to creation of value („value chain configuration“) and



the payment model of the business transaction (“revenue model”).

Value Proposition

In use-oriented product-service systems, the property rights of the physical product remain at the manufacturing company which sells the use of this equipment via concepts like pooling, leasing or sharing, making the physical product available for the production of one or several users.

Revenue Model

Result-oriented product-service systems are an approach focusing on the result of production or services whilst disregarding the underlying product. As in use-oriented product-service systems, the property rights of the product the concept is based on are retained by the manufacturing company. The provider of the result can freely decide on how the result is produced.

Service content (intangible) Product content (tangible)

Pure Product

Product Oriented

UseOriented

ResultOriented

Value mainly in service content

Pure Service

Figure 1: Main categories of product-service systems [11]. 2

SERVICE-BASED BUSINESS CONCEPTS MACHINE TOOL BUILDING COMPANIES As shown in figure 2, a business concept defines

Value Chain Configuration

Figure 2: Elements of Business Concepts [13]. In the machine tool building industry, product-service systems can serve as a basis for new business concepts. Combining machine tools with services can generate value for the customer and for the manufacturer as well. Customized product-oriented, use-oriented or resultoriented product-service systems pose as the value proposition offered by machine tool builders to their customers. The cooperation process between the machine tool builder and the customer companies in new business concepts based on product-service systems affects the value chain and thus describes the novel share of work. These cooperation scenarios have several dimensions which vary depending on the characteristics of the sharing of tasks in the business concepts. In order to explicitly consider these dimensions with their corresponding characteristics while designing a productservice system, the morphological box in figure 3 on the basis of Lay et al. [14] is proposed. The comprised dimensions are ownership of the product during the use phase and ownership after the use phase, location of the technical equipment, operation personnel and maintenance personnel as well as the number of customers served with the business concept. The revenue model is also included in the dimension payment mode.

PRODUCT-SERVICE SYSTEM Value mainly in product content





FOR

the value generated for customers and other actors in a business transaction (“value proposition”),

Characteristic features

Options

Operating personnel

MTB

Operating Joint Venture

Maintenance personnel

MTB

Location Payment modus Ownership

During phase of use After phase of use

Single/multi customer operation

MTB Pay per Part

Pay per use

Third party

Customer

Operating Joint Venture

Third party

Customer

Third party

“Fence to Fence” to the customer

Customer

Pay for availability MTB

MTB

MTB

MTB MTB

In parallel operation for multi customer

Pay for equipment

Fixed rate

Leasing bank

Leasing bank

Customer

Customer

Operation for a single customer

Figure 3: Morphological box on product-service systems for machine tool building companies (MTB).

174

In the machine tool building industry, several kinds of key players involved in the supply chain of a new business concept can be thought of. First of all, the machine tool building company is an actor, offering product and/or services to the customer which is the second actor. A third stakeholder in a new business concept based on productservice systems can be a leasing bank supporting the machine tool builder or customer in the financial matters of the transaction. Beyond these, a potential scenario in the machine tool building industry is the foundation of a joint venture formed between two or more actors. Furthermore, a third party can as well be involved in the business concept. As an example, a temporary employment agency can provide staff to operate or maintain machinery and equipment. The traditional business concept can be found on the right side of the morphological box. Here, the manufacturing company sells the machinery to its customer. The equipment is operated on the customer’s site. The customer pays for the equipment, and is responsible for manufacturing operations as well as for maintaining the machine. The amount of products manufactured meets only the needs of one customer, who is the owner of the machinery. The concept of outsourcing of entire business processes to equipment suppliers is represented by the characteristics depicted on the far left side of the morphological box. In this case, the equipment producer retains ownership of the equipment, employs operating and maintenance personnel, implements the equipment in its production plant, produces parts on this equipment for multi customers and is paid per part. Suitable customer-oriented business concepts based on the combination of products and services can be designed evaluating the technological particularities of machine tools, together with the manufacturer’s and the customer’s organizational structure requirements, like the need for reduction of costs, risks and lead times or increase in flexibility, quality and productivity. Based on the results of large scale surveys and on the findings of case studies [15], several value propositions in the machine tool sector were categorized and described in terms of machines and related services. In this paper, the six most significant of these value propositions are presented below, structured according to the categories of product-service systems cited above and described with regard to the characteristic features introduced in the morphological box. 2.1 Product-oriented Product-Service Systems for Machine Tool Builders Availability guarantee The availability of the production system is a crucial task for an industrial company because it may have impact on the total lead time, the production volume to be realized and the costs related to production loss and the eventual recourse to outsourcing. In situations in which the customer company is not able to reach adequate performance and achieve the needed equipment availability, the machine tool builder can solve this problem by offering an availability guarantee with lower costs and time-saving effects. Leveraging its specialization in maintenance activities, the supplier is able to guarantee higher availabilities on better conditions compared to those of the customer. In this case, the planning of preventive maintenance as well as the entire maintenance strategy can be performed more efficiently, leading to a lower repair frequency (MTBF) and costs. The know-how and specialization in maintenance enables the respective staff to develop competences and skills to allow lower repair time (MTTR) and a more efficient consume of the materials needed for the maintenance,

175

both factors leading to costs savings. The end user is charged for the availability of the production system. The higher the guaranteed availability, the higher the costs for the customer will be. For this value proposition, a cooperation scenario in which the property rights of the machinery are transferred to the customer can be appropriate. The machine tool builder maintains the production system and is paid based on the availability it is able to guarantee to the customer. This concept can therefore be counted among the category of productoriented product-service systems. Solving customer qualification deficits Companies acting in industrial sectors with up-to-date technologies and rapid technological change may hesitate to invest in new production systems. One reason might be the lack of sufficiently qualified personnel. If the customer's workforce does not have the required skills to operate and maintain modern machinery, this can result in the management’s decision to forgo capital-intensive investments and to continue working with an out-dated machine. This attitude can be observed particularly in developing and emerging countries. This is where new service-based business concepts can be applied. A machine tool builder can sell its advanced production system in combination with the offer to solve the customer's qualification deficit either with a training solution or a production service solution. In the first case, a professional training for the customer's workforce is conducted by skilled personnel of the machine tool building company before the implementation and during the operation of the new production system, especially in the ramp-up phase. This business concept is based on a product-oriented product-service system. The alternative solution, a temporary production service offer can take place at the customer’s plant on a short- to mid-term level, as mentioned above during the equipment ramp-up and in the initial production phase by providing the customer’s facilities with qualified personnel that runs the machines. This can be categorized as well as a product-oriented product-service system. Alternatively, a production service in terms of outsourcing entire processes can be offered which is described below. Reconfigurable production systems The concept of reconfiguration can be described as the technical customization of the equipment to technological change and varying customer requirements. Especially fields in which machinery is in use for long periods of time but is needed to manufacture many different product variants and consequently has a high retooling frequency have to be mentioned in this context. Furthermore, unpredictable market demands and requests force the industrial companies to be as flexible as possible in order to respond to them at the right time and with low costs. Customers often do not want to invest in the latest equipment since they fear the high investment costs. A manufacturing company can offer a modernization concept to customers based on reconfiguration and refurbishment of modules or entire production systems. Thereby, the machine tool builder offers to modify rapidly and economically in terms of targeted production volume and mix. Offering such a service for a production system allows the companies to be flexible, to reduce costs and time needed to respond to the market and to reduce the technological and operative risks related to the use of static machines. This cooperation scenario is characterized by the ownership of the end user and a production system addressed to increase its performance and efficiency level which can be very low in out-dated machinery. The business concept is based on a productoriented product-service system.

Lean machine business concepts Many up-to-date production systems offer a multitude of features customers can use for the production of advanced products and a multiplicity of product variants. Yet, there are cases in which customers do not need these advanced machines with many functions and with a high performance. For the manufacturing of basic products or components, especially in low-tech industries, simple machines with a restricted number of functionalities are sufficient. In this case, the use of overdimensioned and oversized machinery may result in high production costs, since the investment and maintenance costs for a production system with a bulk of special features are higher than those for a basic model. The outcomes of this are high production costs per unit. Customers aiming at purchasing a new production system could be offered an alternative to high-end solutions, i.e. the "lean machine" concept with only those functions needed and consequently with stripped-down equipment. This cooperation scenario is characterized by the ownership of the end user and therefore the basis of this business concept is a product-oriented product-service system. 2.2 Use-oriented Product-Service Machine Tool Builders

Systems

for

Concepts for levelling irregular and temporary customer capacity requirements In cyclical industries, but also in make-to-order production and in manufacturing of prototypes, sample parts and trial parts or in cases of high market risks and uncertainties, the capacity utilization of the machinery is low on the average. However, these types of production are often also characterized by significant capacity variances. In this situation, customers have two possibilities: they can either provide capacities that cover capacity peaks and therefore may have unexploited capacities during the production time which results in high costs per unit. The alternative is to dimension the production system for smaller capacity requirements. In this case they may risk delays in the production process. In the worst case orders might have to be rejected. This is where the value proposition of levelling capacity requirements applies. A machine tool builder can provide the customers with a mobile production system in times of capacity peaks. In this case, the customers can either pay a fixed rate depending on the time the machine is needed or the payment is based on the number of operations carried out on the mobile machinery. This product-service system can be regarded as a use-oriented one. 2.3 Result-oriented Product-Service Machine Tool Builders

Systems

for

Concepts for levelling irregular and temporary customer capacity requirements As an alternative for levelling customer capacity requirements, the machine tool builder can also run a machine similar to the customer's one and produce parts simultaneously for him. The customer situations described above vary between temporarily restricted time spans, e.g. in the ramp-up phase of a product variant and can last up to the entire life span of a production machine in seasonal customer industries. If producers of machine tools succeed in increasing capacity utilization by means of offering service-based business concepts so that their customers merely have to invest in machinery and equipment to satisfy the average demand whereas demand peaks are covered by the manufacturer’s facilities, then economic advantages can be achieved for both parties. The parts produced by the machine tool builder are sold to the customer on a pay-per-unit basis. The manufacturer might be able to bundle orders from

several customers and benefits from economies-of-scale effects. For this value proposition, a cooperation scenario in which the machine tool builder owns, operates, maintains the production system can be proposed. The customer pays based on the realized production. The location of the operation is not fixed in this cooperation scenario and therefore all options are open. In case of an unstable and short-term production, the manufacturer can operate simultaneously for more than one customer in such a cooperation scenario. Therefore, this product-service system can be regarded as a result-oriented one. Production service Due to their specific knowledge of the machinery, in certain cases machine tool suppliers can produce higher quality with lower costs in less time than the end users producing autonomously at their own plants and hence turn into a part supplier. The reason for this might lie in the fact that with time the machine tool builders are able to acquire those competences and skills that enable them to obtain higher performances in a particular type of production, not least owing to economies of scale and learning curve effects. In these cases, the suppliers can guarantee the equipment availability for long periods to their customers, running the production system and producing for them whilst being responsible for the manufacturing staff as well as for maintenance and location of the production system. On the other side, the end user can externalize the production to the supplier, purchasing the end products directly from the supplier, paying per unit. In this way, risks can be reduced and the customer’s flexibility increases. In unpredictable production situations, the machine tool builders can operate simultaneously for more than one customer using such a cooperation scenario, realizing the economies of scale mentioned above. Offering a production service ranks among the result-oriented product-service systems. All product-service systems depicted above can be found in Tukker’s categorization scheme in figure 4.

ProductOriented • Availability guarantee • Solving customer qualification deficits • Reconfigurable production systems • Lean machine business concepts

UseOriented

• Levelling irregular and temporary customer capacity requirements

ResultOriented

• Levelling irregular and temporary customer capacity requirements • Production Service

Figure 4: Product-service systems for machine tool building companies. 3

PRACTICAL EXAMPLES FROM THE MACHINE TOOL BUILDING INDUSTRY In the following paragraphs, examples for potential business concepts based on the three types of productservice systems will be depicted.

176

Characteristic features

Options

Operating personnel

MTB

Operating Joint Venture

Third party

Customer

Maintenance personnel

MTB

Operating Joint Venture

Third party

Customer

Location

MTB

Third party

“Fence to Fence” to the customer

Customer

Payment modus Ownership

During phase of use After phase of use

Pay per Part

Pay per use

Pay for availability

MTB

MTB

MTB

MTB MTB

Single/multi customer operation

In parallel operation for multi customer

Pay for equipment

Fixed rate Leasing bank

Leasing bank

Customer

Customer

Operation for a single customer

Figure 5: Morphological box for product-oriented product-service system case study. tools are needed to assist the company in the transition process. The morphological box with the characteristic features of the training business concept of this case study company can be found in figure 5.

3.1 Practical examples for product-oriented productservice systems for machine tool builders: Solving Customer Qualification Deficits The case study company is a large company operating worldwide as one of the international market leaders in designing, producing, maintaining and supplying equipment and components for the railway industry. It consists of three divisions which up to now independently offer training to their customers. A basic training is offered to solve their customers’ qualification deficits during the installation of a machine or in the end of this phase, in the beginning of the testing and ramp-up phase. It mainly consists of a verbal explanation of the instruction book and the programming of the machinery. This training takes place at the customer’s facilities and is conducted by a technician of the case study company. The price for the training is included in the price of the machine. Additionally, for machines with new technologies or for special parts that are supposed to be produced, special training courses are offered as well. They are conducted according to the customer’s requirements at the customer facilities or at the manufacturer’s facilities. The price for the training is negotiated in every additional training case. For the future, alternative ways of giving training to the customers are assessed as the case study company plans to outsource the training activities to a subsidiary company which will conduct training for the three divisions of the company, bundling resources and realising economies of scale. However, in case of a change of the supply of training, in the case study company, its divisions and in the subsidiary company a multitude of processes needs to be modified. Furthermore, a decision has to be made on the hierarchical integration of the subsidiary into the corporate group. Hence, guidelines and supportive

3.2 Practical example for use-oriented product service systems for machine tool builders: Levelling irregular or temporary customer capacity requirements The case study company is a small enterprise specialized in electrical discharge machines, like sinking machines, wire electrical discharge machines and filtration units. The main customers are companies producing parts for electronic and mechanical applications as well as for filtration and process technologies. Furthermore, the customer companies which are served by the case study company range from very small enterprises with less than ten employees to multinational corporate groups. In addition to the traditional sale concept the company offers their machines in a new business concept to increase its market share. Customers are expected to benefit from this offer as they can use the electrical discharge machines without having to buy them since the property rights of the machines either stay with the producing company or are transferred to a third party, in this case a leasing bank. If a customer is interested in renting the machinery instead of buying it, a leasing bank is contacted. After verifying the reliability of the customer, the bank buys the equipment from the manufacturer and leases it to the customer. The machine tool builder is paid for the equipment by the bank while the customer pays monthly or quarterly fees to the leasing bank.

Characteristic features

Options MTB

Operating Joint Venture

Third party

Customer

Maintenance personnel

MTB

Operating Joint Venture

Third party

Customer

Location

MTB

Third party

“Fence to Fence” to the customer

Customer

Operating personnel

Payment modus Ownership

During phase of use After phase of use

Single/multi customer operation

Pay per Part

Pay per use

Pay for availability

MTB

MTB

MTB

Fixed rate MTB MTB

In parallel operation for multi customer

Leasing bank

Leasing bank

Pay for equipment Customer

Customer

Operation for a single customer

Figure 6: Morphological box for use-oriented product-service system case study.

177

Characteristic features

Options

Operating personnel

MTB

Operating Joint Venture

Maintenance personnel

MTB

Location Payment modus Ownership

During phase of use After phase of use

Single/multi customer operation

MTB Pay per Part

Pay per use

Third party

Customer

Operating Joint Venture

Third party

Customer

Third party

“Fence to Fence” to the customer

Customer

Pay for availability

Leasing bank

MTB MTB

MTB

MTB MTB

In parallel operation for multi customer

Pay for equipment

Fixed rate

Leasing bank

Customer

Customer

Operation for a single customer

Figure 7: Morphological box for result-oriented product-service system case study. When the contract expires, the customer company has to decide whether the equipment will be bought from the bank for the residual value or not. If not, the machine tool builder has the chance to buy the equipment back from the bank, to refurbish it and to put it onto the second hand machinery market. In this renting concept, the operating personnel are provided by the customer like in the traditional business concept, but the maintenance is carried out by the manufacturing company. The machine is operated in the facilities of the customer company for only a single customer. Like in the product-oriented product-service system case study, a need for tools and guidelines can be derived. The machine tool building company needs hints on how to deal with financial matters. Furthermore, like in the case described above, when a third party, in this case the leasing bank, is involved additionally besides the customer, processes change. For this transition guidelines are needed as well to make the changes visible and to support the manufacturing companies. The morphological box with the characteristic features of the renting business concept of this case study company can be found in figure 6. 3.3 Practical example for result-oriented productservice systems for machine tool builders: Production Service The result-oriented business concept case study company is a SME producing milling and grinding production systems for precision mechanical components for the aerospace industry. As these products are large, complex systems and have an advanced technology level, they are sold at high prices. To also serve customers which cannot afford this equipment, the company offers a production service as a new business concept to its customers. This leads to a win-win situation as the customer can access an advanced technology for which the machine tool building company owns specific know-how. Parts are produced at the facilities of the customer on refurbished equipment. The company’s personnel are responsible for operation as well as for maintenance of the machinery. The company is paid according to the results of the production, the number of parts produced. In order to reduce the machine tool builder’s risk, the customers have to sign contracts with minimum purchase agreements. After a predefined period of pay-per-part production, in which customers can acquire experience of the process and test the market potential, customers have the option to buy the machines, to renew the pay-per-part contract or to re-negotiate it. In this case study, it became clear that machine tool building companies which plan to implement a product-

service system like the one described above need assistance in evaluating the profitability of the production service, considering the costs and revenues during the entire life cycle of the business concept. The morphological box with the characteristic features of the production service business concept of this case study company can be found in figure 7. 4 CONCLUSIONS Product-service systems are widely spread in the machine tool building industry. Although ways of offering use-oriented and result-oriented combinations of products and services to stand out from competition have been investigated in literature, product-oriented product-service systems, especially in the form of product-related services, still prevail in practice. The reasons for this can be seen in the increasing complexity of use and resultoriented concepts requiring the integration of products and services already in the early stages of development [8] and hence in the lack of appropriate instruments for machine tool building companies supporting them in the transition process from offering the traditional business concept of selling the products with obligatory services to the creation of new relationships between manufacturers and their customers. From the analysis of the case studies, it became clear that tools and guidelines to support machine tool builders in this process are needed, especially with regard to the selection of the appropriate business concept as well as concerning financial issues, technological features of the products and organizational questions. 5 ACKNOWLEDGEMENTS The research results presented in this paper stem from the integrated project “NEXT” funded by the European Commission within the Sixth Framework Programme (IP 011815). 6 REFERENCES [1] Meier, H., Völker, O., 2008, Industrial ProductService-Systems – Typology of Service Supply Chain for IPS2 Providing, The 41st CIRP Conference on Manufacturing Systems, 485-488. 2] Meier, H., Uhlmann, E., Kortmann, D., 2005, Hybride Leistungsbündel. Nutzenorientiertes Produkt-verständnis durch interferierende Sach- und Dienstleistungen, wt Werkstattstechnik online, 95: 528-532. [3] Tischner, U., Tukker, A., 2006, Product-services as a research field: past, present and future.

178

[4]

[5]

[6]

[7]

[8]

[9]

Reflections from a decade of research, Journal of Cleaner Production, 14: 1552-1556. Davis, A., 2004, Moving Base into High-Value Integrated Solutions: A Value Stream Approach, Industrial and Corporate Change, 13: 727-756. Wise, R., Baumgartner, P., 1999, Go Downstream – The New Profit Imperative in Manufacturing, Harvard Business Review, 77: 133-141. Oliva, R., Kallenberg, R., 2003, Managing the transition from products to services, International Journal of Service Industry Management, 14: 160172. European Committee for Co-operation of the Machine Tool Industries, 2007, Production of Machine Tools in the CECIMO Countries, http://www.cecimo.be/content/default.asp?PageID=1 01. Welp, E.G., Meier, H., Sadek, T., Sadek, K., 2008, Modelling Approach for the Integrated Development of Industrial Product-Service Systems, The 41st CIRP Conference on Manufacturing Systems, 2008: 525-530. Azarenko, A., Roy, R., Shore, P., Shehab, E., Tiwari, A., 2007, Technical Product-Service Systems: Business Models for High Precision Machine Tool Manufacturers, Proceedings of the 5th International Conference on Manufacturing Research (ICMR 2007), September 11-13, 2007, Leicester, UK.

179

[10] Sundin, E., Östlin, J., Rönnbäck Öhrwall, A., Lindahl, M., Öhlund Sandström, G., 2008, Remanufacturing of Products used in Product Service Systems Offerings, The 41st CIRP Conference on Manufacturing Systems, 537-542. [11] Tukker, A., 2004, Eight types of product-service system: eight ways to sustainability? Experiences from SusProNet, Business Strategy and the Environment 13: 246–260. [12] Aurich, J., Fuchs, C., Wagenknecht, C., 2006, Life Cycle Oriented Design of Technical Product-Service Systems, Journal of Cleaner Production 14: 14801494. [13] Lehmann-Ortega, L.; Schoettl, J.-M., 2005, From Buzzwords to Managerial Tool: The Role of Business Models in Strategic Innovation. CLADEA, October 22-24, Santiago del Chile, Chile. [14] Lay, G., Meier, H., Schramm, J., Werding, A., 2003,: Betreiben statt verkaufen – Stand und Perspektiven neuer Geschäftsmodelle für den Maschinen- und Anlagenbau, Industriemanagement, 2003 (4): 9-14. [15] Copani, G., Molinari Tosatti, L., Lay, G., Schröter, M., Bueno, R., 2007, New Business Models Diffusion and Trends in European Machine Tool Industry, Proc. 40th CIRP International Manufacturing Systems Seminar 2007.

A Method to Analyze PSS from the Viewpoints of Function, Service Activity, and Product Behavior 1 1 2 T. Hara , T. Arai , Y. Shimomura 1 2

Dept. of Precision Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, Japan Dept. of System Design, Tokyo Metropolitan University, Asahigaoka 6-6, Hino-shi, Tokyo, Japan [email protected]

Abstract As our economy matures, combination of tangible products and intangible services becomes a key issue toward a harmonious balance with economic growth and environment conscious. This paper aims to present a method for analyzing structures of service processes described in the modeling method the authors have proposed. It includes three indices of service delivery process according to customer satisfaction elements: (1) visibility to receiver, (2) interactivity with receiver, and (3) degree of receiver participation. Through an application case study, it is found that the method can indicates the features of services, and contributes to acquirement of clues for improving services. Keywords: Product/Service-System, Service delivery process, Service blueprint, Service modeling

1 INTRODUCTION Service is attracting increasing attention as manufacturing industries shift from being simple sellers of products to being service providers. To serve this need, the engineering target that needs to be analyzed and designed is shifting from simple products to service offerings. Product/Service-System (PSS) [1] is a specific type of value proposition that a business can offer its clients, comprising a mélange of tangible products and intangible services designed and combined so that they are jointly capable of fulfilling the customer needs [2]. Its viewpoints from business models may create much value onto products. The authors have been researching Service/Product Engineering (SPE) [3] [4] [5] to develop the PSS since 2002. It is characterized as top-down approach of service definition and representation. It has a great advantage in computer aided design system as the theory on service must be implemented in the computer to prove its effectiveness. In this paper, a modeling method for PSS in SPE research is presented; it describes services from the viewpoints of function, human activity, and product behavior [5]. The method is useful in that it provides us with visual understanding and sharing of services. However, there has been little discussion on the method to analyze and evaluate the modeled services for improving them. This paper presents another method for analyzing structures of service delivery process described in the modeling method. The rest of this paper is organized as follows: Section 2 first describes the modeling method applied in this paper. Second, it describes the relationship between customer satisfaction elements and service process elements through functions. Section 3 illustrates a method for analyzing service process by introducing three indices: visibility to customer, interactivity with customer, and degree of customer participation. Then, an

CIRP IPS2 Conference 2009

180

application case study of the method presented herein is discussed in Section 4. Section 5 concludes the paper. 2 MODELING METHOD FOR SERVICES 2.1 Overview of modeling method In this paper, service is defined as a deed between a service provider and a service receiver to change the state of the receiver [3]. This definition is broader than typical definitions encountered in traditional management and marketing fields with the obvious difference from products. They emphasize the characteristics of intangibility, heterogeneity, perishability, and simultaneity (e.g., [6] [7]). According to the definition, most business activities are services, including selling physical products. Services to be targeted in this study correspond to PSS (designed to change the state of the receiver), while a pure service (that only comprises human activity) is called a service activity. Figure 1 shows a schematic illustration of service elements and the modeling method applied in this paper (based on [5]). Elliptical nodes represent customers and service entities such as humanware, hardware, and software. Here, software is any component such as the computational code, policies, norms, rules, procedures, practices and any other formal or informal rules that define the way in which the system components interact [8]. In this paper, software is grouped with hardware or humanware: software is either related to hardware or humanware. Rectangular nodes represent service elements such as customer value, functions, and processes to analyze and design the relationships between customers and actual entities. Service activities are tasks performed by humanware and its related software, and product behaviors are tasks performed by hardware and its related software.

The modeling method for services is detailed below, starting with customer and customer value. Scenario using Persona

Why ?

Receiver State Parameter (RSP)

What ?

View model • Function structure for RSP • Notation: Functional tree Legend : “is-realized-by” : entity HuW: Humanware HaW: Hardware SW : Software

How ?

Extended Service blueprint • Service delivery process • Notation: BPMN Activity blueprint Behavior blueprint

Figure 1: Schematic illustration of the proposed method for modeling services.

2.2 Receiver State Parameter (RSP): customer value The upper part of Figure 1 shows the customer, customer value, and the corresponding modeling methods: scenario using persona and Receiver State Parameters (RSP) [3] [4]. A set of RSPs represent customer value and they are indices of customer satisfaction in receiving service offerings. 2.3 View model: service contents and their corresponding functions The middle section of Figure 1 shows the service contents, their functions, and the corresponding “view model” method [3] [4] [5]. After identifying the customer value as RSPs, functions and attributes of entities for each RSP can be described as a view model. A view model works as a bridge between the customer value and actual entities via a tree structure. Yoshikawa's General Design Theory (GDT) [9] provides a basis for our approach. The theory is discussed in terms of two topologies defined by functions and attributes of artifacts. The projection from functions to attributes can be universally recognized as design of products. Assuming that services can also be designed by the same projection, RSPs may consist of parameters in both function and attributes. A function is defined in this paper as “a description of behavior abstracted by humans through recognition of the behavior in order to utilize the behavior” [10]. Here, the term behavior implies both physical phenomena and human activity. According to this definition, a function can be represented in two ways: (1) as symbols represented in the form of to do something and (2) as a set of behaviors. In order to emphasize the flexibility of the description, let us consider the first representation wherein functions in a view model can be represented as lexical symbols (i.e., (1)). Although the symbols are meaningful only to humans, this information, which is associated with the RSP, is essential for clarifying the roles of the design objects. On the other hand, the behavioral aspects of functions (i.e., (2)) are incorporated in the linkage using the service blueprint that is introduced in Section 2.6. Some of lowest-level functions are implemented through humanware (such as staff and customers), and some of lowest-level functions are implemented through hardware

(in the form of machines and facility). Software involves both these functions. Since the customer value (represented through RSP) is related to an embodiment of a service, whose characteristics are recognized as attributes, designers can perform a static evaluation of customer satisfaction based on these entities and their attributes. However, the view model includes little information with regard to the service delivery process. Thus, the ways in which entities complete the connected functions are inevident. The details of the relationships between functions and entities are depicted in a service blueprint. 2.4 Traditional service blueprinting method The service blueprint [11] [12] and the service map [13] are the most famous tools used by marketers to sequentially and visually describe service activities. In the service blueprint, service activities are arranged with respect to two lines: (1) the line of interaction around which the customer and the service provider interact and (2) the line of visibility that separates the “onstage” (visible) activities from the “backstage” (invisible) activities performed by the provider. The service blueprint is known to be an effective tool for analyzing and designing the delivery of services prior to the actual delivery. The service blueprint, however, has the following problems in terms of PSS development. Difficulties in association of the described service process to the customer requirements A number of researches (e.g., [14] [15] [16]) have pointed out that the service blueprint is more an operating manual of the provided service, rather than a depiction of customer requirements. The service blueprint is unable to properly correlate a customer value and service activity. This problem makes it difficult to assess the quality of services from the point of view of the customer. Need for physical processes in addition to human processes Academic literature on the service blueprint has placed considerable emphasis on the interpersonal service delivery system. In this study, however, the authors strive to develop a service offering comparable to PSS, which itself is a combination of products and service activities. Since human processes and physical processes have alternative and/or complementary relationships with each other in PSS, understanding product behavior and its relationship with service activities is essential in the design, evaluation, and simulation of a PSS throughout the lifecycle of products. Therefore, the blueprint of a service such as PSS should contain information concerning the product and its service behavior as well as information on the human activity associated with the service. Lack of normative notation Shostack’s blueprint notation in earlier literature was basically a simple flowchart. Consequently, the detailed meanings of graphical elements are often ambiguous and not well defined [17]. Normative notation and explicit control flow are needed for analyzing and evaluating the described service delivery processes. 2.5 Extended service blueprint: interrelated activity blueprint and product blueprint using BPMN In order to solve these problems, the service blueprint is extended to include product behavior and its relationship with service activities as well as the relationship with customer value as shown in Figure 2. The extended service blueprint consists of an interrelated activity blueprint and behavior blueprint.

181

BPMN elements (partial)

RSP

View model (Function structure)

Task

Sequence flow

Pool

Start event

Message flow

Lane

End event

AND-split

Message start event

Message end event

Correspondence between functions and service activities/product behaviors

Extended service blueprint Receiver’s actions Receiver

Receiver

Ref erence

Line of visibility

Invisible

Hardware A

Activity blueprint

Line of interaction

Visible

Invisible

Humanware A

Visible

Service activities (human process)

XOR-split

Internal collaboration between staff and facilities

Product behaviors (physical process)

Behavior blueprint

Figure 2: Detailed illustration of the modeling method focusing around service delivery process The Business Process Modeling Notation (BPMN) [18] [19] [20] is used for describing the service blueprint so as to have consistent semantics (as shown in Figure 2). The modeling in BPMN is made by simple diagrams with a small set of well-defined graphical elements. The adoption of BPMN supports wide variety of control flows and provides a graphical representation that is readily understandable by all business users, from the business analysts, to the technical developers, and to the business people who will manage and monitor those processes [20]. By connecting the view model aforementioned and the extended service blueprint, it is possible to describe service activities and product behaviors while clarifying their influence on the receiver (i.e., quality of service). In other words, by focusing on customer value and the roles of entities as described in the view model, service activities and product behaviors can be equivalently dealt in the extended service blueprint. The extended blueprint can be especially used as a communication tool for managers, marketers, and engineers in service development. Activity blueprint The activity blueprint corresponds to Shostack’s blueprint and illustrates the activity-oriented aspects of a service. The left section of Figure 2 represents an activity blueprint using BPMN. Each humanware of a service is arranged as a BPMN pool, and the line of visibility is denoted as the border between a visible BPMN lane and an invisible BPMN lane in the pool. Some of the steps performed by the receiver in the activity blueprint are selected from the scenario presented in Section 2.2. The activity blueprint specifies the interactions between the receiver and the staff; these interactions are represented as BPMN message flows. Human processes, which are represented by a set of service activities and BPMN sequence flows among them, are subject to organizational rules, employee manual, and so on.

182

Behavior blueprint The product blueprint illustrates the behavior-oriented aspects of a service. Physical processes in the behavior blueprint are described as well as the activity blueprint using BPMN for the sake of achieving a simple user interface. Since BPMN is a general-purpose modeling language for business process, it can be applied to a technology-oriented process in PSS. The behavior blueprint specifies the interactions between the receiver and the products including self-service machines. These interactions are represented as BPMN message flows. Physical processes, which are represented by a set of product behaviors and BPMN sequence flows among them, are subject to physical lows and/or computational algorithms. Relationships between two blueprints By preparing a similar user interface for both activity and behavior blueprints, marketers and engineers can easily understand both blueprints. In addition, as shown in Figure 2, there is an interrelation between the behavior blueprint and the activity blueprint. Some BPMN message events (shown by a letter with a circle) are symbols that show two types of collaborations between the two blueprints. The first type of collaboration involves an interaction between the receiver and the product hardware, while the second involves interactions between the staff and equipment or facilities. Information about such collaborations and service delivery denotes how the products are used, which is useful for product design.

2.6 Relationships between view models and two blueprints The middle section of Figure 2 presents the relationships between functions in a view model and the service activities/product behaviors in a service blueprint. Each of the lowest-level functions is mapped to a process that produces a service; the process can comprise service

activities, product behaviors, and receiver actions. Such relationships represent the behavioral aspects of the lowest-level functions. Therefore, they are subjective and exhibit a many-to-many correspondence, according to the discussions on function and behavior in conventional design studies [21]. In the case where a mapped process includes receiver actions, the corresponding function needs customer participation as a co-producer of the service. Some of the humanware/hardware entities in view models, such as staff and machines, are correlated with BPMN pools in the corresponding activity/behavior blueprint. The remaining entities (i.e., static objects) in the view models can be correlated with BPMN data objects. According to the above relationships, the typical steps to describe service blueprints based on view models are as follows: y y y y

Add BPMN pools that correspond to entities in the view models. Deploy each of the lowest-level functions in the view models into a series of activities and/or behaviors. Add BPMN data objects that correspond to rest of the entities in the view models. Organize all processes to ensure the totality of the delivery process.

3

METHOD FOR ANALYZING THE STRUCTURE OF SERVICE DELIVERY PROCESS 3.1 Representation of service delivery process In this paper, the term “task” is inclusive term for process elements in a service blueprint, namely, receiver action, provider’s activity, and product behavior. A set of all tasks in a service blueprint is represented as (1) (2) where denotes a set of actions by a receiver of a service, denotes a set of human activities by service provider, and denotes a set of product behaviors by facility and product. In this paper, a service process that comprises elements is called “service receiving process,” while a service of process that comprised elements of is called “service providing process”. Hence, whole service process that comprises both receiving process and providing process is called “service delivery process.” A subset of tasks corresponding to an RSP in is defined through the relationship between functions and service activities/product behaviors explained in Section 2.6. In the following sections, such a subset corresponding to the i-th RSP is represented as . 3.2 Analysis of service delivery process Table 1 (in Section 4) presents a framework for analyzing the structure of a service delivery process according to customer satisfaction elements. RSPs and their relative importance are listed on the vertical axis. Indices of service delivery process are listed on the horizontal axis: Visibility to receiver and interactivity with receiver are indices of activity blueprint and behavior blueprint; the ratio between service activity and product behavior is an index of the balance between two blueprints; and degree of receiver participation is an index of the entire service blueprint. These indices are explained below.

Visibility to receiver This index represents recognizability of provider’s tasks in a service providing process. Here, the recognizability of a task implies receiver’s perception of the task through not only the visual sense but also the auditory and olfactory sense. Visibility to receiver with regard to the i-th RSP is defined as

,

1

(3)

where ( : 0,1 ) denotes a map from task to its visibility. In Eq. (3), is replaced with or depending on which blueprint is focused on. If the index is higher, it is implied that a receiver can obtain more information to evaluate the service from visible provider’s tasks: there is a high possibility that the providing process may influence other unexpected RSPs in addition to the original target RSPs. For instance, in case of a restaurant service wherein process of cooking in a kitchen is visible, other RSPs such as “a feeling of security” and “cleanliness” might be identified and affected by the process in addition to the original target RSPs such as “delicious dish.” Thus, when analyzing a service process with high visibility, designers should make sure that the receiver state is sufficiently analyzed regardless of the importance of the original target RSPs. Interactivity with receiver This index represents the ratio of interaction tasks with a receiver to all of the tasks in a service delivery process. Interactivity with a receiver with regard to the k-th RSP is defined as (4) where and denote a set of sending and receiving tasks of messages from provider to receiver respectively; while and denote a set of sending and receiving tasks of messages from receiver to provider, respectively. In Eq. (4), is replaced with or depending on which blueprint is considered. If the index is higher, it is implied that a service providing process and a receiving process are more interdependent; scheduling and timing of each process are strongly emphasized. On the contrary, if the index is lower, these processes can be performed independently. Degree of receiver participation This index represents the ratio of receiver actions to all the tasks in a service delivery process. The degree of receiver participation with regard to the i-th RSP is defined as (5) If the index is higher, it is implied that a receiver contributes more to the realization of a service delivery process. In this case, since the process depends on the attribute and ability of the receiver, it can be difficult for the provider to control the quality of service functions actualized by the process. This leads to uncertainty and variability of such functions due to the receiver. One of countermeasures for solving this issue is to provide a complementary service or product that helps the receiver to easily perform his/her actions.

183

4 APPLICATION 4.1 Public transportation service: Manyo Line The aforementioned methods have been applied to the Manyo Line (tram) [22] (Figure 3), which runs between Takaoka City and Imizu City in Toyama Prefecture.

Motorman

Passenger

activity blueprint. Therefore, it is found that reconsidering the method of paying fares can contribute to not only satisfaction regarding the RSP “ease of paying fare” but also satisfaction regarding the RSP “punctuality.”

Figure 3: Photograph of Manyo Line. Table 1 presents the result of the application case study. As shown in Table 1, the following parameters were identified as RSPs: transportation to destination, tram availability, tram punctuality, ride comfort, and ease of paying fare. The RSP weights for the passengers were computed numerically according to the AHP method [23], using bilateral comparisons between parameters. An overview of the result focusing on the visibility column and the interactivity column indicates that the recognizability of a motorman’s activities is much lower than that of the tram’s behaviors. This is because passengers only feel the t ram’s behaviors controlled by the motorman at the forefront of the tram while riding in it. However, considering the information about the ratio between service activities and product behaviors (the middle part of Table 1), the result indicates that such activities of motorman surely contribute to the realization of the service. Regarding the result of RSP “tram availability,” the degree of receiver participation in the RSP is zero. This implies that the quality of service functions affecting such RSP can be mostly designed prior to the actual tram operations: (i.e., embedded in planned operation hours). Let us now focus on the RSP “punctuality,” which has the highest importance among the identified RSPs. The ratio of service activity and the degree of receiver participation in the RSP are both high. This results from the high ratio of dialogues between the passengers and the motorman to the process affecting the function of controlling standing time for the RSP (e.g., dialogues pertaining to lost numbered tickets and exchange of money). Figure 4 shows interactions between the passengers and the motorman when getting off the tram in the described

Figure 4: Interactions between passengers and motorman when getting off the tram in the described activity blueprint. 5 CONCLUSION The aim of SPE research is to develop methodologies to evaluate and design PSS. This paper presented an analytical framework for analyzing the structures of service delivery process according to customer satisfaction elements as a next step of service modeling. The framework presented in this paper includes the following indices of service delivery process: (1) visibility to receiver, (2) interactivity with receiver, and (3) degree of receiver participation. Through an application case study, it is found that the method can indicates the features of services from the viewpoint of service function, human activity, and product behavior. Such implications contribute to evaluation of services and acquirement of clues for improving them. Future research will include the feasibility assessment on more complex product–service combinations and the enrichment of modeling criteria for ease of applying the analytical method presented in this paper. 6 ACKNOWLEDGMENTS This research was partially supported by JSPS Research Fellowships for Young Scientists. We would like to thank Dr. Kazuya Kobayashi and the RACDA Takaoka for providing information for the case presented here.

Table 1: Result of analyzing service delivery process regarding public transportation service: Manyo Line. Service delivery process Service activity

Product behavior Ratio between service activity and product behavior

Visibility to receiver

Interactivity with receiver

0.50

0.70

0.29

0.17

0

0

0

0

0.67

0.33

0.56

0.20

0.29

0.47

0.53

0.88

0.15

0.25

0.20

0.80

0.63

0.56

0.50

Receiver State Parameter (RSP)

Importance weight

Visibility to receiver

Interactivity with receiver

Transportation to destination

0.24

0

0

0.50

Tram availability

0.26

0

0

1.00

Tram punctuality

0.29

0.33

0.21

Ride comfort

0.11

0

0

Ease of paying fares

0.04

0

0

184

Receiver Degree of receiver participation

7

REFERENCES

[1] [2] [3]

[4]

[5]

[6] [7] [8]

[9]

[10]

[11] [12] [13]

[14]

[15]

[16]

[17]

Tukker A. and Tischner U., 2006, New Business for Old Europe: Greenleaf Publishing. Tischner U., Verkuijl M., and Tukker A., 2002, First Draft PSS Review. Arai T. and Shimomura Y., 2004, Proposal of service CAD system - A tool for service engineering. CIRP Annals-Manufacturing Technology, 53(1): 397-400. Sakao T. and Shimomura Y., 2007, Service Engineering: a novel engineering discipline for producers to increase value combining service and product. Journal of Cleaner Production, 15(6): 590604. Hara T., Arai T., and Shimomura Y., 2008, Integrated Representation of Function, Service Activity, and Product Behavior for Service Development. in In Proceedings of the 13th Design for Manufacturing and the Life Cycle Conference DFMLC2008 -: The American Society for Mechanical Engineering (ASME). Lovelock C.H. and Wright L.K., 2002, Principle of Service Marketing and Management. Fisk R.P., Grove S.J., and John J., 2000, Interactive services marketing. Boston: Houghton Mifflin. Rizzo A., Pasquini A., Di Nucci P., and Bagnara S., 2000, SHELFS: Managing critical issues through experience feedback. Human Factors and Ergonomics in Manufacturing, 10(1): 83-98. Yoshikawa H., 1981, General Design Theory and a CAD system, in Man-Machine Communication in CAD/CAM, T. Sata and E. Warman, Editors, NorthHolland Publishing Company: Amsterdam. 35-38. Umeda Y., Takeda H., Tomiyama T., and Yoshikawa H., 1990, Function, behavior, and structure, in In Applications of Artificial Intelligence in Engineering, J.S. Gero, Editor. 177-193. Shostack G.L., 1982, How to Design a Service. European Journal of Marketing, 16(1): 49-63. Shostack G.L., 1984, Designing Services That Deliver. Harvard Business Review, 62(1): 133-139. Kingman-Brundage J., 1991, Technology, Design and Service Quality. Technology, Design and Service Quality, 2(3): 47-59. Brooks R. and Lings L., 1996, A Hierarchy of Customer Satisfaction, The Inadequacies of Service Blueprinting, in Proceedings of the 25th Annual Conference of the European Marketing Academy: Budapest, Hungary. 147-164. Pires G. and Stanton P., 2004, The Role of Customer Experiences in the Development of Service Blueprints, in ANZMAC 2004 Conference. Stahel R.W., 1997, The Functional Economy: Cultural and Organizational Change, in The industrial green game: Implications for Environmental Design and Management, D.J. Richards, Editor, National Academy Press: Washington DC. 91-100. Congram C. and Epelman M., 1995, How to Describe Your Service - an Invitation to the Structured Analysis and Design Technique.

[18] [19] [20]

[21]

[22] [23]

185

International Journal of Service Industry Management, 6(2): 6-23. BPMN Information Home [cited; Available from: http://www.bpmn.org. Harvey M., 2005, Essential business process modeling. Sebastopol, CA: O'Reilly Media, Inc. White A.L., Stoughton M., and Feng L., 1999, Servicizing: the Quiet Transition to Extended Producer Responsibility, Tellus InstituteServi: Boston, MA. Umeda Y., Ishii M., Yoshioka M., Shimomura Y., and Tomiyama T., 1996, Supporting conceptual design based on the function-behavior-state modeler. Ai Edam-Artificial Intelligence for Engineering Design Analysis and Manufacturing, 10(4): 275-288. Manyo Line [cited; Available from: http://www1.coralnet.or.jp/manyosen/ (in Japanese). Satty T.L., 1980, The Analytic Hierarchy Process: McGraw-Hill.

Use-Oriented Business Models and Flexibility in Industrial Product-Service Systems A. Richter1, T. Sadek2, M. Steven1, E. G. Welp2 Chair of Production Management, Ruhr-University Bochum, Universitaetsstrasse 150, 44801 Bochum, Germany, {marion.steven, alexander.richter}@rub.de 2 Chair of Engineering Design, Ruhr-University Bochum, Universitaetsstrasse 150, 44801 Bochum, Germany, {welp, sadek}@lmk.rub.de 1

Abstract Today’s corporate environments are characterized by growing dynamics and uncertainties. Here, flexibility gains importance as a critical success factor. This is especially true for long-term customer-supplier relationships. As a solution to the mentioned uncertainties connected with such a business relationship, one can think of flexible systems. The contribution at hand focuses on contracts to control for customer-supplier relationships. By reallocating property rights in use-oriented business models it is possible to distribute incentives and risks to better balance the interests of customers and suppliers. Our contribution points out the importance of flexibility and describes the opportunity to detect the optimal degree of flexibility of an IPS². Keywords: Use-oriented business models, Industrial Product-Service Systems, incomplete contracts, flexibility

1 INTRODUCTION The change from a mere product business to selling customized problem solutions has lead to the establishment of terms such as business models, performance contracts, life cycle costs and productservice systems. This conception, which focuses on securing sustained earnings through services besides the one-off sale of products, originates in the change of customers’ requirements and is driven to a great extent by the reallocation of risks and incentives. In a business environment characterized by increased uncertainty, the aspects “availability” and “flexibility” of an Industrial Product-Service System are thus of special significance.

high publication density regarding the issue of useoriented business models raises the following questions (figure 1):

2

Uncertainty

?

Industrial Product-Service Systems

?

Use-oriented Business Models

?

P 29 S TR

customer

?

Flexibility

supplier

economic assessment

? economic assessment

t0

t1

Figure 1: Questions Here, the classical term “production system” is consciously replaced by the term “Industrial ProductService System” (IPSS) which, according to its definition, is characterized by a life-cycle-oriented integration of the industrial supply of products and service parts [1]. This substitution of the term which is discussed both in academic and in industrial circles as well as a currently

CIRP IPS2 Conference 2009

186



What is the relationship between uncertainty, use-oriented business models and IPSS?



What is the significance of the aspect flexibility in use-oriented business models and how can flexibility be integrated into IPSS?



How can IPSS be designed in an economically sensible way and how can you quantify the value of flexibility of IPSS?

UNCERTAINTY, USE-ORIENTED BUSINESS MODELS AND IPSS Contracts, which create an institutional framework within which rights, obligations and responsibilities are regulated, constitute the basis of a business relationship between the supplier and the customer. Thus, contracts determine business models and, depending on these, they can be of formal and/or informal nature, i.e. explicitly stipulate terms and/ or include implicit agreements. The design of contracts and, thus, of business models is, in particular, characterized by the factor "uncertainty". On account of "uncertainty", long-term contracts have to remain “incomplete” so that they provide room for opportunistic behaviour and therefore influence the players’ incentive to behave in the sense of the business relationship. Besides the negative consequences related to this uncertainty regarding conduct, the contracts' incompleteness offers the possibility to flexibly react to future environmental situations. Thus, uncertainty does not only generate risks, but also, most importantly, chances.

Cost-Plus

Fixed-Price

(customer Ownership)

(customer Ownership)

Costs/ Revenues life cycle revenues1 (accumulated)

Use-oriented business model (supplier Ownership) customer k

C/R

C/R

LCR k

K/E

LCR3

LCR2

P3

LCC k

profit1

P2

customer l C/R

LCC2

life cycle costs1

t

t

risk needs to be borne by customer

risk needs to be borne by supplier

purchasing a production facility

change

t

LCC3

LCR l

LCC1

LCC l t

t

win-win-situation for customer and supplier

purchasing a performance

P1 < P2 < P3 Figure 2: Costs, revenues and profits of selected business models It is the aim of innovative business models to evenly spread risks, chances and incentives between the supplier and the customer [2]. In figure 2, the evolution of business models is depicted. The expansion of the spectrum of business models from the cost-plus to the use-oriented business model is rendered possible through innovative technologies, the rearrangement of the ownership of capital goods (customer ownership versus supplier ownership) and, in particular, through the expansion of industrial services. This directly affects the flows of accumulated costs (LCCi), revenues (LCRi) and profits (Pi). In cost-plus offers, industrial services are mainly only intended as add-on and are limited to the maintenance and servicing of certain components of the production system. Further services are merely optional, as the customer is the owner of the machine and is responsible for the availability of the machine. In this business model, the financial risk caused by a system failure is the responsibility of the customer and results in fluctuating life-cycle costs (LCC1) (figure 2). As the supplier does not assume any risk in this business model, he has no incentive to carry out sustained changes to the machine and to thus reduce the life-cycle costs. Fixed-price models, however, include the customer’s requirement that the product's life-cycle costs are guaranteed, depicted in the linear course of the LCC2 curve in figure 2. This leads to the transition from a transaction-oriented, short-term business relationship to a relational, long-term business relationship [3] in which a substantial proportion of the risk of a failure is transferred to the supplier. In order to determine and/or reduce the risk of a failure and the costs related herewith, the supplier expands his industrial service offer and provides condition-oriented maintenance and servicing especially aligned to the machine. In the contrary to the cost-plus business model, it is necessary to integrate the development of products and services in this case. Through bundling product and industrial services, incentives have now, however, been created for the supplier to reduce the product-service system’s life-cycle costs, but not to increase the productivity of this system. Likewise, the customer has no incentive to operate the technical system in “manner which protects the material”.

The comparison of cost-plus and fixed-price business models leads to the conclusion that a one-sided distribution of risks and incentives is no basis for solving the problem in a way which is ideal for both parties. Useoriented business models which go beyond existing approaches such as “use-oriented maintenance” [4] offer an approach for achieving an efficient distribution of risks and incentives. The difference lies in the title to the machine, whereby it no longer becomes the customer’s property but remains the property of the supplier. In this case, the range of the offer of these contracts reaches up to the “temporary availability” of a partial production system, at least in theory. Similar to the basic principle of a car rental service, it is intended to sell an availability service within a determined time interval to various customers in this case (figure 2). The temporary availability includes both the system’s guaranteed quantitative operational readiness and its availability in terms of location and time. Besides adapting the technical system, it is also necessary to extend the industrial service portfolio (logistics, ramp up, etc.) in order to ensure this. Through aligning the business partners’ interests, incentives are now created for the supplier to not only improve the service offer’s quality but also its productivity, whereby the flexibility of the product-service system required for implementing this business model takes advantage of the incompleteness of contracts. It becomes clear that product-service systems and useoriented business models are not to be understood as synonyms, but that they complement one another. 3 RELATED LITERATURE: CLOSING THE GAP A great deal of literature has been dedicated to the problem of contract design both in the literature of economics and with regard to supply chain management. In the area of research of economics the theory of incomplete contracts deals with the question of how to design an optimal contractual relationship, so as to induce efficient transaction-specific investments (for a literature review see Schmitz [5]). According to this theory investments are made with aim of either reducing production costs or increasing the value of the good. The term “investment” does not only refer to financial aspects, but is used to describe all activities which increase the profitability of a business relationship. Selfish and

187

customer

RISK

technology

IPS2development

market t0 C

C

RR

t1 [time]

life cycle

C

C C T

T

RR

T

RR

M

M

M

M

M [t]

LClong

Shortening of lifecycle

[t]

LC

T

T

T

M

LCshort

C

Determining uncertainty

LC

[t]

Changeability

Figure 3: Development risks and possibilities to reduce risks cooperative investments need to be distinguished. While selfish investments from the supplier’s point of view aim at reducing production costs, cooperative investments result in an increased value of the good. Industrial services such as maintenance and repair can in this regard be classified as cooperative investments. Renegotiations which take place after uncertainties regarding the environment have dissolved implicate underinvestments in this theory. These result from the fact, that while the investing party needs to cover the costs of its investment alone, it has to share the surplus by renegotiation with the other party (hold-up problem). Renegotiations therefore may have negative economic consequences. Contrary to this, contract analysis in supply chain management (SCM) focuses on investments in capacity and inventory (for an overview over supply chain contracts see Tsay, Nahmias and Agrawal [6]). The basic difference between these lines of argumentation is the modeling of renegotiations which forms no part of the SCM-literature. Rather, in return to a commitment customers obtain the flexibility to revise their initial decision conditional on updated information in subsequent periods. Risks are thereby implicitly transferred to the supplier. For example with so called „quantity flexibility contracts“ customers have to communicate ex ante forecasts of their estimated demands to suppliers. These can be adjusted ex post after the market demand has been realized. To earn this right the customer has to ex ante commit to a minimum purchase quantity. Other contract forms distinguish between the reservation of capacity and its actual usage. As is the case in the SCM-literature, aspects of flexibility and contract design are a central part of our contribution. In contrast to an analysis on the level of components, which is the focus of the previously described literature, we concentrate on the system’s level. In this context, in addition to the reallocation of incentives and risks through contracts the problem of “product architecture” gains importance. Baiman et al. examine the interplay of performance measures for contract design, the product

188

architecture and incentive efficiency in a supply chain [7]. Their results show that when choosing a product architecture not only consequences of manufacturability need to find consideration, but also the incentive efficiency of the whole supply chain. However, they do not consider aspects of flexibility and services in their contribution. In the following, this article neglects the strategic interaction between supplier and customer and focuses on the development of the system and its flexibility. We motivate this approach because, as discussed in the previous chapter, the choice of the use-oriented business model creates the “right” incentives for suppliers. Consequently, the incomplete character of contracts is seen as an opportunity and not a risk of opportunistic behavior. What needs to be discussed is the degree of flexibility IPSS must have, in order to realize future opportunities when delivering the system. 4

SIGNIFICANCE OF FLEXIBILITY AND ITS INTERGATION INTO IPSS Below, the aspect of “flexibility” which is frequently discussed in literature will be dealt with and brought into relation with IPSS (compare [8]). For the time being, we shall shelve the definition of the term in favour of a general contemplation. Figure 3 points out the initial problem. In today’s development decision (t0), uncertain values of tomorrow (t1) flow in along the life cycle which can be bundled in the development risk. In this case, the interplay of the risk factors “technology", “market” and “customer” represent both endogenous and exogenous factors which will affect the industrial productservice bundle in future. Here, it becomes evident that, besides aspects such as uncertainty regarding the further development of a technology or fluctuating sales volumes required by the market, in particular instable customer preferences increase the development risk [9]. In order to reduce this,

system adapts itself without external actuation

Agility Agility

system can be changed rapidly

Robustness Robustness

Flexibility Flexibility

system can be changed easily

no implementation of changes from external necessary

implementation of changes from external necessary

Adaptability Adaptability aspects of changeability

system is insensitive towards changes within its environment

Figure 4: Four aspects of changeability alternative solution mechanisms exist which are depicted in the bottom half of figure 3. On the one hand, the period of contemplation and/or the life cycle (on behalf of a service) can be shortened to, in particular, reduce the risk implied by the market and the technology. On the other hand, there is the possibility to invest more time into examining the uncertainty in order to take the likelihood of future environmental situations occurring as well as their effect into account when developing goods. On account of relational contracts' being of a long-term nature and the systems becoming more complex for integration reasons, these mechanisms are, however, not sufficient for reducing the risk involved in the development of Industrial Product-Service Systems. In combination with the incompleteness of use-oriented service contracts, it is of great importance to take the third mechanism, the system’s changeability, into account, whereby changeability is generally understood as the ability of a system to react to relevant changes induced by the system or the environment by means of available internal or external factors [10]. As you can see in figure 4, robustness, flexibility, adaptability and agility (flexibility in a broader sense) are immanent components of changeability according to Fricke and Schultz [11]. However, it only makes sense to use these differentiations regarding the terms, which are partially extended by the aspect of reconfigurability, in combination with defined system boundaries. As this article is based on a superordinate understanding of changeability and, in particular, the character of the possibility to act is intended to be emphasized, the term “flexibility” is used below. There are many possibilities for integrating flexibility into a system, whereby especially the aspect of modularisation comes into the fore, besides principles such as the independence of the system's elements, the reduction of the system's complexity or non-hierarchic coupling [11]. As, according to Wiendahl [12], modularisation is understood to be the substantial enabler of flexibility in production technology, the idea seems reasonable to also apply this approach with regard to designing the architecture of Industrial Product-Service Systems. Besides the production-oriented approaches of Schuh et al. [13], the modularisation approach for industrial services according to Böhmann et al. [14] must be

mentioned in this context, whereby the modularisation of Industrial Product-Service Systems also provides for integrated hybrid modules, besides mere products or service modules. 5

VALUE-ORIENTED DESIGN AND MANAGEMENT OF IPSS To this point, the explanations have revealed that in particular in the context of Industrial Product-Service Systems increased significance is placed on the strategic success factor “flexibility” in product development. Generally, technical methods do, however, follow the simple heuristic “more flexibility is good”. In this section, a superordinate reference value shall be introduced with the economic value by means of which the process of determining the ideal degree of flexibility, i.e. the assessment of the value contribution of flexibility, can be supported. Baldwin and Clark [15] stipulate in a similar context "Designers see and seek value flexibility in new designs" and Sullivan et al. [16] interpret the design and/or the development process to be “one of investing valuable resources under uncertainty with the goal of maximising value added”. In this context, establishing flexibility in Industrial Product-Service Systems implies increased investments which are caused by an increased use of resources on the one hand and opportunity costs due to opportunities of alternative allocation of resources which have not been seized, on the other hand (figure 2). The higher the economic costs for adapting a product are, the lower is flexibility. Thus, the ideal extent of flexibility is a parameter of the economic trade-off of high preproduction costs and low follow-up costs. In this context, controlling has the task of planning, steering and coordinating the establishment of flexibility in development, whereby the challenge lies in assessing the economic value of flexibility, whereas it is relatively easy to quantify the costs related to establishing flexibility. The economic value of flexibility is the result of the ability to react to changing conditions in the future. Thus, the payment flows related to this depend on the uncertain future conditions.

189

risk profile under consideration of flexibility due to options

risk profile without considering flexibility due to options

probability

probability

symmetric risk profile

expected value

e.g. contraction option

asymmetric risk profile z.B. expansion option benefit from chances

expected value

risk is limited

value of capital investment

value of flexibility

Figure 5: Taking flexibility through options into account Traditional assessment procedures of investment planning and budgeting (e.g. net present value method) or of cost management (e.g. target cost management) are no longer sufficient against this setting, as assumptions regarding an expected scenario of payment flows are implicitly related with them [17]. The respective restrictions of these instruments, which literature classifies as deterministic, result in the investment object being systematically underrated and lead to the real option approach being recommended (figure 5). A key finding of this approach is that increased uncertainty of the payment flows triggered by the investment increases the value of flexibility and/or the value of the real option. However, frequently there is a lack of flexibility to carry out actions at a later date at an adequate cost if the Industrial Product-Service System has not been designed for this purpose from the beginning. In the following section, the idea of the real option approach will be presented and transferred to the development of Industrial ProductService Systems. 6

THE REAL-OPTION APPROACH FOR ASSESSING FLEXIBILITY An option is defined as the right, but not the obligation to purchase (call) or sell (put) a specified asset (a share, a contract or a design) at a price stipulated in advance (basic price or exercise price) within an agreed period of time (term of the option). The right to only exercise an option if it is in the interest of the option holder implies an asymmetric spread of risks of payment flows (figure 5). Thus, options differ from contracts which include the obligation to purchase or sell an item of property at specified conditions in the future from today on [18]. Therefore symmetric payment flows, which can develop both positively and negatively are involved in contracts. Due to the options' asymmetric characteristic, risks can be limited on the one hand and chances arising in the future can be used on the other hand. Against this setting, there is the wish for more uncertainty so that the option's value increases. On the contrary to financial options which are traded in financial markets and for which there are elaborate

190

assessment procedures, real options relate to real items, such as design, technologies or production processes [15]. Typical options existing in business practice are: • • • •

the defer option, which describes the possibility of deferring investment until new information is received, the abandonment option, with which the investment can be carried out stepwise and, if necessary, be called off, the expansion or contracting option, which provides the possibility of adapting the extent of the investment as well as, the switching option, which enables its holder to change the way an item of property functions (e.g. to the input or output of a flexible production plant) [17].

There are certain similarities between financial and real options. For example, if the interim results did not meet expectations, the abandonment option of an r&d project can be compared with a call option in the financial market which has not been exercised if the value of the object offered under subscription is below the exercise price. Despite certain similarities, there are, however, significant differences in the investment environment which make it difficult to directly apply the assessment methods for financial options to real options [18], as the application of these procedures requires a complete capital market in which the uncertainty of the investment through traded assets can be duplicated. The risk within the development processes is, however, of private nature and cannot be represented in the market. Thus, the approach of dynamic programming is pursued in literature - an approach which does not require this strict assumption, but which is, however, not able to determine the correct, risk-adjusted interest rate [19]. Finally, instead of looking at the assessment procedures, we shall examine how the idea of service bundling is dealt with in the context of an optionbased conception. The IPSS is to be understood as a portfolio of assets whose value is determined by options [16]. It is the aim to treat the flexibility of an IPSS as an option and therefore

be able to quantify it easily. According to the explanations in section 3, options are generated through as system’s architecture. As this is, however, frequently determined through technical decisions in practice, the maximization of the value added of an Industrial Product-Service System is possibly not given. Thus, the modularization does possess a value through creating a portfolio of options which, if you assume that the aggregated value function remains the same, supersedes the value of an option on a portfolio [16]. However, negative (economic) consequences must be expected with an over or undermodularisation [20]. By means of the real-option approach, the ideal strategy, i.e. the decision regarding the ideal number of modules can be supported here. Besides duplicating the options in form of a "mix & match" of modules, there is also the possibility to change this locally, i.e. to exercise the options described above on individual modules.

[4]

[5]

[6]

[7]

[8] 7 SUMMARY The contracts for regulating customer-supplier relations which remain incomplete on account of being of a longterm nature and therefore being uncertain imply problems regarding incentives and thus inefficiencies. In particular, through re-allocating property rights it is possible to spread incentives and risks more evenly in use-oriented business models and to align the customer’s and the supplier’s interests. Thus, the leeway which exists thanks to the incompleteness of contracts is not understood as a risk, but rather as a chance which can be seized through correspondingly developing a flexible IPSS. The article reveals the significance of flexibility and describes the possibility of determining the value and therefore the ideal extent of flexibility of such a system by means of the realoption approach. In particular, through determining both the business model and the IPSS, it can be ascertained that these are not synonyms, but two constructs which complement one another. Future research needs a combined view on IPSS and their related innovative business models. The objective is to have a better understanding of the relation between the (architecture of an) IPSS and the business model. A first approach was provided by Richter and Steven who specify the dominance of use- and result-oriented models subject to the complexity of product design [21].

[9]

[10]

[11]

[12]

[13]

[14] ACKNOWLEDGMENTS This research is financially supported by the German Science Foundation (DFG) through SFB/Tr29. [15] REFERENCES [16] [1]

[2]

[3]

Meier, H., Uhlmann, E., Kortmann, D., 2005, Hybride Leistungsbündel – Nutzenorientiertes Produktverständnis durch interferierende Sach- und Dienstleistungen, wt Werkstattstechnik online, 95/6: 528-532. Kim, S.-H., Cohen, M. A., Netessine, S., 2007, Performance Contracting in After service Supply Chains, Management Science, 53: 1843-1858. Oliva, R., Kallenberg, R., 2003, Managing the Transition from Products to Services, International Journal of Service Industry Management, 14: 160172.

[17]

[18]

[19]

191

Schuh, G., Gottschalk, S., Odak, R., Kempf, M., Kupke, D., 2007, Verfügbarkeitsorientierte Instandhaltung – Stellhebel zur Aufrechterhaltung und Steigerung der Verfügbarkeit in produzierenden Unternehmen, ZWF 102/9: 516-519. Schmitz, P.W., 2001, The Hold-Up Problem and Incomplete Contracts: A Survey of Recent Topics in Contract Theory, Bulletin of Economic Research, 53: 1-18. Tsay, A., Nahmias, S., Agrawal, N., 1999, Modeling Supply Chain Contracts: A Review, in: Tayur, S., Ganeshan, R., Magazine, M. (Edit.), Quantitative Models for Suplly Chain Management, Boston, Kluwer Academic Publishers, 301-330. Baiman, S., Fischer, P.E., Rajan, M.V., 2001, Performance Measurement and Design in Supply Chains, Management Science, 47: 173-188. Karger, M., Richter, A., Sadek, T., Strotmann, W. C., 2008, Flexibility of Industrial Product-Service Systems – An Assessment Based on Concept Modelling, 3rd Annual International Conference on Business Market, St. Gallen/CH. Thomke, S., Reinertson, D., 1998, Agile Product Development: Flexibility in Uncertain Environments, California Management Review, 4: 8-30. Ferguson, S., Siddiqi, A., Lewis, K., de Weck, O.L., 2007, Flexible and Reconfigurable Systems : Nomenclature and Review, Proceedings of the ASME 2007 International Design Engineering Technical & Computers and Information in Engineering Conference, Las Vegas, USA. Fricke, E., Schulz, A.P., 2005, Design for Changeablility (DFC): Principles to Enable Changes in Systems Throughout Their Entire Lifecycle, Systems Engineering, 8: 342-359. Wiendahl, H.-P., ElMaraghy, H. A., Nyhuis, P. Zäh, M. F., Wiendahl, H.-H., Duffie, N. A., Brieke, M. 2007, Changeable Manufacturing – Classification, Design and Operation, CIRP Annals, Manufacturing Technology 56/2: 783-809. Schuh, G., Harre, J., Gottschalk, S., Kampker, A., 2004, Design for Changeability (DFC) – Das richtige Maß an Wandlungsfähigkeit finden. wt Werkstattstechnik online 94/4: 100-106. Böhmann, T., Krcma, H., 2003, Modulare Servicearchitektur, in: Bullinger, H.-J., Scheer, A. W., Service Engineering – Gestaltung und Entwicklung innovativer Dienstleistungen, Berlin, Springer-Verlag: 391-415. Baldwin, C. Y., Clark, K. B., 2000, Design Rules: The Power of Modularity, Cambridge, MA: MIT Press, 2000. Sullivan, K., Chalasani, P., Jha, S., Sazawal, V., 1999, Software Design as an Investment Activity : A Real Options Perspective, in : Trigeorgis, L. (Edit.), Real Options and Business Strategy: Applications to Decision Making, London: Risk Books, 1999: 215262. Trigeorgis, L., 1996, Real Options – Managerial Flexibility and Strategy in Resource Allocation, Cambridge, MA: MIT Press, 1996. Santiago, L. P., Vakili, P., 2005, On the Value of Flexibility in R&D Projects, Management Science, 51: 1206-1218. Huchzermeier, A., Loch, C. H., 2001, Project Management Under Risk: Using the Real Options

Approach to Evaluate Flexibility in R&D, Management Science, 47: 85-101. [20] Ethiraj, S. K., Levinthal, D., 2004, Modularity and Innovation in Complex Systems, Management Science, 50: 159-173.

192

[21] Richter, A., Steven, M., forthcoming, On the Relation between Industrial Product-Service Systems and Business Models, in: Operations Research Proceedings, Berlin, Springer-Verlag, 2009.

Analysis of Integrated Product and Service Offerings from Current Perspectives of Providers and Customers T. Sakao, E. Sundin IEI – Dept. of Management and Engineering, Linköping University, Linköping, 58183, Sweden [email protected]

Abstract This paper reports the current status of how companies address IPSO (Integrated Product and Service Offerings)-typed business. It will consist of perspectives both from providers and customers mainly from Sweden and Germany. Especially, it selects how providing firms address uncertainty as one focal issue. As a result of interviews, factors from customers are the major source of uncertainty for an experienced company, while services are the major for little experienced companies. In addition, there was found to be a reasonable wish of providers to obtain a formalized way leading to quantitative management of uncertainty. On the other hand, customer incentives are not always clear. While some customers find the IPSO preferable from economic reasons other customers have the opposite recognition. Keywords: Integrated Product and Service Engineering (IPSE), Product/Service Systems (PSS), development, risk, opportunity, uncertainty, business model.

1 INTRODUCTION Manufacturers today regard service activities as increasingly important. Some manufacturing firms are shifting from a “product seller” towards a “service provider” [1]. One reason from the demand side is servicification of customers’ activities, which in some cases means a shift from customers’ owning physical products to getting access to the functionality of products. In the supply side, parallel to the trend above, concepts such as Functional Sales [2] are already found in not only theoretical but also practical fields in industries. Other related concepts include Total Care Products (Functional Products) [3, 4], which comprises combinations of hardware and support services, Product/Service Systems (PSS) [5, 6], Service Engineering [7-9], and Industrial Product Service Systems (IPS2) [10]. Especially, in the field of technical services in the production industry, addressing business models and service design processes are argued to be important [11, 12]. This is also one way to decrease the environmental impacts from the usage phase such as energy and resource consumption, since providers could control the product usage in a better way with their knowledge on products than their customers. It should be noted that in many cases the dominant environmental impacts originate the product usage phase. Our group at Linköping University and the Royal Institute of Technology in Sweden has been, based on our research of Functional Sales, further researching a new engineering way termed Integrated Product and Service Engineering (IPSE) [13, 14]: The group has worked together with over 20 SME on a methodology for developing Integrated Product and Service Offerings (IPSO). IPSO is defined as an integrated offer of physical products (artefacts such as hardware and software) and service activities (activities by people to provide customer value). The IPSE methodology aims to create better prerequisites for firms to develop IPSO that are beneficial for the supplier firm, the customer, and for the society at large.

CIRP IPS2 Conference 2009

193

This paper reports the current status of how companies address IPSO-typed business. Our previous survey was only with providers [15], while this paper will consist of perspectives both from providers and customers. In addition, it selects how providing companies address uncertainty as one focal issue based on a conclusion in our previous work [16]. 2 THE GOAL OF THIS RESEARCH In spite of the high attention on IPSO/IPSE by industries, knowledge/experience to support those companies is not sufficient at present. Especially, the implications on how companies had better run business have not yet been much presented theoretically. Thus, those companies wishing to enter IPSO-typed business cannot be supported enough at present. The final goal of this research is to develop a method/tool to support companies with plan/design IPSO in a better way. Specifically, it is to develop a method/tool to help them with address properties peculiar with IPSO. In our previous work [16], it was found that one of the most crucial impacts on the business with IPSO originates from the shift of these offerings from being static to dynamic. Being dynamic means here that the offerings (what and when to do) may not be determined completely due to the uncertainty along the time dimension. For instance, repair in a full-service contract cannot be projected regarding when and how often. This may be quite obvious, if considering that a contract for service has a time dimension and connote future events as opposed to that for a physical product (except for a period of quality guarantee as free service). Therefore, a crucial issue of IPSE exists in outside of designing/developing a static offer. To be able to design/develop merely a complex but static solution is insufficient, as the business logic of IPSO is often different and, especially, risk-taking tends to increase.

Having the above as the final goal in mind, this paper first of all analyzes IPSO for providers with one focal issue of how to address uncertainty. In addition, the customer perspectives will be brought. As customers need a new mindset for IPSO-type business, the customer incentives and acceptance will be addressed especially in this paper. 3 RESEARCH QUESTIONS OF THIS PAPER The research questions (RQ) for this paper are formalized as follows. The first six are from the provider’s perspective, while the last two are from customer’s. RQ1. What are major drivers/challenges for companies to provide IPSO? RQ2. What are major prerequisites for companies to provide IPSO? RQ3. Which is the major uncertainty factor in providing IPSO? Product, service, or customer? RQ4. How can potential customer value be formalized depending on customer uncertainty/risk and company offer? RQ5. How do companies address customer uncertainty at present during their planning/design and how do companies wish to address them? RQ6. What is a good way to support companies to address customer uncertainty with IPSO during their planning/design? RQ7. What are the incentives for customers of buying IPSO instead of physical products? RQ8. How are IPSO accepted on the market by customers? The three options in RQ3 i.e. product, service, and customer are raised through inspiration by the three important dimensions for designing PSS identified in [17], which are offer, customer, and provider. Definition for some of the terms used in the questions and the dialogue during the interviews is as follows. The categorization of the services, i.e. product-oriented service, use-oriented service, and result-oriented service, is adopted from [6].

♦ ♦ ♦ ♦ ♦

♦ ♦

Uncertainty: state of deficiency of information related to a future event Risk: negative effect of uncertainty on objectives. Risk can be expressed in terms of a combination of the consequences of an event and their likelihood. Opportunity: objectives.

positive

effect

of

uncertainty

on

Product-sales typed contract: a contract in which a company sells a physical product to a customer with a fixed period of quality guarantee without any extra price. Product-oriented service: extra service provided in addition to sales of product. E.g. supply of consumables, maintenance, financing, take-back, and advice on product usage. This is labelled type a. in this paper. Use-oriented service: product leasing/renting/sharing (without ownership transfer) and pay-per-service unit (e.g. contract for using a copier). This is labelled type b. Result-oriented service: activity management service (e.g. office cleaning service and catering) and providing functional result (e.g. keeping harvest losses to an agreed minimum level). This is labelled type c.

4 PROVIDERS’ PERSPECTIVES To answer to the questions regarding the providers’ perspective, nine companies were selected from large-sized manufacturers in Sweden who are interested in providing IPSO. A semi-structured approach with those questions above in mind was adopted for the visiting (face to face) interview except for one (done via telephone). The results are summarized in Tables 1 and 2. The rest of this section explains only three Companies C, E, and H in more detail due to the constraint of space. It should be noted that further description of companies themselves is not given, since it risks the identification of one of large companies in Sweden. 4.1 Company C Company C provides industrial machines which consume a lot of energy at their usage phase. About 75% of the total cost beard by customers is in general energy cost. The cost of the customer, importantly, could be reduced by 30-40% through efforts of the provider. This is critical for Company C to be successful in IPSO business. 1. Needs Company C aims to reach a win-win situation (higher availability/less cost on the customer and higher profit on the company) together with the customer (e.g. by applying the company’s knowledge). They wish to enlarge the types of services; from a. Product-oriented services to include more from b. Use-oriented services and c. Result-oriented services. At the same time, they are interested in including as various contents in their offerings as possible. However, an operator (human) is not needed to be sent from Company C to the customer site, since only software works for operation. 2. Challenges One of the biggest challenges to sell b. Use-oriented and c. Result-oriented services is finding a good way of convincing customers including pricing (payment). Initial investment, if needed, might make obstacle for customers, even though the saving to be obtained exceeds the investment after the pay-back time. In addition, customers often like to take prices originating from cost-based calculation by the company (i.e. dislikes value-based prices) due to a psychological (mental) reason. How to break this mental obstacle is a key challenge. Furthermore, in some cases, there is contradiction among divisions within the customer company (e.g. maintenance division is not willing to accept maintenance services, since some employees in the division lose their jobs). Another challenge is preparing the human resource efficiently. Preparing a service base only for Company C to be ready for services whose time of occurrence can not be forecasted is too expensive against the density of customers at present. Third, their development process should be changed so that the opportunity/risk is evaluated before neither product nor service is fixed in terms of the specification. This process is wished to be carried out by people from the divisions of service development and marketing. Last but not least, the mindset of the employees should be also changed, partially because the functions of the people at the customer company they recently have to face are different from before: They used to work with the technical divisions in most of the cases. 3. Uncertainty Company C considers that uncertainty is among the critical issues associated with IPSO. In the case of c. Resultoriented service, uncertainty both for the company and the customer is the bottle neck to be provided. Among them, customer factors have the largest uncertainty.

194

195

2/3

Company A

N.A.

RQ5: How to Current situation address customer uncertainty Wished situation

Main issues

Factors from customers.

Having high reliability of service and product functioning.

Convincing customers and reducing uncertainty. Preparing service capacity close to every customer.

Convincing customers. Improving skills of sales staff.

Providing a formalized procedure Establishing a more rational providing quantitative data of procedure and a system by making N.A. economy to be shared and managed use of their past experiences. by different divisions.

N.A.

It is vital to address economic costs for the company and some other figures on usage at customers. It may be in a form of software.

Development of offerings.

It varies much.

Factors from customers.

Negotiation with customers. Identifying customer needs.

N.A.

N.A.

Neglecting quantitative uncertainty Including demonstration in the and taking representative numbers period of “trial” service, whose risk to calculate customer benefits. is on the company.

N.A.

Factors from customers.

Ensured customer need/order, and Ensured customer need/order, reliability of services and products. services, and products.

They wish the service development In a formalized way, leading to section to be involved more in the measurement and management. process.

Calculation of costs.

N.A.

They wish to increase both types b and c.

Type a is major at present, and they already have several contracts in types b and c.

20%

Company E

The knowledge of their sales staffs Identifying customer needs. on the benefits and identifying how Organizational structure and to convince customers in the case company culture. of c. result-oriented service.

N.A.

N.A.

Convincing customers. Establishing service bases. Notes: N.A. means that the information was not obtained due to some constraints of the interviews (e.g. time).

RQ6: How to support planning IPSO

N.A.

RQ4: Customer value of IPSO

They wish to increase type c.

Their service business is in most cases in types a and b.

30%

Company D

Economy for the company and the Market needs and differentiation customer (win-win situation). with competitors.

Saving (economic) on customers including risk on customers. The opportunity/risk is evaluated, Not in a formalized way, before neither product nor service sometimes depending on the is fixed. persons.

Factors from customers (operation).

Customers and services for the type b and c, respectively.

RQ3: The major uncertainty with IPSO

Developing a system to support generation of offerings.

Economy for the company.

RQ2: Prerequisites for IPSO

Calculation of economic costs of the company.

N.A.

Having high reliability of service and product functioning. Access to customer’s operation is also a prerequisite.

Challenges of IPSO

Type a is major at present, although they already started to provide types b and c.

40%

Company C

They wish to begin the types b and They wish to shift to the type b but They wish to enlarge the types b c. not to c. and c.

Type a is major at present.

1/2

Company B

They regard all of ensured customer needs, high reliability of service and product functioning, and some insurance.

RQ1.

Drivers of IPSO

Wished situation on service types

IPSO Business Current types of services Type a is provided. provided in general

Current ratio of contracts with services

Table 1. Summary of interview results (1/2)

196 N.A.

N.A.

Main issues

N.A.

Factors from services.

Having high reliability of service and product functioning. In case of type a, customer, product, and service in their order must be ensured. In cases of types b and c, service, customer, and product.

Pricing may be a key.

Integration with customers.

Internal understanding of the meanings of IPSO.

Identifying business case. Prioritization in terms of business potential, and resource requirement.

Improving the above.

Productivity of customers. Saving (time). A certain tool is employed before Using qualitative tools like neither product nor service is fixed. decision tree process.

N.A.

In case of the type a, all from customer, product, and service.

Internal understanding of the Collaboration with dealers. Identifying customer needs. meanings of IPSO. Notes: N.A. means that the information was not obtained due to some constraints of the interviews (e.g. time).

N.A.

N.A.

N.A.

RQ6: How to support planning IPSO

Convenience.

N.A.

RQ4: Customer value of IPSO RQ5: How to Current situation address customer Wished situation uncertainty

Factors from services.

Factors from services.

RQ3: The major uncertainty with IPSO

Internal agreement on how to provide the service offerings.

Ensured reliability of services and Proper involvement from (and products in addition to customer control of) dealers (i.e. service base). need/order.

Challenges of IPSO

Differentiation from competitors (esp. in developing countries).

Drivers of IPSO

They are interested in starting type They are not sure, to enlarge types They are interested in starting type c. They attempt to find new business b and c. b. models using services. Differentiation from competitors, Differentiation from competitors. Increasing the company profits. value creation and optimization of offers. Identifying their strategy on how to Dealing with good collaboration Understanding the implication for sell services, and, to do so, with existing dealers (i.e. service the company. identifying the true customer base). needs.

Type a is major at present. They Only type a is provided at present. look at possibilities to provide types b and c.

They look at possibilities to provide more services in general.

N.A.

Company I

0% (nearly)

Company H

1%

Company G

RQ2: Prerequisites for IPSO

RQ1.

They wish to shift from type a to include more from b, but most probably not from c.

Type a is major at present.

25%

Company F

Wished situation on service types

Current types of services IPSO Business provided in general

Current ratio of contracts with services

Table 2. Summary of interview results (2/2)

4. Risk/Opportunity Company C regard the size of consequence of risk on customers as an indicator to how much customers are interested in buying c. Result-oriented services. Complexity of a provided system would be an indirect indicator. 4.2 Company E Much experience of Company E includes the followings: There were actually some contracts in the form of leasing/renting (type b). However, they quit it due to the less attractiveness in terms of finance for the company. Currently, some independent dealers do leasing/renting only in the North American (not in the European) market, where end users wish more mobility. Another type of experience is that they stopped closing a contract in the form of “profit sharing”, where the company’s revenue is determined depending on the machine performance. The reason is the company E and the customer could not often agree with the quantitative level for the performance from the observed data. 1. Needs They wish to increase both b. Use-oriented and c. Resultoriented services. 2. Challenges A challenge is to identify customer needs depending on customer types as it varies so much among customers. Another is changing organizational structure and company culture. 3. Uncertainty Since their physical products are quite reliable, major uncertainty in services of the types b. and c. is regarded to exist in customers. 4. Risk/Opportunity To contribute to reduce the uncertainty of customers, they provide a “trial” period of several weeks. The company E takes the risk during the period so that they will make financial compensation in case the performance was not satisfactory to customers. This can be regarded as demonstration using the customer-specific hardware and conditions. 4.3 Company H B2B business is 100%, and product-sales typed contracts are (nearly) 100%. They keep high margin now from the sales of physical products, which is a result of the high performance by their R&D activities. I.e. they are a price leader in their market. However, they have so far not found a good strategy to sell services. Their knowledge is so powerful to increase the customer’s productivity. E.g. 10-30 % of the customer’s cost can be decreased by the company’s knowledge. This big impact is striking if compared to the actual cost for customers to buy their products (e.g. only 2% of the total cost). 1. Needs They are interested in starting the type c, and attempt to find new business models using services. 2. Challenges The biggest challenge is identifying their strategy on how to sell services in a better way. To do so, they want to identify the true customer needs. In addition, they want to know how much it costs to provide products/services, for which they do not have any working forms like software at present. 3. Uncertainty They find more risk than opportunity in the type c, and find the major uncertainty in type c from customers. This can be interpreted that the company find high risk in the shift of business model from product sales to service provision.

197

5. Development process They wish to change the degree of adaptation of products for the IPSO, especially in the case of type c. At present, they do not adapt the physical products at all. 5 CUSTOMERS’ PERSPECTIVES To answer to the questions regarding the customer’s perspective, a set of some 60 companies with whom one identical company provides IPSO was selected. The provider, called Company J, is a medium-sized company in Sweden. A semi-structured approach with the questions RQ 7 and 8 in mind was adopted for the interview. Customers included dealers and users in Sweden and Germany. The customers were randomly picked from the CRM system of the IPSO-providing company. Company J is a part supplier that has adapted their product to be used in an IPSO. The provided offering is a combination of product and service that can be used for heavy machinery products (either for newly built products or for the aftermarket). The reason to adopt this Swedish provider is that the company had found that there is much difference in how well the market accepts these new kinds of offers upon selling IPSO to their customers. Especially, the acceptance was much different between Swedish and German customers. In this case it was found that the Swedish customers accepted the business approach of IPSO to a larger extent than the German market. As the research results show various reasons to why the customer acceptance is not as high in Germany as it is in Sweden. The following points summarize the most important reasons; German market

♦ ♦

♦ ♦ ♦ ♦

Much of the new IPSO solution can be solved by traditional methods more easily. The price experience of the new IPSO solution is mixed. Some customers say that it is cheaper than a traditional solution while other customers claim that it is much more expensive. The customers are not aware of the new IPSO solution. Some customers that are aware of the new IPSO solution refer to it as an “emergency solution”. The customer finds it easy to use a well-functioning method/solution than scouting for alternatives. Traditional methods/solutions are considered better and the customer can do the traditional solution themselves.



The incitements for the customers are very low to use the new IPSO solution. Swedish market



♦ ♦

The new IPSO solution is directly considered as an alternative when that kind of service solution is needed since all customers know about the new IPSO solution. The new IPSO solution is seen as a less expensive solution by the customers.

The new IPSO solution is much faster to implement than the traditional solutions. To summarize, the new IPSO is more known and accepted on the Swedish market than on the German market. However, the costs for the new IPSO in comparison to

traditional solutions is not clear if it is preferable or not in comparison to traditional solutions. For a better market acceptance the IPSO provider needs to improve the incitements for their customers, especially on the German market. 6 DISCUSSION AND CONCLUSIONS One of the major findings is that there is a difference in how providers view IPSO depending on the maturity of providers in IPSO business: Providers with much experience (Companies A, B, and C) recognize that the major uncertainty exists in customers, while the others (Companies D and E) consider their services are the most uncertain. Coherently with this, the matured providers raised convincing customers and changing themselves as their major challenges while the others consider understanding the meaning of IPSO business to them is their current challenge. What the latter wish to have is, for example, support to identify their business case with IPSO. Another finding is that a company already with much business of IPSO (Company A) regards, indeed, uncertainty is a critical concept and they wish to have a quantitative tool used for them to design/develop IPSO. How to prepare the service base with both economic efficiency and capacity for providing services is found to be a challenge in Companies C and G. In the case of Company G, it could be another possibility that a new organization under even more (or full) control from the Company G be established. One example form could be a showroom that functions as a marketing/advertisement or selling place as well in a shopping area in a big city (from the viewpoint of economic feasibility). In order to reach a good customer acceptance IPSO provider needs to be more clear and conscious about the customer incentives. If the incentives for the customers are better understood then the customers would more likely choose to change to the IPSO solution. In order to reach new markets with the IPSO they need to be marketed as reaching a win-win situation for the provider and especially for the customers. Company B have been successful in their IPSO marketing by clearly describing the customer benefits. The benefits could for example be; economy, risk reduction, reliability, time-effectiveness. The scales of these benefits (i.e. customer value and costs) are very much dependant on in which industry sector that the IPSO is provided in. Regarding the prerequisites for IPSO, all of the reliability of products and services in addition to ensured customer demands were often raised. Two of the nine companies collaborate with other insurance companies, and one of them sometimes regards this as necessity. Changing development processes of their offerings was proved to be a common challenge. Thus, the research questions are answered to as follows in conclusion. RQ1. What are major drivers/challenges for companies to provide IPSO? A1. Drivers include differentiation from competitors due to increased competition, decreasing costs (occasionally, winwin situation between a provider and a customer), market needs, increasing profits, and improving company brand. On the other hand, challenges include convincing customers, improving a process to generate offerings, reducing uncertainty from customers, improving the skills of the sales staff and the organizational structure for an experienced company and understanding the meaning of IPSO for less-experienced companies. Challenges for firms regardless of their level of experience include facilitating the shift of employee mindset as well as developing good

collaboration intended for IPSO with existing dealers (i.e. service base). RQ2. What are major prerequisites for companies to provide IPSO? A2. Having both high reliability of service (incl. proper service base) and product functioning is a prerequisite in addition to ensured customer demands. This is common to almost all firms, although the orders varied from a firm to another. RQ3. Which is the major uncertainty factor in providing IPSO? Product, service, or customer? A3. Factors from customers are the major for more experienced companies, while services are the major for less experienced companies. RQ4. How can potential customer value be formalized depending on customer uncertainty/risk and company offer? A4. It is different among providers. It could be formalized as saving (on economy or time) on customers including risk based on the idea of value in use. RQ5. How do companies address customer uncertainty at present during their planning/design and how do companies wish to address them? A5. Some companies have tested tools for dynamic assessment of risks. However, it is not widespread in industries. Even a company with much experience does not address (quantitative and qualitative) uncertainty in a formalized way. “Trial” service was employed by some companies to demonstrate the service at the customer. Thus, there is a reasonable wish to obtain a formalized way leading to quantitative management of uncertainty. RQ6. What is a good way to support companies to address customer uncertainty with IPSO during their planning/design? A6. One way would be to adopt a formalized and quantitative tool (software) calculating economy of the provider and the customer using some data in the past. Such tool should address information of the usage of products at customers. This originates partially from interpretation of the interviews by the authors since sufficient spoken needs from the companies were not available. However, this is in line with the future research implication in another literature [17]. RQ7. What are the incentives for customers of buying IPSO instead of products? A7. Customers find incentives such as economy and timeeffectiveness as preferable for the IPSO solution. It is important to find win-win situations for the provider and the customers. RQ8. How are IPSO accepted on the market by customers? Customer incentives are not always clear. While some customers find the IPSO preferable from economic reasons another customer could have an opposite experience. The IPSO maturity on the market is important to achieve the preferable win-win situation. Further immediate works include, first, continuing more interviews for providers and enriching/strengthening the answers to those research questions. After fixing the company needs and challenges, establishing a method and a tool to support designers in companies will be a next work. For this method/tool, the results from the interviews will be incorporated especially what kinds of inputs, factors, and outputs for evaluating risks/opportunities they are concerned of. This is already dealt in another paper from us [18]. Many of the findings of this research study is also in line with the IPSO challenges found when analysing a learning network of large companies in Sweden. These results are further described in detail in [19].

198

To reveal the uncertainty on the customer side, a method to analyse the customers activity cycles developed in the marketing field [20] as well as a method to design services by discovering customer risks developed in the engineering field [21] would be of suggestion. The tool will be implemented on computer software, whose advantage is handling quantitative data with more ease. 7 ACKNOWLEDGMENTS This research is partially supported by the Swedish Association of Graduate Engineers (Sveriges Ingenjörer) as well as by Swedish Governmental Agency for Innovation Systems (VINNOVA). The interview studies explained in Section 5 have been supported by Mr. Nils Lommatzsch. Last but not least, the authors also would like to give sincere gratitude to those anonymous companies who participated in this research. 8 REFERENCES [1] R. Oliva R. Kallenberg, 2003, Managing the transition from products to services, International Journal of Service Industry Management, 14: 160-172. [2] M. Lindahl G. Ölundh, 2001, The Meaning of Functional Sales. In 8th CIRP International Seminar on Life Cycle Engineering – Life Cycle Engineering: Challenges and Opportunities, 211-220. [3] T. Alonso-Rasgado, G. Thompson, B. Elfstrom, 2004, The design of functional (total care) products, Journal of Engineering Design, 15: 515-540. [4] T. Alonso-Rasgado G. Thompson, 2006, A rapid design process for Total Care Product creation, Journal of Engineering Design, 17: 509 - 531. [5] N. Maussang, P. Zwolinski, D. Brissaud, 2006, A Representation of a Product-Service System During its Design Phase - A Case Study of a Helium Liquefier. In 13th CIRP International Conference on Life Cycle Engineering. Leuven, Belgium, 555-561. [6] A. Tukker U. Tischner, 2006, New Business for Old Europe, Greenleaf Publishing, Sheffield. [7] T. Arai Y. Shimomura, 2004, Proposal of Service CAD System -A Tool for Service Engineering-, Annals of the CIRP, 53: 397-400. [8] T. Arai Y. Shimomura, 2005, Service CAD System Evaluation and Quantification, Annals of the CIRP, 54: 463-466. [9] T. Sakao Y. Shimomura, 2007, Service Engineering: A Novel Engineering Discipline for Producers to Increase Value Combining Service and Product, Journal of Cleaner Production, 15: 590-604. [10] H. Meier O. Völker, 2008, Industrial Product-ServiceSystems - Typology of Service Supply Chain for IPS2 Providing, In: Manufacturing Systems and Technologies for the New Frontier - Proceedings for The 41st CIRP Conference on Manufacturing Systems, Mitsuishi, Ueda, and Kimura, Editors. Springer: Tokyo. 485-488. [11] C. Aurich, C. Fuchs, F. DeVries, 2004, An Approach to Life Cycle Oriented Technical Service Design, Annals of the CIRP, 53: 151-154. [12] H. Meier, 2004, Lifecycle-based Service Design for Innovative Business Models, Annals of the CIRP, 53: 393-396. [13] M. Lindahl, E. Sundin, A. Ö. Rönnbäck, G. Ölundh, J. Östlin, 2006, Integrated Product and Service Engineering – the IPSE project. In Changes to Sustainable Consumption, Workshop of the Sustainable Consumption Research Exchange (SCORE!) Network, supported by the EU's 6th Framework Programme. Copenhagen , Denmark

199

[14] M. Lindahl, G. Ö. Sandström, E. Sundin, A. Ö. Rönnbäck, J. Östlin, 2008, Learning networks: a method for Integrated Product and Service Engineering – experience from the IPSE project, In: Manufacturing Systems and Technologies for the New Frontier - Proceedings for The 41st CIRP Conference on Manufacturing Systems, Mitsuishi, Ueda, and Kimura, Editors. Springer: Tokyo. 495-500. [15] T. Sakao, N. Napolitano, M. Tronci, E. Sundin, M. Lindahl, 2008, How Are Product-Service Combined Offers Provided in Germany and Italy? – Analysis with Company Sizes and Countries -, Journal of Systems Science and Systems Engineering, 17: 367–381. [16] T. Sakao, M. Lindahl, E. Sundin, A. Ö. Rönnbäck, O. Tang, 2008, Addressing Uncertainty as a Key for Successful Integrated Product and Service Offerings: Literature Review and Company Interview. In The Swedish Production Symposium. Stockholm, 253-260. [17] T. Sakao, G. Ö. Sandström, D. Matzen, 2009, Framing design research for service orientation through PSS approaches, Journal of Engineering Design, provisionally accepted to appear. [18] M. Lindahl, T. Sakao, A. Ö. Rönnbäck, 2009, Business Implications of Integrated Product and Service Offerings. In CIRP IPS2 Conference 2009. Cranfield, in print. [19] E. Sundin, G. Ö. Sandström, M. Lindahl, A. Ö. Rönnbäck, T. Sakao, T. Larsson, 2009, Challenges for Industrial Product/Service Systems: Experiences from a learning network of large companies. In CIRP IPS2 Conference 2009. Cranfield, in print. [20] S. Vandermerwe, 1993, Jumping into the customer's activity cycle: A new role for customer services in the 1990s, Columbia Journal of World Business, 28: 4665. [21] V. Panshef, E. Dörsam, T. Sakao, H. Birkhofer, ValueChain-Oriented Service Management by Means of a ‘Two-Channel Service Model’, International Journal of Services Technology and Management, 11:1, 4-23, 2009.

Uncertainty challenges in service cost estimation for product- service systems in the aerospace and defence industries J. A. Erkoyuncu1, R. Roy1, E. Shehab1, P. Wardle2 1 Decision Engineering Centre, Cranfield University, MK43 0AL, Bedfordshire, UK 2 BAE Systems Integrated System Technologies, Eastwood House, Glebe Road, Chelmsford, CM1 1QW {j.a.erkoyuncu, r.roy, e.shehab}@cranfield.ac.uk, [email protected]

Abstract Contracting for availability is expected to become more prevalent for product -service systems (PSS) in the aerospace and defence industries. These contracts tend to transfer responsibilities for the operational phase from the customer to the supplier. In parallel, with operational life spans spanning several decades, the ability to deal with uncertainty in cost estimation for support activities is becoming critical. This paper outlines challenges within this process derived from literature as well as issues that were highlighted during interviews with four major defence and aerospace organisations. Keywords: Product Service Systems, uncertainty, cost estimation, service

1 INTRODUCTION The product-service system (PSS) approach, which integrates products and services to varying degrees, has lately attracted interest in the defence and aerospace industries as a candidate for availability contracting. In these industries the technical-PSS (t-PSS) concept applies where this is defined by the major characteristics of relatively higher monetary value of product core, a physical product core that is integrated with services, and a business to business relationship [1]. Examples that have followed this trend are the ‘Total Care’ and ‘Power by the Hour’ packages offered by Rolls-Royce which are focused on provision of in-service support against performance measures such as equipment availability. Taking a PSS approach drives a life cycle view of system provision for suppliers. This in essence creates a number of challenges that are illustrated in Figure 1. Supply Chain - Performance - Sustainability

Technical challenges - WLCC estimation - Design

Business Long term PSS challenges

- Profitability - Affordable - Strategies

- Demand

Socio-Technical Figure 1. Long term PSS challenges (Adapted from Shehab and Roy [2]) Equipment availability is a function of its mean time between failure (MTBF) and mean time to repair (MTTR). MTBF is a characteristic of equipment design and is generally independent of the arrangements for in-service support, but MTTR depends directly on the support CIRP IPS2 Conference 2009

200

solution which can be characterised in terms of process, outcomes, and cost. Process includes detection of failures (including planned and unplanned maintenance) and rectification (including repairs, overhaul, retro-fitting, upgrades and obsolescence management) [3]. Under traditional contract arrangements suppliers are typically paid according to the throughput of ‘spares and repairs’ and other transactions such as mitigations for obsolescence. The sales value of each transaction with respect to costs incurred, which may be negotiated caseby-case, determines the supplier’s profitability whilst transaction throughput determines the affordability for the customer. The throughput is under the customer’s control and they will typically manage their demand rate within internal budgetary constraints, e.g. by prioritising transactions and dealing with simple cases in-house. Under availability contracting arrangements the MTTR or other performance criterion is made the essence of the contract. At the time of bidding, the supplier offers a fixed price to the customer whilst assuming responsibility for estimating the cumulative number of transactions needed to sustain the MTTR. The supplier must accept the risk that, if they underestimate the number of transactions necessary, profitability will be reduced. Such estimates need to anticipate a range of contributory technical, commercial, financial, and behavioural risks and uncertainties that are exacerbated because of the need to look-ahead over a long period of time. The affordability for the customer is now independent of demand rate but they still have an interest in using this measure as a basis for comparing offers from alternative suppliers to achieve the best value-for-money. Risk is the threat of a loss (e.g. financial, timescale, or performance) from an unwanted event. Uncertainty is the difference between an anticipated or predicted outcome (e.g. a cost estimate) and the confirmed outcome (e.g. the actual cost). To be able to handle uncertainty, one needs to examine its sources. Broadly, these are incomplete information, disagreements between information sources, imperfect communication, and variation in circumstances. The PSS sets the context for this study in which the goal of the research is to enhance rigour in cost estimates by means of better handling of uncertainty, particularly during the in-service phase. Availability contracts, which are the

commercial arrangements under which the PSS is procured and delivered, take the whole life perspective of the equipment/system life cycle. In essence, when the PSS supplier takes decisions such as whether to bid for a contract or accept one when offered, they need to do so based on an understanding of profitability over the duration of the lifecycle including the inherent uncertainty. This necessitates better prediction of uncertainty for availability contracts than has been typical of traditional contracts in the past because the contract timescales are much longer, and ownership of uncertainty has been transferred from customer to the supplier - typically on a fixed-cost basis. Figure 2 illustrates how the main concepts within this paper are interlinked. PSS

Whole life Cost cycle cost Estimation (WLCC)

Uncertainty Modelling

Figure 2. Inter-linkage of concepts in the paper The research covered by this paper focuses on defining the current practice in uncertainty modelling in cost estimation, highlighting some of the major estimating uncertainties arising from the shift to contracting for availability and challenges incurred in incorporating service uncertainty into cost estimation. It has been conducted within the Product-Service System Cost Project (PSS-Cost) at Cranfield University. Section 2 explains the methodology; Section 3 gives the outcome of a literature review on key concepts to the paper. Section 4 describes the current practice in uncertainty modelling for cost estimation, service uncertainties in availability contracts and the major challenges for industry. Sections 5, gives conclusions and implications for further work. 2 METHODOLOGY The research commenced with a literature review to understand the drivers of the move towards service oriented contracts within the defence and aerospace industries. Subsequently, the research aimed to understand how the concept of WLCC has become associated with, and necessary to, the PSS approach. The literature review also briefly covered trends in cost estimation methods including uncertainty prediction, although such prediction at the in-service phase was found to be very limited. The main databases used were ProQuest, Scopus, Web of Knowledge, Science Direct, EBSCO, and Google Scholar. No specific limit was set in terms of date of publication during the search process but the references found were all published between 2001 and 2008. This may be due to the field’s short history. The following key words were used in the search: (whole life cycle) cost estimation, bid phase, uncertainty definition, uncertainty modelling, product service systems, service, service cost estimation, service cost uncertainty. The outcome of the literature review was used to devise semi-structured interviews with four major partners in the project. These comprised three defence companies and one defence customer (UK MoD). The research with the partners was designed to confirm which of the techniques for uncertainty prediction suggested by the literature review were actually being used, to capture experience of their use, and to elicit challenges for future research. The questionnaires used in the interviews were piloted with BAE Systems’ sponsoring manager.

A total of over 33 hours of semi-structured interviews were conducted with cost engineers, project managers, support managers, engineering managers, and functional experts (e.g. on risk and uncertainty). The PSS-Cost project members mostly attended the interviews together, apart from interviews that were held with functional experts. As a result, the linkages among research topics core to the project (design rework, obsolescence management, uncertainty and affordability assessment) were better understood. The duration of each interview did not exceed two hours. The researchers took notes during each interview and observations were reflected back in the form of a report for validation. The interviewees shared documents with the researchers prior to interviews to speed-up the process of learning about current practices in the collaborating industrial organisations. Some interviews focused on enhancing understanding of these documents. The range of current practices included: stakeholder involvement in the bidding process, life cycle management frameworks, software and engineering estimation guidelines, service definition elements in availability contracts. Owing to the limited time that industrial participants could provide the researchers needed to select their questions carefully. To begin with, these included basic questions to confirm shared understanding of the definitions of terms such as uncertainty and risk, the elaboration of types of uncertainties, and the way uncertainties change during the lifetime of an availability contract. Results from interviews were analysed by developing mind maps designed to highlight commonalities and differences in current practices and challenges experienced across the projects studied. These were again reflected back for validation. The deliverable outputs at the end of year one of the PSS Cost project (December 2008) comprised reports on (1) the state of the art in cost estimation, (2) key challenges in cost estimation within the defence and aerospace industry, and (3) candidate activities for improvement in years two and three. 3

LITERATURE REVIEW

3.1 PSS and service PSS offerings have generated interest in the defence and aerospace industries because of (1) pressure in national defence budgets in most countries including the UK, (2) the UK defence customer’s ambition to transfer financial uncertainty from itself to industry, and (3) UK industry’s ambition to grow its share of the diminishing defence budget in terms increased span across both the lifecycle (e.g. CADMID1) and defence lines of development (e.g. TEPIDOIL2). These specific interests in the PSS approach on the part of the defence and aerospace industries are backed-up by others in industry at large. In the literature these include environmental benefits and system level cost reductions. These benefits are driven by increasing effectiveness in utilisation of equipment [4] and an emphasis on the 1

The Concept, Assessment, Demonstration, Manufacture, Inservice, Disposal cycle has been used by the United Kingdom Ministry of Defence (MOD) since 1999, when it was devised as part of the Smart Procurement initiative, since replaced by Smart Acquisition, to deliver equipment capability within agreed performance, cost and time parameters. 2

The United Kingdom’s defence lines of development are training, equipment, personnel, information, doctrine and concepts, organisation, infrastructure and logistics

201

functionality and capability of the combined productservice system rather than the product itself [5, 6, and 7]. Furthermore, contracting for availability helps achieve value for money through co-creation of value between supplier and customer [8]. The most recent research looks at how the desire for co-creation of value is driving traditional product suppliers to transition into service delivery organisations. 3.2 Whole Life Cycle Cost (WLCC) When considered in a rigorous manner WLCC analysis guides formulation of the PSS proposition during the bidding process from the technical, economic, and contractual perspectives and helps analysts to compare alternative propositions by taking account of all future costs. Technical metrics may involve functionality, performance, effectiveness, reliability, maintainability, supportability, or recyclability. Economic metrics may include initial cost, affordability, or profitability. An availability contract may make reference to any of the technical or economic metrics. WLCC analyses offer better uncertainty assessment techniques, which are limited in standard methods [9]. Better consideration of uncertainty improves the chances of actual cost outcomes being within cost predictions, and this in turn allows contingency for uncertainty to be taken out of the price quoted by the supplier to a customer. This is beneficial to the supplier, especially in industries that have tight competition [10], because it improves their position against competitors. In the defence industry, the importance of the reliability of a cost estimate has increased since the public procurement policy was put in place during the 1980’s [9]. This policy put the concept of ‘value for money’ at the forefront. Despite these motivating factors the overall growth in adoption of WLCC has been relatively slow [10]. This can be attributed to a number of factors. For instance, estimators tend to be sceptical about adopting emerging techniques for WLCC estimating whose efficacy is unproven. Also, in committing one self to a long lasting contract, difficulty arises when the uncertainties in the estimate exceed the nominal profit. Other reasons for the slow uptake may be attributed to the short term view of management and/or the influence of reward systems that favour lower costs on an immediate basis [10]. A systematic approach that enables the risk and uncertainty in WLCC estimates to be reduced will enhance uptake in the industry [9]. 75 to 80 percent of the WLCC is often committed before contract award, or shortly after, because the early design activity must scope the product solution and service solution concurrently owing to interdependency. In fact this behaviour is often driven by requirements in the customers request for proposal that ask for early visibility of the WLCC predictions. The percentage uncertainty in cost estimates for the product solution (including development, manufacturing, testing, integration, certification, and acceptance) is often much less than that of the support solution. Furthermore, by the time cost outcomes become evident in the first year or two of the support phase it is difficult to modify the support solution from either the technical or contractual viewpoint (e.g. in the event that it becomes necessary to pre-empt predicted cost overruns). Some availability contracts mitigate this problem by means of the ‘evergreen renewal’ principle that permits renegotiation at intervals, e.g. every five years for a thirty year contract, but this is not seen as an ideal solution.

202

For these reasons it is a priority to address uncertainty in the cost estimating of the support phase of availability contracts, and to prepare initial estimates at an earlier phase of the lifecycle than under traditional arrangements where the product solution and support solution are decoupled under separate, sequential contracts. It was noted in the course of interviews conducted by the PSS Cost project team that disposal or termination rarely form part of either traditional or availability contracts in the defence and aerospace industry. The prediction of costs and consideration of uncertainty for this phase of the lifecycle is therefore not covered here but remains an opportunity for further work. 3.3 Representation of WLCC The segmentation of the life cycle enables a cost breakdown structure (CBS) to be produced for use in allocating budgets to individual cost centres and recording actual spend. The approach taken in developing the CBS may be variously driven by a legacy structure, customer requirement, the product breakdown, project organisation, or functional organisation. Ideally, industry would like to work towards CBS standardisation but this is difficult to achieve. In practice, estimators apply their experience to map data from one CBS to another in order to inform future estimates from data collected on past projects. It is inherent in the process of retrieving data from past projects, possibly as long as several decades ago, that the original knowledge of the scope of each element will have been lost. The cost estimator typically mitigates this by applying his/her own experience and researching the experience of others but, particularly in the defence and aerospace industries where project lifecycles can exceed a working lifetime, better knowledge management techniques are required so that this understanding can be persisted without relying on the memory of individuals. 3.4 Cost estimation methods The stages of the lifecycle for the PSS, for example as in the CADMID cycle, are widely varied in scope and scale and require a variety of methods to most effectively predict the respective costs. Three well-recognised methods are mentioned in [11] and a fourth has been identified from research conducted under the PSS Cost project: estimating by analogy: reads-across the cost outcomes from past projects having similarities with the one being estimated, traditionally used for the non-recurring engineering costs during development and unit cost during manufacture. activity based costing: identifies activities in an organisation and allocates the cost of each (e.g. in terms of man-hours, facilities, and materials) to products and services according to their actual consumption. the parametric method: derives cost estimating relationships (CERs) that can predict cost as a function of the basic attribute(s) of an item (e.g. weight, volume, complexity) [12]. Models in this group include regression analysis, fuzzy logic, and neural networks [13]. extrapolation: particularly for operational and support phases, experience from prior contracts (e.g. spares and repairs) can inform projections of future costs when the incumbent supplier now needs to estimate for follow-on availability contracts.

Selection of the method largely depends on the available data. Relatively new methods such as fuzzy logic and evidence theory [13, 14, 15] are available but have not been widely adopted in industry because the uncertainty in the estimates is too large to assure the supplier of profitability and the customer of affordability. Ideally, more than one method is applied to a given estimating challenge, and this in itself can reduce uncertainty in the prediction of cost. 3.5 Uncertainty in cost estimation The understanding of the level of uncertainty in the available data, such as a cost estimate, is an important factor in making reasoned decision [16], such as when the customer is choosing between the PSS solutions offered by alternative suppliers. Well-established techniques have been developed to manage uncertainties in predicted costs that arise during the development and production phases of the PSS lifecycle (e.g. owing to technical, financial, timescale, and quality factors). Much research has applied methods that derive from general probability theory to establish suitable methods for particular scenarios in estimating uncertainty. Uncertainties in predicted costs that arise during the inservice phase have, however, proved more difficult to manage (e.g. owing to obsolescence and supply chain disruption). Understanding of risk and uncertainty was shaped by social scientists Frank Knight and John M. Keynes who raised definitions for these terms. In 1921 Frank Knight differentiated between them. Risk is concerned with the loss that might arise (e.g. cost, time, or performance) depending on whether a given event may or may not happen. Uncertainty is concerned with events which are certain to happen (e.g. obsolescence) but whose effect is hard to predict (e.g. the number of obsolescence events over the lifetime of the PSS and the cost of their mitigation). Figure 3, depicts the way in which the accuracy of cost predictions for a given scope of supply, such as the support solution being discussed here, improves over time as the uncertainties are progressively resolved. Such depictions frequently show cost predictions bounded by high (straight line) and low (dotted line) confidence levels, e.g. 90% and 10% respectively. Project Phases

100 80

Worst range of expected accuracy

Uncertainty

60

Best range of expected accuracy

40 20 0 -20 -40 -60

Rough Feasibility Preliminar Order of y estimate Magnitude studies

Definitive

Detailed Estimates

estimate

Calendar Time (No Scale)

Figure 3. Estimating accuracy trumpet [17] 4

IN-SERVICE PHASE COST UNCERTAINTY: INDUSTRIAL PERSPECTIVE Interest in becoming service oriented is certainly evident amongst suppliers and customers in the defence and aerospace industries. However, in seeking to devise, negotiate, and deliver availability contracts that are both

profitable for the supplier and affordable for the customer, an expanded set of uncertainties need consideration compared with those arising from traditional contracting arrangements. This is driven by the transfer of uncertainties from the customer to the supplier. This section, by taking an industry perspective, aims to highlight current practice in integrating uncertainty into cost estimation, describes the major factors that influence uncertainty in availability contracts, and indicates the key challenges that are faced in the defence and aerospace industries at the bidding stage. 4.1 Current practice in uncertainty and cost estimation at the bidding stage Based on the industrial interviews it was found that availability contracts are currently being awarded on the basis that they span the manufacturing and in-service phases of the CADMID lifecycle but the bids are often prepared and submitted in earlier phases. The in-service phase for current contracts typically runs from ten to thirty years but there is often an ‘evergreen renewal’ arrangement in the terms and conditions which allows the contract to be re-baselined at shorter intervals, typically five years. On each iteration of this interval the estimating uncertainties become smaller as experience of cost outcomes increases and the time to contract completion decreases. Depending on the concept of the PSS solution, the individual equipments of which it is comprised may have a shorter design life than that of the PSS as a whole. This approach can mitigate problems such as obsolescence at the equipment level provided the successor equipment has the same form, fit and function, i.e. on the “open architecture”, “open standards” principle. In these circumstances it can be helpful to consider uncertainties on two levels – first at overall PSS level, and second at equipment level. Furthermore, each level has its own CADMID lifecycle. For a PSS of significant size (e.g. £100 million or greater in contract value), industry tends to start working on design solution and in-service support solution as long as three to four years before winning a contract, i.e. at the concept or assessment phase of CADMID. During these phases a number of technical and business reviews are conducted to assessing the feasibility, affordability, and profitability of the potential project. These reviews, which take place on a cross-functional basis (e.g. engineering, procurement, operations, commercial, and finance) inform decisions such as “bid / no bid” and whether to accept an availability contract for the manufacturing and in-service support phases if offered. Cost models are established at the earliest possible phases of the CADMID lifecycle and are evolved as the lifecycle proceeds. Lower levels of detail are progressively added to the design solution and in-service support solution, e.g. by clarifying and elaborating requirements of the PSS with the customer, by performing trade studies to examine design or in-service support solution options, and by producing derived requirements to capture design decisions. Although it is usually possible to enumerate risks and uncertainties early on in this process, the challenge for industry is to quantify them sufficiently, and with adequate levels of confidence, to support discussions on affordability (with the customer) and profitability (internally and with the supply chain). For example, cost estimates at the concept phase are based on high level assumptions and use parametric or analogy based tools. It is often not

203

until the assessment stage or later, when the maturity of the design is progressing, that it becomes possible to quantify uncertainty by means of the more accurate estimating methods. This challenge can be characterised slightly differently for each of the bidding scenarios in the UK defence and aerospace industries. These are: 

The competitive situation: the price that the customer agrees is governed by the competition in the market



The single bid situation: the price is established through negotiation between customer and supplier. This involves the customer having visibility of the supplier’s costs and cost-to-price calculation including profit margin. The trend from traditional contracting towards availability contracting sustains the challenge for suppliers to be confident in the affordability of their offering (e.g. to be assured of both winning the bid in the competitive situation, and of winning the value-for-money argument with the customer in the single bid situation). It has also increased the challenge for them to be confident in their own profitability as a result of the transfer of risk and uncertainty from customer to supplier, particularly in the single bid situation. The usual response of industry to these challenges is to manage uncertainty using a framework as illustrated in Figure 4. Each step has an overlap with the next and may require individuals to return to previous steps to make updates (e.g. in the light of new data) or to consider new uncertainties that had not been previously identified.

which contains a suitable database developed in collaboration with BAE Systems3. Estimating practitioners in both the supplier and customer communities have a preference for commercial tools because this simplifies verification and validation of cost models. Unfortunately, commercial tools are not always able to cope with specific circumstances, for example the phased introduction or withdrawal of platforms from a fleet concurrent with mid-life update and/or technology refresh and/or spares scavanging. In this case special-to-purpose models in Microsoft Excel (or similar) are required and investment in verification and validation for these must be accomodated in the cost of bidding. Although risk management is outside the scope of this paper, functional experts interviewed in the industry highlighted difficulties in segregating uncertainty and risk in cost estimation. This means that, in some instances, risks and uncertainties are incorrectly categorised and may cause unreliable cost estimates. Furthermore, this problem does not only occur when risk registers are first created at the beginning of projects because they may change their categorisation with time. Driven by time constraints (e.g. tight timescales during the bidding phase), some suppliers find that there is insufficient time to analyse risks and uncertainties sufficiently. 4.2 Service uncertainty in availability contracts Moving from traditional contracts towards those based on availability necessitates consideration of a wider set of uncertainties. There are two major drivers for this: 

Step by step approach to deal with uncertainty

Identify Assess Analyse Reduce

Control

Resolve issues deriving from uncertainty

Additional uncertainties arising from the inservice phase of CADMID. Under traditional arrangements these additional uncertainties were managed under follow-on contracts signed towards completion of the manufacturing phase (e.g. for spares, repairs, training services, obsolescence management, technology refresh, and disposal). Under availability contracts consideration of uncertainties is bundled and concurrent for both suppliers and customers. There is also the challenge of offering new services that were not traditionally offered (e.g. training).



Figure 4. Managing uncertainty The uncertainty associated with a cost estimate at a terminal node in a CBS is ideally expressed as a probability distribution with “minimum”, “maximum” and “most likely” costs and is typically represented as a line item in for example a Microsoft Excel spreadsheet. The estimate at a terminal node typically refers to an equipment cost in the manufacturing phase of CADMID (e.g. development or manufacture of a “line replaceable unit”, a “hardware configuration item”, or a “software configuration item”), or a service operation / transaction in the in-service support phase (e.g. spares and repairs). Equipment costs in the manufacturing phase are typically associated with one-off non-recurring engineering tasks or unit production. They are often aggregated at higher levels of the CBS using Monte Carlo simulation tools such as the riskHive suite [19], the Predict! suite [20], or Crystal Ball [21]. Support costs in the in-service phase typically arise repeatedly. These are often aggregated using tools that can simulate the rate of occurrence. Examples include Vari-Metric [22], OPUS 10 [23], or Tecnomatix PSST [24]

204

Availability contracts also demand ‘left shift’ of the point-in-time at which some uncertainties are addressed yet the information needed to resolve some of them may not have been developed at bid time (e.g. the design of the support solution cannot be firmed-up before the product solution has been designed, the supplier’s initial assumptions on in-service environment have to be clarified and confirmed, and the boundary between supplier and customer responsibilities may not have been negotiated). These drivers necessitate better consideration of the types of uncertainties at the bidding stage to facilitate cost estimates having the best possible accuracy. Based on interviews with industry the major categories of uncertainties concerned are illustrated in Figure 5.

3

The PSST tool in combination with the BAE Systems database was formerly known as RAMLOG.

Materials: Time?

Technology:

Location? Variety?

Obsolescence?

Availability?

Refresh?

Contract requirements?

Uncertainty Service network: Performance? Sustainability?

Macro factors:

Service

Economy? Regulation? Labour requirements: Skill set?

Figure 5. Types of uncertainty 4.3 Challenges in incorporating service uncertainty into cost estimation Efficient consideration of uncertainty in cost prediction is essential in order that availability contracts can be successfully negotiated and delivered in the future. The evidence from both the literature review and the industrial interviews is that uncertainty is driven by both the lack of information and poor timeliness in its availability. Along with this aspect, provision of new services requires the development of a new knowledge set for suppliers. This section presents a number of key challenges highlighted during interviews: 

Equipment reliability (MTBF), repair time (MTTR), and the demand rate for spares are important sources of uncertainty in the support costs for a PSS. Even if the performance criteria for an availability contract are not directly based on these metrics the uncertainty involved in them will affect performance indirectly. The challenge is to improve prediction of these drivers.



At all phases of the CADMID cycle, and particularly at the in-service support phase, the ability to deliver the contracted level of performance is highly dependent on the supply network. Sustaining performance of a supply network is a challenging task. Furthermore, the long duration of availability contracts increases the chance of disruption(s) occurring.



Improved monitoring of supply chain performance would give greater opportunity for proactive intervention before problems arise, and hence reduce uncertainty. This concept is inspired by Earned Value Management (EVM) as applied at the manufacturing phase of the CADMID cycle but the concept will need adaptation for the in-service phase.



The need for improved estimating techniques that can take account of the increased range and scale of uncertainties typical of availability contracts compared to traditional contracts.



To be able to agree support based contracts it is necessary to have a common understanding between the customer and the supplier of the uncertainties. This is a particularly challenging aspect which requires a synergy in approaches to consider uncertainties. This may be achieved by means of a common framework utilized to capture uncertainties at the bidding stage.

5 CONCLUSIONS AND FUTURE WORK The trend towards availability contracting adds to the challenges faced by the defence and aerospace industry

in estimating long term contracts with sufficient accuracy. A substantial element of this challenge is the uncertainty that attaches the in-service support phase of the contract which, at the time of bidding, can extend several decades into the future. To improve understanding of this challenge this project proposes to develop a list of generic uncertainties and potential mitigations that typically apply to availability contracts, and to test this with the industrial collaborators. At the same time, the suppliers are often invited to take on additional scope such that they increasingly need to take a holistic view of the contracts and focus less on intermediate deliverables and more on overall outcomes. This increase in scope can introduce additional sources of uncertainty. One of the major sources of uncertainty that lies outside the control of the supplier is supply chain disruption, yet in the limit, the supplier must accept the cost impact of supply chain failures and the responsibility of resolving these. As part of future research the PSS Cost project will develop an estimating framework for availability contracts that encapsulates management of supply chain disruption within the estimating methodology. The research will particularly focus on reflecting the dynamic nature of service supply chains. 6 ACKNOWLEDGMENTS The authors would like to thank the EPSRC, Cranfield IMRC for funding this research. The continued support given by Industrial Collaborators is also appreciated. 7 REFERENCES [1] Tukker A., Ursula T., (2006) “Product Service as a research field: past, present and future. Reflections from a decade of research” Journal of Cleaner Production. 14(17), pp 1552-1556 [2] Shehab E.M. and Roy R., (2006) “ProductService Systems: Issues and Challenges”, The 4th International Conference on Manufacturing Research (ICMR 2006), Liverpool John Moores University, 5th – 7th September 2006, pp 17-22 [3] Ling D., Roy R., Shehab E., Jaiswal J., Stretch J., (2006) “Modelling the cost of railway asset renewal projects using pair wise comparisons”, Proceedings of the Institution of Mechanical Engineers, Part F, Journal of Rail and Rapid Transit, IMechE, 220(4), pp 331-346 [4] Mont O., (2002) “Clarifying the concept of Product service system” Journal of Cleaner Production, 10, pp 237-245 [5] Ayres R.U., (1998) “Towards a zero emissions economy” Environment Science Technology, 32(15), pp 366- 367 [6] Maxwell D., Vorsty R., (2003) “Developing sustainable products and services “ Journal of Cleaner Production, 11, pp 883-895 [7] Baines T., Lightfoot H., Evans S., Neely A., Greenough R., Peppard J., Roy R., Shehab E., Braganza A. Tiwari, A. Alcock, J. Angus, J. Bastl, M., Cousens A., Irving P., Johnson M., Kingston J., Lockett H. & Martinez V., (2008) "State-of-the-art in product-service systems", Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, 221 (10), pp 1543-1552 [8] Irene C. L. Ng, 2008, “The pricing and revenue management of service, A strategic approach”, Routledge

205

Advance in Management and Business Studies, Routledge Taylor & Francis Group, pp 81-85 [9] Boussabaine A., Kirkham R., (2004) “Whole Lifecycle Costing: Risk and risk responses” Blackwell Publishing, 1st Edition [10] Bradley M., Dawson R., (1999) “Whole Life Cost: The Future Trend in Software Development” Software Quality Journal, 8, pp 121-131 [11] Asiedu Y., Gu P., (1998) “Product life cycle cost analysis: state of the art review” Internatioal Journal of Production Research, 36(4), pp 883-908 [12] Curran R., Raghunathan S., Price M., (2004) “Review of aerospace engineering cost modelling: The genetic causal approach” Progress in Aerospace Sciences, 40, pp 487-534 [13] Oberkampf W., Helton J., and Sentz K., (2001), “Mathematical representation of uncertainty”, AIAA Paper 2001-1645, April. [14] Kishk M., (2004) “Combining various facets of uncertainty in whole-life cost modeling” Construction Management and Economics, 22(4), pp 429-435 [15] Harding A., Lowe D., Emsley M., Hickson A. and Duff R., (1999), “The role of neural networks in early stage cost estimation in the 21st century” COBRA 1999: The Quantitative and qualitative cost estimating for engineering design“[Online] Webpage, http://www.rics.org/Practiceareas/Management/Business management/Finance/role_of_neural_networks_1999010 1.html, Access date: 12/01/2008 [16] Thunnissen D., (2005), Propagating and Mitigating Uncertainty in the Design of Complex Multidisciplinary Systems, PhD Thesis, California Institute of Tech nology Pasadena, California [17] Smart Acquisition 1, Safety and Engineering, “Through life project management” [Online] Webpage, http://www.ams.mod.uk/content/docs/tlmguide/definitn.ht m, (Access date: 17/12/07) [18] Baines T. S., Asch R., Hadfield L., Mason J. P., Fletcher S., Kay J. K., (2005), “Towards a theoretical framework for human performance modelling within manufacturing systems design”, Simulation Modelling Practice and Theory, 13, pp 486-504 [19] riskHive suite, [Online] Webpage, http://www.riskhive.com/,(Access date: 10/10/2008 [20] Predict! Suite, [Online] Webpage, http://www.riskdecisions.com/, (Access date: 10/10/2008) [21] Crystal Ball, [Online] Webpage, http://www.oracle.com/crystalball/, (Access date: 10/10/2008) [22] Zamperini M. B., Freimer M., (2005), “A simulation analysis of the vari-metric repairable inventory optimization procedure for the U.S. Coast Guard”, Proceedings of the Winter Simulation Conference, 4-7 December 2005. [23] OPUS 10, [Online] Webpage, http://www.systecon.se, (Access date: 10/10/2008) [24] Tecnomatix PSST, [Online] Webpage, http://www.plm.automation.siemens.com/en_us/products/t ecnomatix, (Access date: 11/10/2008)

206

Identifying Risk and its Impact on Contracting Through a Benefit Based-Model Framework in Business to Business contracting: Case of the defence industry I. Ng and N. Yip Business School, University of Exeter, Streatham Court, Rennes Drive, Exeter EX4 4PU, United Kingdom [email protected] and [email protected]

Abstract Two defence contracts for availability are studied in the attempt to better understanding the provision of service in a maintenance, repair and overhaul environment that is contracted on the performance of the equipment, rather than merely providing equipment. The nature of the contract changes the dynamics of the delivery, bringing behavioural issues into the forefront, with both customer and firm focused on value cocreation, rather than each party’s contractual obligation. Our study provides a customer focused approach that exposes gaps in the way organizations approach their service provision in MRO. We argue that customer involvement and behavioural issues in the co-creation process has to be factored into the design and delivery of traditional MRO delivery systems. This paper uncovers four areas that pose risks to performance based contracts and are crucial in the design of services under such a contractual environment and provides a research agenda for future studies in this area. Keywords: Co-Creation, Performance Based-Contracts, Risks; Contract Design

1 INTRODUCTION Performance based contracts (PBC) are about contracting on performance, rather than tasks or inputs by the service provider. For example, in the case of Rolls Royce, the service provided to maintain engines is being remunerated on the basis of how many hours the engine is in the air – a concept known as ‘power by the hour’. Recently, it has been reported that there has been increased interests in PBC from service firms keen on witnessing significant improvements in costs, customer satisfaction and financial audits [1]. A critical element of PBC is the clear separation between the customer’s expectations of service and the firm’s implementation [2]. In short, the contract explicitly states the outcome of the service without specifying how it is to be achieved, e.g. consistent power by an engine. The contractor then determines how to achieve that outcome, usually will less intervention from the customer. As a result of this flexibility in the arrangement, PBC should promote new and improved ways to manage tangible and intangible resources by the firm to achieve outcomes that are of benefit to the customer. Such a radical change in the approach to contracting has caused confusion among suppliers. Nonetheless, little is still understood about the characteristics of PBC, further suggesting that academic literature offers little guidance with respect to how such contracts should be executed [3,2]. Under the service dominant logic in marketing literature, Vargo and Lusch (2004) argue that “customers are CIRP IPS2 Conference 2009

207

always co-producers” and “co-creators of value”. As such, marketing researchers have proposed that firms do not really provide value, but merely value propositions [4] and it is the customer that determines the value and cocreates it with the firm. Co-creating value then implies that customers’ roles are moving from being isolated to being connected to the firm, passive to active and being unaware to being informed. Within PBC, where outcomes are a result of value co-creation, there is then a need to understand the role of the customer in the firm’s processes and systems, and the role of the firm in customer’s processes and systems [5]. The understanding of value co-creation was given greater specification in the Benefit Based Model (BBM) as proposed by Ng et al [6]. In the BBM, Ng et al argue that the principle of co-created value implies that both customers and firms provide a value proposition and the resultant co-creation during the encounter provides benefits to both (benefit to the customer and revenue to the firm). By linking benefits to co-created value, the model provides an end-to-end visualization of service contract and delivery. In this paper, we use the BBM model as a framework to qualitatively analyze two types of defence contracts for availability in the attempt to better understand the provision of service in a maintenance, repair and overhaul (MRO) environment that is contracted on the performance of the equipment, rather than merely providing equipment. In analyzing these contracts, we enquired about the differences between the traditional contracting and PBC and how it impacts on the

effectiveness and efficiency of service delivery as the nature of the contract clearly changes. Our study provides a customer focused approach that exposes gaps in the way organizations approach their service provision in MRO. We argue that customer involvement and behavioural issues in the co-creation process has to be factored into the design and delivery of performance based MRO delivery systems. The rest of the paper is organized as follows. After a brief review of related literature in section 2, we present our methodology. In section 4, we present our analysis before a general discussion in section 5. 2. LITERATURE REVIEW Aircraft MRO covers an entire spectrum of line and heavy maintenance including repair, overhaul and modification of complete aircraft. It is a highly complex service involving a network of suppliers. A prime contractor for an Aircraft MRO contracts often on the basis full MRO provision, and such a contract often has a book value of over US$100m, creating thousands of jobs and involving some of the largest organisations in the world such as Honeywell, Boeing, Rolls Royce and BAE Systems. The value of the worldwide commercial jet transport MRO market for 2006 was $38.8 billion [7]. Conventional industry wisdom has it that for each aircraft built, the cost to service it over its lifetime is approximately three times its manufacturing cost. Furthermore, with improved design and engineering technologies that extend equipment life, manufacturers are now reporting that more than 50% of their revenues are earned from MRO service. This focus on service has brought manufacturing and engineering curiosity into what constitutes service and what research has been conducted that could assist in their understanding of it. 2.1 Co-creation of Value and the Benefit-Based Model Defining the nature of service has been a challenge to researchers and they have stressed that while there seemed to be a widespread consensus on the importance of service, precise definitions are difficult, owing to the varied nature of service industries [8,9]. Much of service research have also been contextual [10,11] and the lack of adequate service research at an abstract level has resulted in knowledge of service becoming increasingly sector driven with practitioners and researchers socialised within their own industries perpetuating more contextual and jargonised language that is less inclusive, resulting in more embedded and tacit knowledge. While academic service journals aim to be more inclusive for transfer of knowledge across industries, much of their focus is on service management, often with a focus on more intangible service provisions such as healthcare and hospitality. Currently, there is still a lack of understanding on the role of tangible products, of which “Design & Engineering” play a crucial role within a service delivery system, such as an MRO service. The manufacturing and engineering response to a better understanding of products and service within a system of delivering value to the customer was the launch of the

Product-Service-System (PSS) initiative [12], which is tasked to enable innovative ways of transforming the "product-service mix" [13] to achieve sustainable consumption and production. In 2004, Vargo and Lusch proposed the service-dominant logic (SDL) claiming that goods are appliances used in service provision. They suggest that economic exchange is fundamentally about service provision; in short, everything is a service. As such, they argue that customers are always co-producers and co-creators of value when compared to the traditional view where the firm and consumer are separated upon the purchase. Hence, marketing researchers have proposed that firms do not really provide value, but merely value propositions [4] and it is the customer that determines the value and co-creates it with the firm. Co-creating value then implies that customers’ roles are moving from being isolated to being connected to the firm, passive to active and being unaware to being informed. Therefore, a firm’s product offering is merely value unrealized until the customer realizes it through co-creation and gains the benefits. In this light, Ng et al (2008) proposed the Benefit-Based Model (BBM), with a symmetric model of parties in the cocreation process. Similarly, Woodruff and Flint (2006) suggest that in the bi-directionality for mutual satisfaction, as part of the co-creation of value, customers have an obligation to assess the needs of the provider and their own resources. In doing so, there is a need to understand the role of the customer in the firm’s processes and systems, and the role of the firm in customer’s processes and systems [5]. In the BBM representation (Figure 1), Vt is the convex combination of value proposed by customer and the firm. The point between A & B is dependent on the quality of the encounter between firm and customer. The BBM argues that although firms and customers have the power to co-create better value, they also have the power to influence value leading to reduced benefits. As both parties co-create value, roles may overlap implying that not all co-creation result in the highest benefits. In some cases, the overlapping may result in benefits that are lower than what was contracted on. Expected Benefits, B

B( t ,Vt (v f , vc ))

B( t ' ,Vt (v f , vc )) P

A Value Proposition of the Firm & Suppliers/Contractors (F/S/C) – delivered through firm’s operand and operant resources and  processes v f

Vt

B Value Proposition of the Customer – delivered through customer  operant and operand resources vc

Figure 1: The Benefit-based Model for Value Co-Creation

208

The contention then is that bi-directional thinking, together with current thinking in relationship marketing, has to consider that the co-creation towards mutual satisfaction does not always lead to optimal benefits to the customer and the firm [6]. 2.2 Service Contracts: Traditional vs. PerformanceBased Contracts (PBC) Notwithstanding the interest in service, a new way of contracting in MRO has brought the issue of service to the forefront. Traditional MRO contracts are contracted under a MRO service level agreement where the cost of spares could be excluded, or where spares are included in the price [14]. The contractor could also provide a costplus contract provide detailed costs structures (inclusive of a schedule of cost of spares) to the customer to determine reimbursement with a profit percentage that has been pre-determined [2]. Recently, there have been a growing number of MRO contracts that focuses on outcomes rather than inputs or tasks known as PBC. This mode of contracting is starting to re-shape how MRO service contracts are being formed. In essence, PBC is about contracting on performance, rather than tasks or inputs by the service provider. For example, in the case of Rolls Royce, the service provided to maintain engines is being remunerated on the basis of how many hours the engine is in the air. As an analogy, imagine being paid to deliver English lessons to a student not in terms of the number of lessons or materials but on the basis of how many English words is used by the student after the lessons are over. PBC focus on achieving required outcomes rather than a contract for the supply of a set of prescribed specifications [3,15]. In short, the buyer purchases the result of the product used (utilisation of service or performance outcomes) and not ownership of the product. Interestingly, the customer no longer directly manages or possibly even owns resources such as the inventory of spares. Hence, researchers argue that in the long term, suppliers may find it in their interest to invest in designing more reliable products and more efficient repair and logistics capabilities to increase profitability [16]. This implies that contracting on PBC has an ability to elicit desired behaviours arising from the incentives within the contract, thus reducing the cost of MRO over the longer term for the customer. Nonetheless, these different types of contracting methods have different risk implications for both buyer and supplier. For example, a fixed-price contract puts all the risk on the supplier but few performance incentives. A cost-plus contract shares the risks between customer and supplier but provides few or no incentives for the supplier to reduce cost [2]. Under PBC, there are important differences in terms of risks and responsibilities between supplier and customer. For example, suppliers tend to have full responsibilities for performance, such as the transfer of the risk for investments, ownership, maintenance, utilized capability and re-sales [17].

209

Overall, there are more equitably aligned risks and incentives between suppliers and customers in PBC contracting than in traditional contracting [2]. As such, we are beginning to find more B2B services contracts moving towards performance-based incentives with hopes of witnessing significant improvement in costs and customer satisfaction [1]. Yet, PBC is not a new form of contracting. Literature shows in the 1960s, US government bodies have begun initiating contracts to optimise public spending. In defence contracting, questions such as addressing “incentives to produce good performance” and “incentives apart from profits to induce innovation” were subjects of discussions [18] to ensure that the roles of the parties concerned in the governance of the service contracts for the public are jointly engaged. PBC are also widely used in other public services such as health services [19] and transport services [20]. In health services, PBC has been promoted by the US Institute of Medicine as a cost-effective mechanism to manage and ensure the “effectiveness of public medical services’’ through funding of certain treatment outcomes. They use PBC with local health centres to monitor and evaluate their performance in order to ‘‘redirect funds, away from less efficient programs within the communities towards programs which have proven themselves. Similarly in transportation services, Hensher and Stanley (2008) argue that PBC are excellent mechanisms aimed at promoting economic effectiveness and efficiency through the life of the contract. Recently, the Office of the US Secretary of Defence has initiated and the US Air Force has aggressively implemented a policy aimed at the widespread adoption of performance-based services acquisition (PBSA), an outcome-oriented approach in which the buyer tells the supplier what it needs rather than how to meet that need. From its successful implementation, efforts were made to define positive performance-based practices. It resulted in a study that showed that the Air Force personnel were generally pleased with the results of PBSA as well as with many of the practices it encourages [21]. From the exposition above, it is clear that firms could contract its services on a spectrum of levels between the traditional and performance-based. For each extreme of the spectrum, it would then be up to the customer’s responsibility to create the rest of the value to achieve the benefits. Hence, if a firm is contracted only for a resource based contract, the customer would either manage the rest of the value within their own value proposition or contract with multiple firms, leading to the make-buy decision facing many organisations. For MRO services, there is evidence to suggest that increasing number of contracts are moving towards performance-based type of incentives to ensure effectiveness and efficiency of both the firms’ and the customers’ resources [15,22]. Despite this growing interest in PBC from both the public and private sectors in terms of application, little research has been established in understanding the dynamic relationship between the firm and the customer under a PBC where value is co-

produced and created. To continue with the English lesson analogy, where previously an English teacher skills set include the expertise of the English language and the skill to teach the language, under the new PBCdriven business model where the student’s ability to speak the language is the performance outcome implies that the English teacher needs new skill sets of motivation, pedagogy and even psychology to ensure that the student is able and willing to co-create value with the teacher. Hence, there is a question of risk that is borne by the contracting parties in value co-creation under a PBC if they do not have the competency to ensure that the customer is able to co-create value to achieve the outcomes. With evidence to show that the utilisation of PBC in MRO service contracts are increasing, this change of the business model from the traditional contracting poses some serious questions. First, are the processes, systems, behaviours and activities designed under the traditional business model just as efficient in the new business model? Inefficiencies could arise from a combination of two local optimums rather than from optimising globally across two systems. This then results in an increase in overall system costs which would make the contract more expensive than it has to be. Second, are the processes, systems, behaviours and activities designed in the traditional business model just as effective in the new business model? Ineffectiveness could arise from the combination as well as from both parties’ inability to explicitly build a combined system. And as both parties focus on their individual system efficiencies, the transaction cost increases from the interactions. In other words, as both parties build more efficient individual systems, the overall effectiveness of the contract may suffer leading to suboptimal outcomes. Although there has been research on PBC within the construction industry [3], there are not many studies that examine fundamental delivery issues arising from the service concept particularly from the perspective of identifying potential risks and delivering benefits to the customer. Also, literature opens up the debate on balancing formal contract and relational governance, the proportion of goods and services to offer a proper value for the customer. Hence, using the aircraft maintenance industry as a context for MRO services, this paper attempts to answer the questions on identifying the potential risks that arise from a PBC under a co-located MRO service environment given that both the firm and the customer are co-creators of value in the relationship. 3. METHODOLOGY This study represents a qualitative study. There are a number of different methods to be used in qualitative research and it can be distinguished between four major methods: observation, analysis of texts and documents, interviews, and recording and transcribing. The logic behind using multiple methods is to secure an in-depth understanding of the phenomenon in question.

In our study, we analysed two defence contracts between a defence contractor and the UK government (in this case the Ministry of Defence or MoD) which were based on a type of performance based contract that delivers the aircraft as a performance outcome of the contract availability. We conducted in-depth interviews with stakeholders from the firm and the customer and these included technical managers, executives, commercial managers, directors, army officers and commanding officers. In order to capture an in-depth understanding of the relevant stakeholders’ perception of the two defence contracts, the questions asked were mainly open-ended and aimed at establishing the interviewee’s perception of the benefits derived from the contracts. The interviews were recorded and subsequently transcribed, coded and categorised. Participant observation on the MRO sites was also employed to document the interactions between customer and firm. The two contracts analysed were awarded to 2 different organisations. Both were awarded for the MRO of the equipment’s ‘through life’. The performance-based nature of the contracts is expected to bring a total of USD1.2b savings to the customer (MoD) over their combined serviceable life. Unlike conventional outsourcing solutions, the contracts were unique in the sense that the companies had to use people and assets that ‘belonged’ to the MoD in delivering the service, and also be colocated physically at the customer’s site. While the MRO service is outsourced, the MoD’s had a big role in the partnership which is to provide Government Furnished Materials (GFX) including supplying physical facilities, material, IT and manpower to facilitate the company in achieving its outcomes. The cost of GFX is generally not included in the contract price. The first contract was the MRO service for a fleet of aircraft used by the Royal Air Force, including spares provision, technical support and maintenance training. The contract is broadly based on a fixed annual price with the performance of the MRO service assessed principally through the outcome provision of the availability of a bank of flying hours of the aircraft. In addition, there was a noncontractual KPI that measured the performance of GFX which measured the MoD’s performance in delivering the necessary assets and manpower for the programme. Previously, the MoD’s Integrated Project Team (IPT) was responsible for the overall MRO service of the aircrafts. With the programme in place, the scope of work for the IPT reduced significantly and together with a downsizing of manpower, the role of the IPT shifted from being the ‘provider’ to an intelligent ‘decider’ that enabled what support was required from the company to achieve the performance. Consequently, the company’s responsibility was to ensure the required aircraft (at an agreed capability) was provided to the RAF front line when they were needed. The business risk was thus transferred from the MoD to the industry. The second contract was a broadly agreed annual fixed price MRO service with its performance assessed through the availability of a weapon system for the British Army.

210

The solution for the weapon’s readiness and availability included a company and customer IPT support centre colocated at an army base with on-site maintainers training aids, fleet management, and a joint delivery team. The programme also employed civilians who would support the equipment both in barracks as well as in operation (wartime) availability of equipment. The measurement of performance output differed between the availability in barracks and the availability in the operating theatre (e.g. in Afghanistan) with a higher availability in barracks than in the operating theatre. 4. ANALYSIS AND FINDINGS 4.1 Analysis The data obtained was subjected to grounded-theory analysis to identify data that was salient, recurring and themes that could emerge from the interview data representing the categories that had some meaning to the respondents. Researchers re-examined the transcripts to evaluate the plausibility of the categories identified for their informational adequacy, credibility, usefulness and centrality. The interaction between the categories was discussed extensively and the initial coding categories were refined with another round of interview data being coded based on the categories found. Data from the same categories were then grouped to assist the final evaluation of the categories. We then apply the benefit based model into the findings for a more complete valuebased understanding of MRO service. 4.2 Findings In analysing the two MRO contracts through a valuebased approach, we apply the BBM framework presented in Figure 1. Our analysis show the difference between the traditional business model and the new business model based on outcomes. In a traditional MRO or logistics environment, contracts are not usually based on outcomes but rather fixed inputs. Revenues were based on Vf in the traditional contract (i.e. only the value proposition of the firm), which implies that the firm has no incentive to be pre-emptive in maintenance, to invest in reliability for spares or to be innovative in solutions. Under this new business model, by contracting for availability at a fixed sum, the contract is now on the basis of outcome (Vt in Figure 1) where the value is coproduced and co-created by the both the firm and the customer. By applying the BBM perspective, the firm alone cannot deliver on the outcomes without the cooperation of the customer. Our findings found six challenges under the new business model. They differ in terms of degree and intensity across the two contracts, but would exist in some form in both contracts. Six Challenges in MRO Service Delivery under PBC 1. Complexity and Unpredictability in Costs - There is real difficulty in calculating costs when the team is being reactive to changes, where predictions are difficult, and when the service provision aims to be innovative and preemptive. Being innovative and pre-emptive would reduce

211

the spares used and in turn reduce the overall costs of the contract as well as achieve higher level of satisfaction. Yet, this implies that there is less predictability in the system and the lack of predictability would make cost estimation and forecasting difficult. Our findings found this tension to be challenging to the employees delivering the service. On one hand, there is a need for predictability to report to headquarters and to forecast costs so that the service can be delivered economically and below the price contracted and on the other, there is also the need to manage and change usage and provide more innovative solutions so that overall acquisition of spares would be reduced. This constant negotiation is clearly not sustainable over the longer term and caused tension within the firm when compared to the traditional business model. 2. Cultural change from traditional contracting - The new business model is adopted differently by different people. Our findings showed that many of the firm’s employees negotiated within themselves what their value is within the company as the company moved towards delivering value under a PBC. Identity issues abound as company personnel try to grapple with their own place within the organisation. In addition, the concept of delivering value to the customer has changed from being a design and manufacture organisation to that of a service organisation and our findings showed that people struggled to reconcile the changes. 3. Loss of perceived control by customer - The changes caused by the outsourcing the contract came amid other changes within the customer. Findings show that the customer faced a loss of perceived control. Where they were previously in charge, the role change caused discomfort and disruptions 4. Loss of perceived control by the firm - The complexity and lack of predictability manifested itself through the organisation from strategic to the operational and tactical. This resulted in an increase in a lack of control and security, manifesting itself in higher monitoring and transactions. 5. Lack of Boundaries (Rigidities and Fluidities) - With both the firm and the customer co-producing the service to ensure availability and benefits for the end-user, the outcome driven nature of the service and the coproduction resulted in a lack of boundaries as to what is ‘acceptable’ under the contract. Our findings suggested that there have been instances where boundaries were held rigidly (“this is their problem”) and where boundaries have been fluid with out of contract requests being accommodated so as to build better relationships. This was clearly viewed differently by different people within the organisation. For those who were more understanding and accommodating of the customer, others within the organisation viewed them as having ‘gone native’. 6. Coordination with suppliers - A big challenge to availability based contracting was how to reconcile and align contracts with sub-contractors. Where previously an

order from a customer could be sub-contracted out and the orders join up in terms of costs, resources and delivery, it now wasn’t clear what the role of subcontractors are and how they fit into the value coproduced by the firm with the customer. In a PBC contractual environment, 85% availability of the equipment did not translate easily to 85% of its component spares. Four key findings on Value Co-Creation Based on the challenges that surfaced, we re-analysed the data using the Benefit-Based Model (BBM) and categorised our findings on four aspects of BBM: understanding value-in-use, service behaviours and service skills, capacity in service value proposition and value co-creation & co-production. For each of these findings, we find potential risks at the design stage of the contracting. Key Finding 1 - The need to understand value-in-use (i.e. multi-state benefits) in availability based contracting is crucial because of the way value-in-use impacts on customer satisfaction, costs and delivery of the service. MRO service requires a good set of historical data on how often the equipment and its spares broke down. Historical use ascertained through data obtained before the signing of contracts was used to inform the way the contracts were priced and cost estimated. However, our study shows that the past may not be a good reflection of the future. Understanding usage, and more specifically, “changing usage” could bring about a more efficient and effective support solutions that would result in benefits to customer and firm. Our data showed numerous examples of these types of “usage-change” that impact on customer satisfaction, costs and the delivery of the service. For example, a pilot that is more careful about the use of the equipment such as “taking care when removing the communication plug” instead of carelessly and unknowingly flinging it, and hitting the windscreen can save the firm £18,000 per piece of glass. Similarly, there is evidence in the data to suggest that when “rudders” are broken, they are simply “thrown into the sand and lost in the desert forever” rather than brought back to base. As one personnel observed; “the rudder could probably be repaired for £1,000 rather than buying a new one for £22,000”. In both cases, the costs savings translate to benefits for both parties and therefore We also found that an understanding of usage has an impact on how the service is being delivered. Due to the state-contingent nature of value-in-use, the usage of equipment would vary, as would the service delivered to ensure the most effective usage would vary as well. For example, it was noted that pilots were using “cables as a foot rest”. Rather than moving the cable away, it was reported that the firm “put a guard over it and sort it that way”, hence ensuring that understanding the usage of the customer resulted in better service delivery leading to higher customer satisfaction. Similarly, the example of solving the problem of “water getting into the windscreen causing it to delaminate” by putting a “rubber compound

outside the windscreen to prevent water getting in” saved the customer (and the firm) “thousands of pounds”. In the past, the firm would have taken care of the repairs which would cost thousands whereas the rubber compound only cost around £10-£20. In essence, if the firm’s activities, design and systems do not join up towards value-in-use and both the organisation’s and the customer’s role in co-creating value towards the benefits is not made explicitly clear, the organisation may not realise the conflicts that may occur due to different people delivering to different perceived value. This is particularly acute when having to negotiate the tension between the predictability of costs and the need to reduce costs of service delivery. Predictability is based on historical data whilst costs reduction is based on understanding, managing and changing value-in-use, and it’s clear that the skills, resources and processes towards the two are quite different. This poses a risk to effective service delivery and cost of delivery. Key Finding 2 - The study finds that the firm do not sufficiently emphasise the role of people in delivering value. There is a high dependency on processes and activities that are equipment focused, without much attention on the behaviours required to achieve them. For example, an employee of the firm noted that “if you’ve had a supplier that’s going to deliver something to your house you wouldn’t disown the responsibility of managing that supplier to make sure that he delivers what you need”. However, if we “sit back and do nothing and don’t get involved with the customer (GFX, who is also the supplier), they will just carry on the way and not deliver any of it because at the moment, they probably haven’t got their head around that they are also a key supplier”. It appears then, that behaviours of the firm and customer are important for co-creating value in the relationship. The BBM model also suggest that there are skills involved in coordinating people, leading different situations, developing relationships, thinking as a team, reducing misinformation, establishing trust and projecting a good image of the company. However, much of these skills seem to be attributed to individuals and personalities and while individuals will always be important, they would need systematic support within the design of the system so that the service delivered could be better replicated. Nonetheless, we found that there seem to be inadequate provision to capture the learning into the organisation. As noted by an employee of the firm; “it’s no good waiting for things to go wrong before we do something about it, we’ve got to be knowing, we’ve got to know what’s going on all the time so that we can be ahead of the game in making sure that the user’s always got, available to do the job that he needs to do and we have readiness and availability criteria set out in the contract which are very clear about what we need to achieve”. This observation implies that efforts to instil suitable attitudes and behaviours depended very much on the individual and team initiatives, rather than being explicitly and systematically designed into the system and structured in concert with other activities. It does not mean, however,

212

that human factors and behaviours are ignored within the organisations. Indeed, for both the firm and the customer, such factors are recognised and individual managers go through considerable effort to ensure that people do have the right attitudes and behaviours. Hence, service delivery requires high fixed costs that could have a major human component. Capacity and capability of human resources are paramount in service delivery and a lack of focus on human resources may result in lower or inconsistent service quality and an increased cost due to higher transaction, monitoring, scrutiny and mistakes. By not designing and systematically structuring behaviours into the system, the service delivery system runs the risk of not capturing potential conflicts between processes/activities and behaviours in delivering the service. Key Finding 3 - Our study finds that the firm does not have a clear understanding of where and how value is created within the service contracts and the contribution of components and resources to value. There needs to be equitable focus on both equipment capability as well as embedded human capability in understanding the capacity for delivering the service. For instance, the firm has a fixed set of resources, both tangible and intangible to deliver on its service offerings to the customer. If the firm is to design and structure its capacity to deliver the service, the firm would have to understand which component of its costs deliver how much of value to the customer and the degree of importance of all resources within that system. Hence, the service capacity of the contract becomes important to the firm. Our findings show that such a systematic analysis is lacking. Additionally, from our understanding of the existing system and processes within the firm and customer, there is inadequate understanding of the degree of importance of human and equipment factors within the system when delivering the service. For example, from the interview, there was an observation that revealed the firm was unable to “carry out the inspection and repair because there are various loopholes in military documentation or military procedures that don’t allow them to actually carry out the inspection or repair, there’s anomalies within the military system that people don’t understand”. These comments suggests that it is important to understand the resources and components of human and equipment capability as well as the links between resources, costs and service attributes to employ the optimal service capacity in delivering the service under the contract. By not analyzing service capacity, the firm would not be able to determine the supply availability for service offerings and the lowest costs to deliver the same service if it needs to be scaled up, or if the service is to be transferred or repeated in another contract and this poses a potential risk to the organisation. Key Finding 4 - Our study finds that the firm is more focused on its value proposition and less focused on the

213

co-produced value proposition. This is to be expected as traditional contracting allows the firm to concentrate on its value proposition to the customer. However, availability based contracts are contracted on the basis of cocreation between the firm and the customer. Under the BBM, this is illustrated as Vt (Vf, Vc). This then implies, to a large extent, that the business model of the firm has changed. Where traditional business model was fulfilled upon delivery of Vf (value proposition of the firm), contracting for availability demands that the firm fulfils its obligation to deliver to Vt (Vf, Vc). Hence, the customer’s value proposition in the value co-production of the contracts Vc becomes the responsibility of the firm. Interestingly, under the MRO contracts that we examined, the customer’s responsibility to deliver certain aspects of the assets under the contract, allowed the firm to abdicate some of its responsibilities. However, that does not change the fact that the firm has chosen to contract on availability, and with that choice comes the responsibility of understanding their customer’s value proposition to coproduce value. As such, there is a need for both the firm’s and the customer’s value proposition to be understood well by both parties in order to deliver the maximum benefit. However, our data saw very little evidence of this being an important factor in the firm’s systems and processes. In fact, we found on many instances, inconsistencies in the interactions between the people (firm) involved in the co-production and the customer. Where demands of the customers are unreasonable, they are sometimes met with the objectives of building relationships whereas less unreasonable demands are tolerated as a one-off exercise. For example, in the interviews, while one respondent commented that there is a view that “if you can get that amount of passion, one team, one goal delivering the end product, the fall out will be that it will come cheaper and it will come quicker”, another customer commented “what’s it got to do with you and the shutters would just go straight up you know there was just no interaction”. Yet, there was another comment by an employee who noted that “my engineering people have still got a little way to go because they are not that close to the customer” and the “contract is a service which is, I am trying to think of the word (which) it’s a bit of a contradiction with engineering”, also the engineers have the attitude of “what else can I do for you sir and a curtsey and scope creep again”. These comments appear to suggest that the firm is still focused on its own value proposition without the need to incorporate the customer’s value proposition in delivering the service. Under these circumstances, the risks for the firm is in the danger of focusing on Vf instead of Vt is that the optimal system for Vf may not be that of Vt. Without understanding what is required to optimise resources under Vt (i.e. a thorough understanding of both firm and customer proposition) to deliver better service at a lower system cost, the combination of 2 optimums (the customer Vc and the organisation Vf) may result in higher system costs due to transaction costs and misalignment issues such as seen from the data.

5. DISCUSSION Our research indicates that services contract (MRO) that move from traditional based contracting to PBC poses serious issues that impact on both the firm and customer in terms of risks. It appears that the processes, systems, behaviours and activities which were associated with traditional contracting are not as efficient or effective under the performance-based environment. Also, our investigation opens up the debate on balancing the proportion of goods and services to offer a proper value for the customer. Using the aircraft maintenance industry as a context for MRO services to support our research, this paper answers the questions on identifying the potential risks that arise from a PBC under a co-located MRO service environment given that both the firm and the customer are co-creators of value in the relationship. In our study, we find four key-findings underpinning the contracts. First, we found that the firm has an unclear understanding of “value-in-use”. The need to understand value-in-use (i.e. multi-state benefits) in availability based contracting is crucial because of the way value-in-use impacts on customer satisfaction, costs and delivery of the service. When the customer uses the service, how they use the service and understanding the manner in which they use the service is vital to bringing efficiency and effective support solutions that result in benefits to both the firm and the customer. Second, the study found that the firm do not sufficiently emphasise the role of people and their behaviours in delivering value. There is a high dependency on processes and activities that are equipment focused, without much attention on the behaviours required to achieve them. For example, within the value proposition of the firm, lies an array of attributes performed by the firm that consume the firm’s resources in order to deliver services for the customer. These attributes include transporting the spares, storing and managing equipment and diagnostics. In essence, this implies that the successful delivery on these attributes (which effects the achievement of measured availability of the contract) in co-production with the customer requires resources from the firm, including human resources. Furthermore, the delivery of many attributes requires suitable behaviours within the organisation as well as within the customer organisation. However, our findings suggest that the firm does not explicitly and systematically focus on human factors within the attributes e.g. cultivating the right behaviours within the organisation towards co-production. Third, the study finds that the firm does not have a clear understanding of where and how value is created within the service contracts and the contribution of components and resources to value. As mentioned earlier in the second key finding, within the value proposition of the firm lies an array of attributes performed by the firm using both the firm’s tangible and intangible resources in delivering services for its customer. If the firm is to design and structure its capacity to deliver the service, it would have

to understand which component of its costs (resources) deliver how much of value to the customer and the degree of importance of all resources within that system. The service capacity of the contract then becomes important. Our analysis show that such a systematic analysis is lacking. Additionally, we found that there is inadequate understanding of the degree of importance of human and equipment factors within the system when delivering the service. In other words, there needs to be equitable focus on both equipment capability as well as embedded human capability in understanding the capacity for delivering the service Finally, based on our BBM analysis, the firm has to contract on Vt which is combination of the firms’ (Vf) and the customers’ value proposition (Vc). This directly brings the customer’s value proposition into the firm’s delivery of the service. Hence, an understanding of co-production and co-creation of value is required on the part the firm. However, we found in our data that the firm is less focused on the value co-production and co-creation. In fact, this change from the traditional way of doing business (i.e. charging for MRO activities) to availabilitytype contract has caused discomfort in terms of understanding the activities involved within the scope of the contract. Boundaries of what should be delivered under the contract are no longer obvious. While clear performance indicators relate to availability, many in the firm are unaware that the performance is unachievable without the cooperation of the customer (Vc). As such, our study of these MRO performance based contracts demonstrates that there exist potential risks associated with moving from a traditional based contracting platform to performance-based contracting. 6. CONCLUSION On the basis of our research, it is clear that with performance-based contracts, firms may find themselves exposed to customer-focused risks that threaten their capability towards delivering service value that is replicable, consistent and scalable across future service projects. If the performance depends on the co-creation between the customer and the firm, both parties would need to understand their own value proposition fully before contracting to avoid the risk of lowered benefits with the contract performing at a different value Vt. Furthermore, if the contract is renegotiated over time, consistent reduction in benefits for whatever reason may result in a re-negotiated price that is lower than the optimal benefits, even if the firm is highly efficient in its own value proposition. This implies that both parties need to come together to achieve an effective value cocreation/co-production model and in turn the appropriate contractual mechanisms to achieve a consistently high benefits that are financially viable. In addition, inefficiencies can arise from a combination of two local optimums rather than optimising globally across the two systems. This then results in an increase in overall system costs which would make the contract more expensive than it has to be. Additionally, ineffectiveness

214

could arise from the combination as well as from both parties’ inability to explicitly build a combined system. As both parties focus on their individual system’s efficiencies, the transaction cost increases from the interactions. In other words, as both parties build more efficient individual systems, the overall effectiveness of the contract may suffer (due to more altercations and transactions), leading to sub-optimal outcomes.

10. Knight, Gary (1999), "International services marketing: review of research, 1980-1998," Journal of Services Marketing, 13 (4/5), 347.

As such, our qualitative study provides an insight into the co-creation process between the firm and the customer under a performance-based contract. We identify areas of potential risks from four key findings. For further research into this area, we intend look at some aspects of quantitative analysis to validate some of these findings.

12. Mont, O. K. (2002), "Clarifying the concept of product-service system," Journal of Cleaner Production, 10 (3), 237-45.

REFERENCES 1.

2.

3.

Gordon, S. (2001), Performance-Based Contracting. Washington D.C.: International City/County Management Association. Kim, S.H., Cohen M. & Netessine S. (2007), "Performance Contracting in After-Sales Service Supply Chains," Management Science, 53 (12), 1843-58. Gruneberg, S., Hughes, W. & Ancell D. (2007), "Risk under PBC in the UK construction sector," Construction Mgmt & Econ, 25 (7), 691-99.

4.

Vargo, S. and Lusch R. (2004), "Evolving to a New Dominant Logic for Marketing," Journal of Marketing, 68 (1), 1-17.

5.

Woodruff, R.B. and Daniel J. Flint (2006), "Marketing’s service-dominant logic and customer value," in The Service-Dominant Logic of Marketing: Dialog, Debate and Directions, Lusch R.F & Vargo S.L., Ed. New York: M.E. Sharpe.

6.

Ng, I., Lei G., Scott J., and Yip, N. (2008), "Towards a benefit-based framework for understanding B2B services and its impact on contract and capability" in 10th International Research Seminar in Services Management: Insights into Service Activities La Londe, France.

11. Pilkington, Alan and Chai Kah-Hin (2008), "Research themes, concepts and relationships," International Journal of Service Industry Management, 19 (1), 83-110.

13. Manzini, E. and C. Vezzoli (2003), "A strategic design approach to develop sustainable product service systems: examples taken from the environmentally friendly innovation Italian prize," Journal of Cleaner Production, 11 (8), 851-57. 14. Van Weele, A. (2002), Purchasing and Supply Chain Management, Analysis, Planning and Practices (4th ed.). London: Thomson Learning. 15. Bramwell, J. (2003), "What is performance based building?," in Performance-based Building, First International State of the Art Report, P.S. Lee A. and Barrett, Ed. Vol. CIB Report 291. The Netherlands: CIB Publication 16. Martin, L. (2003), "Making PBC perform: What the federal government can learn from the state and local governments " in The Procurement Revolution, Abramson M.A. and R. Harris, Eds.: Rowman & Littlefield. 17. Glas, A. & M. Essig (2008), "Public Performance Contracting of solutions: Benefits and consequences," in 17th Annual IPSERA Conference, Australia: International Purchasing & Supply Education & Research Association. 18. Sumner, M. (1964), "Studies of Defense Contracting," Harvard BR, 42 (3), 20-184. 19. Shen, Yujing (2003), "Selection Incentives in a Performance-Based Contracting System," Health Services Research, 38 (2), 535-52.

7.

Jackman, F. (2006), “MRO Market Up Modestly As Efficiencies Take Hold”, Overhaul and Maintenance, April 2006 Issue.

20. Hensher, D. and Stanley J. (2008), "Transacting under a performance-based contract: The role of negotiation and competitive tendering," Transportation Research Part A 42, 1143-1151.

8.

Lovelock, C. & Gummesson E. (2004), "Whither Services Marketing? In Search of a New Paradigm and Fresh Perspectives," Journal of Service Research: 7 (1), 20.

21. Ausink, J., Camm F. and Cannon, C. (2001), “PBC in the Air Force: A report on experiences in the field”, Rand Corporation.

9.

Ng, Irene C.L. (2008), The pricing and revenue management of services: a strategic approach London: Routledge.

215

22. Foliente,

G.C (2000), "Developments in performance-based building codes and standards," Forest Products Journal, 50 (7/8), 12-21.

Cost Modelling Techniques for Availability Type Service Support Contracts: a Literature Review and Empirical Study P. P. Datta, R. Roy Decision Engineering Centre, Cranfield University, Cranfield, Bedford, MK43 0AL, UK [email protected]; [email protected]

Abstract The research in this paper is focused on enhancing existing knowledge in cost estimation models at the bidding stage of service support contracts. The difficulty of this task lies in the long lasting contracts, which in some cases may reach even 40 years. The paper first reports the existing knowledge through detailed review of literature. The paper studies different service support contracts and reports the cost modelling techniques used in availability type contracts in the context of defence and aerospace industry. The contribution of this paper is to identify the key areas of improvement and business priorities in the area of cost modelling of availability type service support contracts. Keywords: Cost Modelling, Availability Type Service Contracts, Literature Review

1 INTRODUCTION Upon entering into 21st century, the business environment for the manufacturing industry has changed significantly. Manufacturers now tend to include more services in their total offering to: facilitate the sale of their goods; lengthen customer relationships; create growth opportunities in matured markets; balance the effects of economic cycles with different cash-flows; and respond to integrated service solution demand [1-3]. A major shift in support and maintenance logistics for complex systems over the past few years has been observed in defence and aerospace industry. Availability contracting, a novel approach in this area, is replacing traditional service procurement practices. The premise behind availabilitycontracting is summarized in the official UK Ministry of Defence (MoD) guidelines (http://www.berr.gov.uk/files/file33168.pdf, 2007): “Contracting for Availability (CfA) is a commercial process which seeks to sustain a system or capability at an agreed level of readiness, over a period of time, by building a partnering arrangement between the MoD and Industry.” Cost assessment of such service offerings remains a challenge and has not been addressed in literature. Some of whole life cycle cost literature focuses on assessing the maintenance or in-service costs [4-7]. But most of this literature is focused towards costing the service associated with stand-alone products. In most of these studies, services are viewed as “add-ons” to products and treated as mere features of the products. The literature on assessing the costs of availability type support service solution is rare. This paper aims to bridge the gap. This research first studies the literature on different cost estimation techniques, whole life cycle costing, service and maintenance costing. Then it carries out an empirical study of cost assessment techniques used in availability contracts in service-based manufacturing organisations in the defence industry undergoing transition from being a manufacturer to a service provider. This research summarises the key best practices and areas for further improvement. The paper presents an integrated framework for costing service contracts using different modelling techniques and information for aiding the cost assessment of availability type service contracts.

CIRP IPS2 Conference 2009

216

2 LITERATURE REVIEW A structured approach is adopted for the literature review. Studies were identified through an electronic search of the databases such as, Ingenta, Emerald, Science Direct, Engineering Village, Web of Science, IEEE Explore, library files and reference lists. In addition, the literature search was extended to the Internet, to NATO, NASA websites, as well as government (MoD, Department of Defense), academic institutions (MIT, Stanford, Georgia Tech) and large cost engineering companies’ web pages for unpublished online information. Relevant literature is classified into several specific groups to set up inclusion, exclusion criteria. They are: whole life cycle costs, service and maintenance costs, cost estimation techniques. The literature review is reported below. 2.1 Cost Estimation Techniques There are three well-recognized estimating models of cost [5]:  the analogous method: compares costs according to similarities and differences with other projects (for parts geometrical characteristics influence), which has been applied a lot in the aerospace industry;  the bottom-up method: collects all product cost values that are available, making it a highly data intensive method, activity based costing poses an example; and  the parametric method: derives cost estimating relationships (CER’s) and associated mathematical algorithms to reach cost estimates [8]. Models in this group include regression analysis, fuzzy logic, and neural networks [9]. The most often applied parametric estimation method is regression analysis [10]. These are known to be top-down applications. At an advanced level, a range of methods such as feature based modelling, have also found application. This method uses relational drivers of cost, which means a direct relationship, is developed between the associated feature of a product and its cost [8]. Fuzzy logic is also a relatively new method, which applies highly sophisticated mathematical models to estimate costs, in situations where imprecision and uncertainty is very affective. This method has commonly been applied to represent vague

and imprecise knowledge [8]. Harding et al [11] proposed that, within the construction industry, neural networks offered a useful route for cost estimation at the early stages. As the system receives new information it incorporates it into the decision making process [12]. 2.2 Whole Life Cycle Cost Whole life cycle cost (WLCC) of products comprise all costs attributable to a product from conception to those customers incur throughout the life of the product, including the costs of installation, operation, support, maintenance and disposal [13], [14]. There have been a number of proposed approaches to dealing with the WLCC over the years, presented in a review by Christensen et al [7] and majority of them are reported to be suffering from lack of available data. The use of historical maintenance data has been examined [15, 16] and it was found that historical information is most often combined within broader definitions that do not provide the transparency required for the development of accurate WLCC models. Parametric CER is a recognised technique for prediction of costs based on the known behaviours of past projects with similar key cost drivers [17, 18]. However, lack of appropriate/available data is a huge obstacle in developing these relationships in the first instance. In order to avoid this, researchers [6] have used iterative approaches in which feedback is embedded into the practice. However, the main limitation of this linear model is, all the costing decisions are based on design level cost drivers alone. But within WLCC framework, the key drivers are not transparent within the developmental phase but extend through the operation/maintenance phases. Early et al [19] treated WLCC as an evolutionary process rather than a linear one and proposed a reverse spiral WLCC model. This model described a process by which a dynamic feedback model could be implemented in order to inform the WLCC, in which an integrated developer/operator modelling approach is adopted. The time-intensive nature of the process and the need for data transparency between developer-operator are the obvious disadvantages of such approach. In construction and other capital intensive industries there is a further, more complex dimension that refers to longterm life cycle costs. Once cost modelling spans decades, instead of months or years, the issue is not only that of the availability and reliability of data, but also the compatibility of the timescale with the ever-shorter time horizons utilized by private and public corporations in their decision making processes, as well as the need to compare different economic and financial future scenarios [20]. Nicolini et al [20] showed through case study research in UK construction industry, the industry often operates without full understanding of costs throughout the supply chain, which changes with time. Nicolini et al. [20] found, most players in construction currently either develop prices through “top down” estimating based on rates for building elements or else can only price something once quotes have been received from the various subcontractors or materials suppliers involved in delivering a fixed design. The researchers thus found a gap in current WLCC research on to what extent or in which contexts the elemental costing can be replaced with “bottom up” estimation of actual labour, plant and materials costs throughout the supply chain. WLCC is one of the most effective cost approaches in railway industry for prediction of total railway system costs. Jun and Kim [21] have showed the application of this technique in cost modelling of the brake module of a train vehicle. Several scenarios were created with reduced or increased man hour requirements. Several

issues as uncertainty of input data, reality of the simulation models are also pointed out by the authors as issues with WLCC approach. Nilsson and Bertling [22] presented a WLCC analysis of wind power systems with possible maintenance management benefits in strategies using condition monitoring services. A present value method is applied to understand the present value impact of annual WLCC across the years of maintenance contract to understand the best maintenance strategy. Several scenarios are constructed to evaluate different maintenance strategies and their WLCC impacts. The importance of WLCC for defence planning and military equipment has long been recognised [23, 24]. For most of the fighting equipment development work, there always exists a dominant proportion of operation and support cost compared with acquisition cost [25]. In defence context, it is clear that the detailed decisions on how the system is to be manned, supported and how it is to be repaired and overhauled (and the frequency of these tasks) will have a profound influence on the operating and support costs [26]. From the above literature review on WLCC, several issues and research gap have evolved. First, all studies in WLCC are focused on product and are mainly carried out at design stage with very little knowledge of maintenance or operation associated cost drivers and processes. Secondly, parametric methods are of use only when enough data is available on the product and its lifecycle activities. Accurate WLCC remains a time-consuming and expensive process. Thirdly, in long term WLCC studies, the ability to estimate when and to what extent the topdown costing can be replaced with “bottom up” estimation of actual labour, plant and materials costs throughout the supply chain remains a challenge. End-to-end cost estimation (starting from supplier to the customer) for long term WLCC remains a critical issue. Fourth, simulation models can be used in WLCC research, especially while considering maintenance/operations but validation of those models remains an issue. Finally, better data collection processes will enhance accuracy levels of WLCC. The role of existing knowledge (historical data or expert judgments) is evermore critical in this process. So, capturing and making proper use of this intangible asset remains an area for further research. Essentially, control of information and knowledge is an important part of this problem. 2.3 Maintenance Costing The review of literature on costing of maintenance activities is carried out to understand the methods used to cost the maintenance activities. Jian and Hong-fu [27] tried to predict the maintenance cost for civil aeroplanes. They first identified the key factors contributing to the maintenance costs by detailed investigation of some Chinese airlines. Common characteristics of each airline are integrated and a general cost breakdown structure (CBS) of aeroplane maintenance is proposed. The final cost is added from the bottom up from this CBS and CERs are constructed to predict the maintenance costs. Bowman and Schmee [28] suggested a discrete event simulation model utilizing historical cost and failure data analysis results to evaluate contract price. The simulation discussed in the paper utilized the statistical failure and cost models to generate the simulated failure and cost models to generate the simulated failures and repair/replacement costs. Lamb [29] attempted to connect business performance and competitiveness to maintenance in the context of paper and pulp industry. He discussed the concept of availability as a measure of business performance and made reliability and

217

maintenance operations as the key drivers for paper mill performance. Lamb approached maintenance costs from a totally different perspective, where the paper mill’s maintenance budget is no longer an extrapolation of the past and rather it is a forecast of costs necessary to achieve the mill’s planned business performance. Hence it can be found from the above limited number of studies that a combination of different cost estimation techniques (analogy-based methods, parametric techniques) in addition to simulation models are employed to cost maintenance offerings. Still the techniques used are data intensive and very much rely on historical data availability on maintenance. 2.4 Service Costing Very few studies are reported in service costing. We are reporting here some of the different techniques used. Top down costing Top down costing first calculates the total costs of the service at the organisational, provider or departmental level, then disaggregates the total costs to the department or the units of services (or products) depending on the richness of available data and the homogeneity of services provided. It can be done through multiple steps, e.g. allocate costs to cost centres (e.g. support services workshop, project management), then divide the total costs of the cost centre by the number of units (e.g. spares supplied etc) [30, 31]. Top-down approach is less detailed and so accuracy can suffer. Furthermore, allocation of resources can be more or less arbitrary. Bottom-up costing/Activity based costing The bottom-up approach records resource utilisation at the individual service level, and aggregates service level utilisation data to identify the type of resources used and to measure resource utilisation in order to calculate the costs of specific services. It is particularly useful when cost data are not available from other reliable sources [31, 32]. The disadvantages of this approach are the huge cost and long time required for costing complex services [31-33]. Mixed Approach Mixed approaches are based partly on bottom up and partly on top-down approaches [34]. The mixed approach could avoid some of the disadvantages of both methods. A mixed method could be cheaper than using only bottom-up approach and it could be more accurate than using only top-down approach because it can reflect variation in resource consumptions. Top-down costing can be used where resource variation is reasonably small, and/or when the level of aggregation is relatively high, as well as where bottom-up costing would be very expensive and/or would not be worthwhile. On the other hand, bottom-up costing can be used where the precision/accuracy of resource measurement is important, and data collection is feasible in an economically sensible way. Study using mixed approach could suffer from the weaknesses of both methods. Local data may not be externally valid, whereas aggregate data may not be locally representative and could over or underestimate real resource utilisation [35]. Target Costing Target costing was first introduced by Toyota in 1965 [36]. In a target-costing approach, the target profit is established and subtracted from the market driven cost (selling price) to determine target cost of services [37]. After target cost determination, the functional cost analyses are performed in order to reach the target cost.

218

Analogy based estimates In some cases, when similar services or activities have already been valued and the unit costs calculated, information can be extracted from published reports or analysis. It may be helpful to contact the authors directly to discover more details about the costing exercise in order to assess the quality and reliability of these estimates [31]. However, published studies may suffer from weaknesses as good internal validity and poor external validity. Extrapolation based on expert opinion Although expert opinion is generally seen as the least reliable source of information about effectiveness and costs, several studies had to rely on multiple sources when assigning monetary value to resources, including expert opinion [38, 39]. Sometimes this helps where the experts are particularly experienced in the service delivery process. 2.5 Summary The above study identified WLCC research used to cost maintenance and service functions across a spread of industry sectors. The best practice choice is guided by a number of criteria (a) the purpose of costing; (b) the type and complexity of the service; (c) the precision required; and (d) the data availability. From the above review it can be found that maintenance or in-service cost assessment studies in WLCC and maintenance costing literature are more focused towards costing the service associated with stand-alone products. In most of these studies, services are viewed as “add-ons” to products and treated as mere features of the products. They involve the provision of traditional reactive services, which focus on ensuring the proper functioning of the equipment consisting of maintenance, repair and overhaul, spares provisioning, technical publications and technical support [40]. The literature on integrated service support solutions is scarce and assessing the costs of integrated solution is rare. Literature on service costing is more limited to pure service type activities (health care, hospitality). In fact, no studies are found to address the cost estimation of service contracts at bidding stages in service based manufacturing sector, where the offering is in the form of a combination of product and services for a long period of time. Recently services in providing support to the client have grown from just maintenance or overhaul activities over the service life of equipment. This entails the alliance between support with client and supplier initiatives and advances client’s processes, strategies and actions. Up to date, examples of such services involve supply and support chain management, integrated logistics support, asset management, equipment health monitoring and reliability trend analysis [40]. These types of services require closer relationships along the service supply chains. So now the definition of services is changing and very few of the WLCC research actually take an end-toend (customer to supplier) orientation focus in generating cost estimates. Most of the techniques stick on retrospective approaches (basing on past historical data) and do not relate to customer requirements for future. 3 CASE STUDY In UK, the MoD in order to improve the through life management of defence equipment now contracts out based on availability. Under the concept of through-life management, the servicing and supply of spares takes second place to the overall goal of providing the availability and upgrades to mechanical and electrical equipment in one contract, building in potential equipment failures into the cost of the contract. Consequently,

industry is incentivised to deliver reliable and capable equipment (e.g. an aircraft should be capable of flying day or night with the weaponry to defend and attack), reduce maintenance downtimes and minimise the number of required spares as part of a total package of maintenance, repair, overhaul, logistics support as well as equipment availability. Contracting for Availability (CFA) is the mechanism through which this level of support is made possible. In such cases, both the delivery and the availability of the service become part of the service offering. While organizations may be aware of the former and could price/contract accordingly, it is often a challenge to contract/price on the latter as availability is of value to the customer even if the actual service does not get consumed. This paper studies the industry practice of life cycle costing of support service solutions across 5 such availability contracts between MoD and two organisations in defence industry. Among the 5 contracts, 3 are for defence platforms already in service phase, one is in manufacturing phase and one is entering into service phase. Unstructured interviews with cost estimators, contract managers and commercial officers are carried out to extract the best practices and issues around the cost assessment practice of such contracts. 3.1 Inputs The various data, information inputs required for cost modelling are listed below in order of requirements. User Requirements All costing of service contracts are done based on user requirements. For example, if the user says he wants to fly 100 hours, all the services necessary to deliver that are listed. There is a fleet plan showing the number of aircrafts coming in and what services to do. Hence the user availability requirements actually give rise to service breakdown structure. One of the customer requirement documents is MDAL (master data assumptions list). For example, how many aircrafts the customer has got, how is he going to fly them, what rates he is going to fly them at, how many flying hours he has got against each type of aircraft etc are all included in the list. And the defence contractors are contracted against this list. User Top-level Budgets Almost all the projects needed to consider customer affordability issues, their top-level budgets. Service Breakdown Structure (SBS) According to a suitably determined service breakdown structure from 3.1.1 above, data related to the operation and maintenance activities of the system to be maintained needs to be collected. This breakdown structure Historical data & Expert opinions Supply support team for one of the contracts looked at every single repairable item by pulling the history on it. They looked at historical price variance and usages to estimate variance of potential demand. Purchases, average prices, number of transactions in the past act as source of data. A new equipment service contract with no history of service breakdown structure is established by speaking to a number of subject matter experts (SME). Even SME helped develop CERs for parametric modelling of costs. In contracts where past historical data is available, understanding manpower requirements for previous maintenance schedules are used to estimate the time required for doing a maintenance job in current system.

Supplier Inputs In these contracts, as the industry are managing spares, they have to maintain procurement setups and have contracts in place with suppliers for annual supply of certain number of parts needed to meet the flying hours stated in MDAL. So supplier prices for these supplies are required to cost service activities under each availability contract. 3.2 Techniques Using industry standards/catalogues, Expert Judgments For finding the mean time between failures (MTBF), the industry and its suppliers use industry standard data. One of the organisations relied on design expert experience in costing obsolescence. Historical data is not useful for costing obsolescence. Generally there is a commercial standard for estimating obsolescence which depends on equipment costs. Most organisations use engineering judgment to go for planned obsolescence cost assessment. Bottom up costing Three of the contracts studied are mainly based on manpower and employs bottom up approach for cost assessment. Once the profiling of activities carried out in repair and support is done, the profiling of labour and maintenance is carried out accordingly and thus requires use of bottom-up costing. Modelling What-if Scenarios Multiple what-if scenarios (based on usage, for example, hours of flying or days at sea) are run by the contract costing teams to understand spares costing for supporting equipment through its lifetime. The spares provisioning, fixed and variable support costs are summed to give an overall cost profile that has direct relation with availability. These costs are optimised to achieve an optimised operational availability within a budget constraint. Top-Down and Bottom Up One organisation used a combination of both top-down and bottom-up cost modelling technique to arrive at the estimates. MoD’s top level budget and the bottom up costs are compared and problems in terms of driving down costs are identified and costed upon. End to end estimating Four contracts from one organisation use end-to-end estimating process before bidding for a contract. It uses the method of estimation after visualising the point of contract the service is in and where the organisation is in terms of the contracting process. For example, if the organisation realizes an opportunity but has not decided to bid, it uses a rough estimate based on the price of similar equipment service activities along with some riskadjustment. When the company decides to bid, the estimator goes through the company processes of life cycle management and adopts a more thorough approach needing better Statement of work, better drawings, plans and involvement of suppliers. Top down Parametric Estimates In many projects top down estimates are used where enough data is not available. For example in one of the service contracts, the industry’s bid team employed a top down cost estimation process by constructing a SBS first and identifying 10-20. Budgetary estimates are formed based on levels of services required by MoD. Thus “As-Is” costs are established after spending 6-12 months understanding MoD’s service process, cost drivers. Then the industry as design authority finds ways to reduce those costs. This gives the to-be costs. Range of possible

219

savings against As-Is are identified – supply chain, fleet management, optimising capability required in aircraft, managing obsolescence, improving maintenance throughput, reducing levels of risings/faults, reducing nofaults-found, improved testing facilities of aircrafts, reducing spares in supply chain repair loop. In order to minimise costs, one organisation cut off support activities as early as possible in the service life of the equipment by creating a stock buffer. Another key feature is to make maximum use of all equipment drawn down from service to minimise obsolescence, calibration and maintenance costs. Analogy based estimates In estimating maintenance costs of new destroyer ship, parametric estimating in an equivalent ship is used. Then based on how bigger the new ship is the estimators came up with a multiplying factor for the time of maintenance tasks. Similarly analogy is used for all maintenance activities for all other contracts. Joint Cost Model The estimates for one contract are produced by speaking with SMEs in MoD and company on activities, manpower taken, CER, levers and drivers between cost models (MTBF, turnaround times, and repair rate). SMEs populated CERs from their equipment, for Monte Carlo and probabilistic modelling put ranges to the estimates. 3.3

Risks in risk register are populated by joint risk reviews. This has helped as they had to set cost targets based on affordability and availability. So getting MoD and suppliers around the table helps to understand the cost constraints, understanding the kind of support solutions needed to meet them and how flexible they need to be. On analysing the different cost estimation techniques used across different availability type contracts it is found that analogy based techniques, use of historical data and expert opinions are the most popular techniques. Projects at the start of bidding normally tend to use more top-down process of estimating when reliable data is not available. Most contracts used prior experience, huge historical data to do bottom up estimating before submitting the bid. Only one contract compared with the top down estimates to fit affordability requirements set by customer. Others used bottom up estimating to study and improve the As-Is costs of customer. At the same time, interviewees suggested use of analogy based estimates for costing new equipment that has not been in service before. Analogy based estimates are also used in cases where data is available on similar equipment services. Though what-if models are used to study the spares provisioning, spares optimisation is carried out only in one case. Almost all the interviewees use catalogues to base their estimates of material prices.

The generic Process

Expert Judgments/ Experience, Past History

Supplier Estimates

USER REQUIREMENTS USER DATA

Affordability, Availability

Figure 1: Life cycle costing framework for Availability Type Contracts Based on the above findings from industry practices and white boxes. On the top is the programme documentation data inputs in cost estimation of availability type service relating to procurement strategy, how the system will be contracts, a generic framework is constructed and shown deployed in operational and peacetime use and how it will in Figure 1. The figure represents the components of the be supported in these environments. The inputs from cost estimation process. The framework in the figure is suppliers and users would support the understanding of broken down into a number of areas. On the left (in the proposed deployment. At the bottom are the ground green) is the SBS reflecting the system, any specific rules and assumptions (MDAL). These are recorded in a options relating to the system and details of the document and the information would be used to populate programme timescale. This is derived from user data the areas of the model where no hard data was available. originating from user provided details of the activities. The This forms the basis of equipment service at initial stages. inputs to the modelling framework are represented in On the right hand side, the risk issues could be included

220

within the model so as to obtain a ‘risk adjusted’ cost. These risks are mainly process risks generated by talking to experts and experienced people focusing mainly on the service process risks from industry’s point of view. The whole framework depends on expert judgments, past historical figures from similar equipment. To avoid an over complicated implementation it may be necessary to develop some suitable input and output screens to assist the user to find costs for different variations of the input parameters and carry out “what-if” type simulations. It is essential that all life cycle cost models are robustly tested and validated to ensure the correct operation of the equations in relation to the input attributes. And for this multiple iterations to match with user requirements and SME judgments are needed to validate and verify the whole framework. So from Figure 1 and the above practices discussed for various availability type contracts it can be summarised that, there are two dimensions to this model. One is the end-to-end cost estimation in terms of the life cycle stages of the equipment. For example as the equipment moves into service stage of life cycle after manufacturing, the cost estimation process at bidding stage is fairly topdown based approach but for equipment already in service, the contract costing for upgrades and reducing maintenance costs a mixed approach is adopted as past historical evidence and expert judgments are available. The second dimension is the cost estimation across the entire supply network required to support the service of the equipment. This requires equal involvement of the customers and suppliers in the cost estimation approach and is found to be used in one of the contracts. Full visibility of supply chain cost drivers involving customer and suppliers is also essential in cost estimation of this type of contracts. 4 IMPROVEMENT AREAS IN INDUSTRIAL CASE Several areas for improvement can be identified from this research. As can be seen from Figure 1, the risk issues identified in current cost estimating practices are mainly concerned with equipment related risks with no consideration for customer value. This remains a potential area for further research. Accuracy and joint cost modelling remains areas for further improvements in cost modelling of availability type contracts. 4.1 Consideration of Customer-focused risks The current situation seems to indicate that organizations are focused on delivering the availability of their respective equipment, but finds it a challenge on several fronts in properly costing the contract. First, the change from the traditional way of doing business to availabilitytype contract has caused discomfort in terms of understanding the activities involved within the scope of the contract. While clear performance indicators relate to availability, many in the companies are aware that the performance is unachievable without the cooperation of the customer. And the impact of costing these within the contracts is not available in practice or literature. In traditional contracts, the delivery of specific activities (e.g. repair) are directly chargeable and the value to the customer is often directly attributable to what activities are rendered by the company. However, under availabilitybased contract, the performance assessed is the output of the company’s collective efforts and activities, the link between customer value and the activities of the company becomes fuzzier. The defence industry’s activities and structure are mostly product-centric i.e. most of the activities, solutions and models are focused around the tangible aspects of the service i.e. the equipment

capabilities (that are to be maintained, repaired, overhauled or made obsolete) that contribute to the value; without much focus on the behavioural and human capabilities that also contribute towards delivering value. The focus of the organizations on tangible aspects of the service may result in the organization while estimating the costs, overlooking the effectiveness, adequacy and completeness of the service design that brings in human and equipment capabilities that deliver value to the customer. This product-centric approach to cost estimation may also limit the ability of the organization to achieve compliance and efficiency gains when delivering the service on through both equipment and human resources. On the basis of these major challenges, the companies therefore may find themselves exposed to customer-focused risks that threaten the companies’ capability towards delivering service value that is replicable, consistent and scalable across future service projects. The customer-focused risk adjusted cost estimation remains an area for future improvement. 4.2 Accuracy Activity Profile Accuracy First issue is around accuracy of the estimates. Though most of the estimators interviewed answered they improved their estimate accuracy through lots and lots of evidence on what is being required, when is being required. Similarly usage is being recorded in the same way. They spent a lot of time at military bases to understand the cost drivers and actual servicing costs. But accurate visualization is absolutely essential in such costing. Industry is still unsure on predicting future activity profiles, especially under uncertain or unknown situations. The most important challenge of costing service contracts at bidding stage is estimating the activities that are new and several years in advance. There are high probabilities to always miss something and industry or literature has no proper answer on how to improve upon this. New Equipment Cost Estimates The issue of accuracy becomes more pronounced for new equipment service costing where no in-service data is available. All assumptions and estimates are around the equipment’s operating cycle, maintenance process, but estimators don’t know how it will perform in service. From the above case study examples the defence industry is found to rely more on expert knowledge and past data trying to relate it to what is known already. No simulation modelling techniques are utilised in the described case for understanding future in-service operation of the equipment. However this is reported in maintenance costing literature (Bowman and Schmee 2001). Cost Estimates for Obsolescence Most companies estimate mitigation cost of obsolescence based on expert knowledge and intuition on when a component is going to be obsolete. So they trust his wisdom, experience and do the costing based on the certain changes s/he said s/he has to make in the years to come. For unknown obsolescence beyond the knowledge of the expert, intelligent engineering judgments are made balancing the cost impacts of addressing such uncertainties and the affordability of MoD. Information Accuracy The usage data for contracts is based on MoD’s maintenance line information and industry is unaware of what happens in frontline where actually the equipment is used. Hence in spite of spending lot of time in collecting past maintenance data on equipment from MoD, the

221

whole picture might be totally different as availability type contracts are totally based on user’s usage. 4.3 Communication Communication with customers Communication with customer is not as open and involved in most of the contracts studied. This problem is more evident in cost estimation phase of the contracts where customers take the traditional bargaining perspective of “everything costs too much”. Customer has issues around different charging rates and profit level industry puts in. For some contracts which are already in service phase, requirements change a lot. Inspite of MoD’s intention of making availability type contracts proactive, it remains a reactive contract as negotiating every little bit takes lot of time when the actual activity suffers delays. Communication within own organisation Even within own organization, several assignee levels slow things down in agreeing a cost estimate and sending that to customer. Instead of supporting the innovative things the project team wanted to include in the bid, the senior management asked their viability “Do you really want to do that?” as they were more focused on winning the contract even at lower price than usual. Senior management commitment is essential in improving contract cost estimates. Relationships with supplier In one of the contracts, the MoD asked the organisation and its prime supplier both to bid for the same contract. If the supplier had been brought in board earlier before the bidding process competition from own supplier could have been avoided. Joint team development with customer and suppliers is needed to gain the maximum benefits in cost estimation of such type of contracts. The repeated change in solution nature for in-service support contracts hampers supply relationships and incentivizing the suppliers. 5 CONCLUSION This paper through a structured literature review and an empirical study of defence organisations identifies areas for future research in the area of availability type service support contract cost estimation. First of all, there is presently no study on cost estimation of availability type contracts in literature. Current industry practice on availability type contract cost estimation actually uses multiple cost estimation techniques at various stages of the bidding process and equipment life cycle stages. The most popular method used is analogy based cost estimation. Organisations make use of expert opinion in most contracts. The best practice identified are use of combined top-down and bottom-up costing and joint cost model build up involving customers and suppliers. However, the challenges identified in costing of availability type service contracts are: reliability of data supplied by user or assumptions regarding equipment failure (e.g., MTBF), too much reliance on expert opinions might limit innovative thinking of uncertainties and risks, uncertainties of customer’s contribution to availability performance, difficulty of not using bottom up cost estimates in every case, communication problems with the customers, prediction of maintenance activities in future (10-15 years), inability to understand cost impact of customer focused risks. From above it is clear that industry still takes a product centric focus while costing service contracts – attempting to put known numbers for a known task and relating everything back to a known thing.

222

This study provides opportunities for improvement in the cost estimation techniques. Firstly, companies should make people own the cost estimation process and involve people across the entire supply chain who deliver it. Secondly, risk, cost and opportunity should be estimated jointly with customers and suppliers. Thirdly, availability type service contract involves service breakdown structure that cuts through traditional departments and so service cost estimates require cross functional transformation making sure people have cross functional understanding and joint teams are most essential for effective cost estimation at early stages. Hence this research identifies some key focal areas for further research in this area. The first is the research on the effects of customer-perceived risks in cost estimates of availability type service contracts. Then inclusion of human behavioural aspects into the cost models to make them more service focused remains a potential research area for future. Another area for future research is to use the entire supply chain cost data to improve estimate accuracy and involvement of suppliers at bidding stage. Simulation techniques applied to cost estimation models to improve the cost estimation of new equipment remains a potential area for future research. Modelling uncertainties and obsolescence costs of such contracts remains an area for further research. 6 REFERENCES [1] Brax, S., 2005, A manufacturer becoming service provider – challenges and a paradox, Managing Service Quality, 15(2): 142-155. [2] Davies, A., 2003, Are firms moving ‘downstream’ into high-value services?, in Tidd, J. and Hull, F.M. (Eds), Service Innovation, Series on Technology Management, Vol. 9, Imperial College Press, London, 21-34. [3] Xu, X., Chen, J.L.-Q., Xie, S.Q., 2006, Framework of a product lifecycle costing system, Journal of Computing and Information Science in Engineering, 6: 69-77. [4] Kirkham R. J., 2005, Re-engineering the whole life cycle costing process, Construction Management and Economics, 23: 9-14. [5] Asiedu Y., Gu P., 1998, Product life cycle cost analysis: state of the art review, International Journal of Production Research, 36(4): 883-908. [6] Fabrycky, W. J., Blanchard, B.S., 1991, Life-cycle cost and economic analysis, Prentice-Hall, Inc, Englewood Cliffs, NJ [7] Christensen, P.N., Sparks, G.A., Kostuk, K.J., 2005, A method-based survey of life cycle costing pertinent to infrastructure design and renewal, Canadian Journal of Civil. Engineering, 32: 250-259. [8] Curran R., Raghunathan S., Price M., 2004, Review of aerospace engineering cost modelling: The genetic causal approach, Progress in Aerospace Sciences, 40: 487-534. [9] Feldman P., Shtub A., 2006, Model for cost estimation in a finite-capacity environment, International Journal of Production Research, 44(2): 305-327 [10] Roy, R., Kelvesjo, S., Forsberg, S., Rush, C., (2001), Quantitative and qualitative cost estimating for engineering design, Journal of Engineering Design, 12(2): 147-162. [11] Harding A., Lowe, D., Emsley, M., Hickson, A., Duff, R., 1999, The role of neural networks in early stage cost estimation in the 21st century, COBRA 1999:

[12] [13] [14] [15]

[16]

[17] [18] [19]

[20]

[21]

[22]

[23]

[24] [25]

[26] [27]

[28] [29]

The Quantitative and qualitative cost estimating for engineering design. Boussabaine, A., Kirkham, R., 2004, Whole Lifecycle Costing: Risk and risk responses. Blackwell Publishing, London, 1st Edition. Shields, M. D., Young, S. M., 1991, Managing Product Life Cycle Costs: An Organization Model, Journal of Cost Management, 5(1): 39–42. Artto, K. A., 1994, Life Cycle Cost Concepts and Methodologies, Journal of Cost Management, 8(4): 28–32. Wilkinson, S., 1996, Barriers to LCC Use in the New Zealand Construction Industry. Proceedings of the 7th International Symposium on Economic Management of Innovation, Productivity and Quality in Construction, Zagreb, 447-456. Ashworth, A., 1993, How lifecycle costs could have improved existing costing? Lifecycle costs for construction, Blackie Academic and Professional, Glasgow, UK FAA, 2002, FAA Lifecycle Cost Estimating Handbook, Investment Cost Analysis Branch, June 3 2002 Stewart, R., 1995, Cost Estimators Reference Manual. Wiley-Interscience; 2nd edition. Early, J., Wang, J., Curran, R., Price, M., Raghunathan, S., 2007, Dynamic feedback models for whole life cost prediction, Proceedings of the 7th AIAA Aviation Technology, Integration and Operations Conference, 18 – 20 September, Belfast, Northern Ireland. Nicolini, D., Tomkins, C., Holti, R. Oldman, A., Smalley, M., 2000, Can target costing and whole life costing be applied in the construction industry?: Evidence from two case studies. British Journal of Management, 11: 303-324. Jun, H.K., Kim, J.H., 2007, Life cycle cost modeling for railway vehicle. Proceedings of International Conference on electrical Machines and Systems, Oct 8-11, Seoul, Korea. Nilsson, J., Bertling, L., 2007, Maintenance management of wind power systems using condition monitoring systems – Life cycle cost analysis for two case studies. IEEE Transactions on Energy Conversion, 22(1): 223- 229. NATO Research and Technology Organization SAS028, 2003, Cost structure and Life cycle cost for military systems, Report Number: RTO-TR- 058, AC/323(SAS-028)TP/37, September 2003. Department of Defence of USA, 1993, Military standard configuration management, Report number: MIL-STD-973. Komarek, J., 2001, Life Cycle Cost Simulation in Defense Planning, RTO SAS Symposium on Cost Structure and Life Cycle Cost(LCC) for Military Systems, Paris, France, 24–25 October 2001, 1– 4/10. Griffin, J.J., 1988, Whole life cost studies: A defence management perspective, Engineering Costs and Production Economics, 14: 107-115. Jian, L., Hong-fu, Z., 2004, The predictive models of maintenance costs for a civil airplane. Proceedings of the Institution of Mechanical Engineers, 218: 347351. Bowman, R.A., Schmee, J., 2001, Pricing and managing a maintenance contract for a fleet of aircraft engines, Simulation, 76(2): 69-77. Lamb, R.G., 1996, Determining true cost of maintenance performance can generate new profits, Pulp & Paper, 70(10): 93-100.

[30] Beecham J, 1995, Collecting and estimating costs. In Knapp M (ed) The economic evaluation of mental health care. Arena. Ashgate Publishing Limited, London, UK, 61-82 [31] Muennig P. and Kahn K, 2002, Designing and conducting cost-effectiveness analysis in medicine and health care, Jossey-Bass. A Wiley Company, 134-157. [32] Gyldmark M.,1995, A review of cost studies of intensive care units: problems with the cost concept, Critical Care Medicine 23(5):964-72. [33] Beck EJ, Beecham J, Mandalia S, Griffith R, Walters MD, Boulton M, Miller, DL, 1999, What is the cost of getting the price wrong? Journal of Public Health Medicine, 21(3):311-317. [34] Department of Health, 2005, NHS Costing Manual, Version 4.1, Internet version. UK. [35] Luce B, Manning W, Siegel J, Lipscomb J, 1996, Estimating costs in cost effectiveness analysis. In Gold M, Siegel J, Russell L, Weinstein M (eds) cost effectiveness in health and medicine, 176-213 [36] Tanaka, T., 1993, Target costing at Toyota, Journal of Cost Management, Spring: 4-11. [37] Miller, J.A., 1992, Target costing for the chapter 11 business, Faulkner & Gray’s Bankruptcy Law Review, 3(4): 51. [38] Carabin H, Edmunds WJ, Gyldmark M, Beutels P, Levy-Bruhl D, Salo H, Griffiths UK, 2003, The cost of measles in industrialised countries, Vaccine 21(2730):4167-77 [39] Carabin H, Edmunds WJ, Kou U, van den Hof S, Nguyen VH, 2002, The average cost of measles cases and adverse events following vaccination in industrialised countries, BMC Public Health, 2(1):22. [40] Ward Y. & Graves A., 2005, Through-life management: The provision of integrated customer solutions by aerospace manufacturers. Bath Working Paper Series, http://www.bath.ac.uk/management/research/pdf/20 05-14.pdf Access date: 16/04/2008

223

Cost Evaluation Method for Service Design Based on Activity Based Costing K. KIMITA1, T. HARA2, Y. SHIMOMURA1, T. ARAI2 Department of System Design, Tokyo Metropolitan University, Asahigaoka 6-6, Hino-shi, Tokyo, Japan 2 Department of Precision Engineering, The University of Tokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo, Japan [email protected], [email protected] [email protected], [email protected] 1

Abstract As our economy matures, customers have begun to demand more services in addition to just industrial products. To address this problem, designers require a novel engineering methodology, called Service Engineering (SE). SE aims to create value by combining services and products. SE focuses more on increasing customer satisfaction, while general service developers need to take into account economic cost in order to be successful in business. This paper proposes a method to evaluate service from the viewpoints of customer importance and economic cost. The proposed method is verified through its application to a practical case. Keywords: Product/Service Systems, Design Support, Activity Based Costing

1 INTRODUCTION Environmental problems have been quite serious over a couple of decades. To solve this problem, the production and consumption volume of artifacts should be reduced to an adequate, manageable size without making quality of our life lower than now. Consequently, our society must be changed to the new paradigm that aims at qualitative satisfaction rather than quantitative sufficiency, and thus the decoupling of economic growth from material and energy consumption [1]. To achieve this paradigm, products should have more values, supplied largely by knowledge and service contents, rather than just materialistic values [2]. On the basis of this concept, ‘Product-Service Systems,’ [3-5] for example, have been attracting considerable attention, since they can create value by coupling a product with a service. However, few studies have focused on the design of such services (for example, [6-7]). Thus, the authors of this paper have launched conceptual research on designing services from the engineering perspective. This series of research is named ‘Service Engineering (SE).’ [8-10] SE aims to create value by combining services and products. SE differs from conventional engineering in that the design target is customer value, and the purpose is to increase customer satisfaction rather than to achieve more functional products or services. For SE, we have proposed service models and developed a computer-aided modeling tool system called as Service Explorer [8-10]. The purpose of SE, which is currently under development, is to fulfill the requirements of customers. However, in order to be successful in business, general service designers need to take into account the economic costs. In other words, service designers should evaluate a service from the viewpoints of both customer satisfaction and economic costs. In order to serve this need, this paper proposes a method to calculate the economic cost of a service and to support service designers in finding concrete ways to reduce the service costs. First, the modeling method of SE is adopted. In SE, service contents are represented as a set of functions and entities. Then, a management accounting method known as ABC (Activity-Based Costing) [11-15] is applied to the SE model. ABC is a costing methodology used to trace

CIRP IPS2 Conference 2009

224

overhead costs for cost objects such as products, processes, and departments. With respect to a service, this is an appropriate method owing to the characteristics of the service: high rate of overhead costs in the total costs. On the other hand, in SE, a method has been developed to evaluate the importance of functions from the viewpoint of customer importance [16]. This method allocates customer importance to functions by adopting two methods: QFD (Quality Function Deployment) [17] and DEMATEL (Decision-Making Trial and Evaluation Laboratory) [18]. The results of this method are used in portfolio analysis to evaluate functions from the viewpoints of economic cost and customer importance. The present method is verified through an example presented herein. The remaining paper is divided into the following sections. Section 2 introduces the concepts and models proposed in SE. Section 3 explains the models and procedures in ABC. Section 4 presents the proposed method, and Section 5 demonstrates its application. Finally, Section 6 concludes this paper. 2

SERVICE ENGINEERING

2.1 Definition of service Contents State Change

Provider

Receiver

Channel

Figure 1: Definition of a service [8-10]. Service is defined as an activity between a service provider and a service receiver to change the state of the receiver [8-10]. According to the definition, a receiver is satisfied when his/her state changes to a new desired state. For the purpose of SE, the design of services must be based on the state change of a receiver. Therefore, it is necessary to find a method of expressing state changes of the receiver. States of the service receivers

are represented as a set of parameters called receiver state parameters (RSPs), which represent customer value in SE [8-10]. All RSPs are assumed to be observable and controllable. RSPs are changed by ‘service contents’ and ‘service channels,’ as shown in Figure 1. Service contents are materials, energy, or information that directly change the receiver’s state. Service channels transfer, amplify, and control service contents. The parameters expressing service contents, which influence RSPs directly, are called contents parameters (CoPs). In the same way, the parameters of service channels are called channel parameters (ChPs), which influence RSP indirectly. Thus, in SE, customer requirements are represented as RSPs, and the design of a service is based on the degree of customer satisfaction represented as the change of RSPs. 2.2 Function design of service contents As mentioned before, the design of services must be based on customer satisfaction. Therefore, service designers need a modeling method that represents the relationship among customer value, service contents and service channels. In SE, a sub-model called view model is proposed to represent a functional service structure [810]. A view model represents the mutual relationships among an RSP, CoPs and ChPs. An RSP changes only according to the contents of the service received. In other words, service receivers evaluate service contents when they receive the service. Service channels are evaluated indirectly by the receiver and thus do not influence the RSP. SE assumes that service contents and service channels are comprised of various functions. To describe these functions, Function Names and Function Parameters (FPs) are defined. Consequently, both CoPs and ChPs belong to FPs. These functions are actualized by entities. An entity in a view model represents not only physical products but also facilities, employees, information systems and so forth. As shown in Figure 2, a view model is expressed visually using a tree structure, and thus allows service designers to obtain relationships among an RSP, functions and actual entities.

factors. For the description of the human activities and product behaviors, in SE, the activity/behavior blueprint is proposed [20]. A service blueprint, which is originally proposed by Shostack [21], is an effective technique for theoretically analyzing and designing the delivery of services, while the activity/behavior blueprint extends the service blueprint to include product behavior and its relationship to human activities as well as the connection among activities/behaviors, functions and customer value. Since activities and behaviors in the activity/behavior blueprint are related to functions in view models, it is possible for service designers to describe human activities and product behaviors while confirming their influence on the receiver’s state. In this model, the BPMN (Business Process Modeling Notation) model [22] is utilized to describe human activities and product behaviors (see Figure 4).

(Receiver State Parameter)

Customer value  View model

Service Contents (Preliminary functions)

How

Physical Environment

Line of Visibility

Line of Visibility

Invisible

Function parameter

Customer’s Action

Provider’s Activity

Legend

Area

Activity (Task)

Sequence Flow

Pool

Start Event

AND-Split

Data Object

Message Flow

Lane

End Event

XOR-Split

Shape

Figure 4: Activity/behavior blueprint described by Business Process Modeling Notation (BPMN) [20].

Desk

Legend: Receiver state parameter

Behavior blueprint

Figure 3: Overview of the modeling methods in Service Engineering [20].

Actor B

Size per customer

BGM

Hardware (HaW)

Visible

Floor staff

(Ha‐SW)

Activity blueprint

Actor A

Volume

Volume

HaW‐related software 

Visible Invisible

Volume of voice

(Hu‐SW)

Seating comfort

Prepare a table

ChP

HuW‐related software 

Line of Interaction

Play music

Volume of voice

Functions by physical products

Customer

Wait on a customer

Functions by human power

Humanware (HuW)

Prepare a seat

Volume of environmental sound CoP

Customer

Why RSP

Comfortable Environment Control environmental sound

Legend: :  entity :  “is‐realized‐by”

Function Entity

Attribute Parameter

Figure 2: A simple example of view model [8-10]. 2.3 Service activity design Functions in a view model are realized by both human activities and product behaviors that are performed by entities. In SE, these entities are classified into four types: hardware (HaW), humanware (HuW), hardware-related software (Ha-SW), and humanware-related software (HuSW), as shown in Figure 3. These classifications are based on the Software, Hardware, Environment, and Liveware (SHEL) model [19] based on the study of human

2.4 Service evaluation In SE, a method is proposed for evaluating services [16] using QFD [17], which correspond to the view models. This method can quantify the importance of the service functions from the customer importance inputs obtained through QFD. A view model, which is represented by a graph structure, is converted into a matrix expression. A table similar to QFD is generated to appropriately reflect the EM (Engineering Metrics) of a product, i.e., a product’s quality characteristics, to meet the customer’s needs, or the VOC (Voice of the Customer). In addition, the DEMATEL method [18] is used to conduct a quantitative analysis by classifying the FPs into CoPs or ChPs, depending on whether or not they directly affect the RSP. Using this method, RSP importance, which is

225

determined by the customers, is converted into FP importance. Consequently, service designers can obtain the function importance that reflects the customers’ needs. 3 ACTIVITY BASED COSTING ABC (Activity-Based Costing) [11-15] is a costing methodology used to trace overhead costs for cost objects such as products, processes, and departments. In ABC, the resource costs, which include the overhead cost, can be allocated to the cost objects based on the activities. Activities are the operations needed to implement tasks, and resources such as labor, electricity, and facilities are consumed to perform the activities. For example, in order to deliver a product, activities such as designing, assembly, and shipping are essential. Moreover, for these activities, the abovementioned resources are consumed. The ABC procedure comprises two stages. In the first stage, the resource costs are associated with activities based on a cost driver. A cost driver is the criterion for cost allocation. In order to appropriately assign the resource cost to each activity, cost drivers have to be appropriately identified for each resource. For instance, the resource ‘salary’ may be driven by the amount of time the employee spends on an activity. In the second stage, costs are allocated to the cost objects instead of activities based on the number of activities the cost objects consume. This stage can be achieved by using cost drivers similar to the previous stage. Thus, ABC calculates the economic costs by allocating resource costs for activities. Figure 5 illustrates the relationships among cost objects, activities and resources. Cost object

Activity

Product A

Product B

Cost driver

Cost driver

Cost driver

Assemble product A

Assemble product B

Weld product B

as the operations needed to actualize these functions. In addition, since entities are utilized to perform these human activities and product behaviors, entities in view models correspond to the concept of resources in ABC. With regard to quantifying customer importance of the function, the service evaluation method, introduced in section 2.4, is adopted. Based on the results of both analyses, critical functions for the improvement of the service can be identified. The remainder of this section introduces the analysis procedures, described in Figure 6, in detail. 1. Function design Functions (cost object) and entities (resource) 2. Service activity design Service activities (activities) 3. Eco. cost analysis

4. Cus. importance analysis

Economic cost

Customer importance

5. Portfolio analysis

Figure 6: The proposal procedures. 4.2 Analysis procedures This analysis begins with the determination of functions in the service under consideration. Service designers, at first, develop view models introduced in section 2.2. In the view models, function hierarchy that realizes customer requirements is determined, and then entities that actualize the functions are identified. Next, the service designers develop service activities, using activity/behavior blueprint. These activities and behaviors are associated with the functions in the view models. Thus, service designers can identify the relationship among functions, service activities and entities (see Figure 7). Cost object

Labor time 6:4 Resource

Employees

Cost driver

Machines Activity

Figure 5: Relationships among resources, activities and cost objects. 4

THE PRESENTED METHOD

Resource

4.1 Approach As mentioned in chapter 2, the design method proposed in SE aims to fulfill the requirements of customers. In order to be successful in business, however, general service designers need to take into account the economic costs. In other words, service designers should evaluate a service from the viewpoints of both customer satisfaction and economic costs. To serve this need, this paper proposes a method to evaluate a function from the viewpoints of customer importance and economic cost. For cost estimation of the function, ABC is applied to the Modeling method proposed in SE. In the Modeling method in SE, functions in a view model are realized by entities. In the activity/behavior blueprint, on the other hand, human activities and product behaviors are described for each entity, and are related to the functions in the view model. Since functions in view models are regarded as cost objects in this method, these human activities and product behaviors can be identified

226

Legend: RSP (Receiver state parameter)

Function

Human activity and product behavior

Entity

Figure 7: Relationships among functions, service activities and entities. Next, the function analysis is performed with respect to economic costs and customer importance. The economic cost analysis is carried out according to the procedures of ABC. Firstly, the cost of the entities, corresponding to resource cost in ABC, is estimated with reference to existing services, and then is allocated to the service activities. For this allocation, the service designers need to define cost drivers for each human activity and product behavior, for example working hours for a human activity and running time for a product behavior. Subsequently, the cost of each activity is allocated to the functions that

are regarded as cost objects. In this allocation, the degree of association between the functions and the activities is utilized as cost drivers. With respect to the customer importance, the service evaluation method introduced in section 2.4 is adopted. Finally, these results are used in portfolio analysis to evaluate functions from the viewpoints of economic cost and customer importance. Depending on the results of quantification of its economic costs and its importance for the customers, a function is placed in the prepared portfolio. According to the area in the portfolio, recommendations of strategies for the following service design stages are given as shown in Table 1. Table 1: Recommendation strategies. Customer importance

Economic cost

High

High

High Low Low

Low High Low

Recommendation strategy Realize function and reduce economic cost Realize function Leave out function No recommendation

5 APPLICATION The proposed method was implemented by considering an example from the support service for the introduction of an IT system. This service involves supporting a client firm in introducing an IT system. It includes a survey of the current business process, taking decisions on the concept of IT introduction, and so on. In this application, the design-planning phase in IT introduction was considered as the scope of the application. This section explains the application by focusing on the evaluation of economic costs. First, function design was implemented to obtain the functions of service contents and entities corresponding to resources. This was executed according to the view model. First, the requirements of the client firm (corresponding to RSP) were determined through a brainstorming session. Consequently, three RSPs were enumerated: design quality of IT introduction, planning efficiency with respect to its design, and planning quality of its design. Next, these RSPs were decomposed into functions and entities that realized the state change of the RSPs. Consequently, eighteen functions were enumerated (see Table 2). With regard to entity, the employees that consist of class-A consultant, class-B consultant and system engineer were chosen. Second, the service activity design was implemented based on the view model established in the previous step. Consequently, 17 activities were enumerated (see Table 3). Next, each function was analyzed from the viewpoint of economic costs by using ABC. The initial step was to allocate the resource costs to activities based on the cost driver. In this case, the resource and labor costs were allocated to service activities based on the amount of time the employees spent on each activity. As mentioned before, since service activity is a means of actualizing a function, they are associated with the functions. Therefore, the entities that are related to a certain service activity are identified from the relationship between the functions and entities, if there exist several entities in service. Figure 8 shows the relationship between the activities and resources (entities). The economic cost of each activity is described on the lowest line in Figure 8. Next, the cost of each activity was assigned to functions. Each activity cost was divided among the functions related to the service activity according to the proportion of contribution. For instance, the cost of the service activity

“Survey the current situation” was divided into two functions: grasp the current situation and assess the current business process with the customer. The ratio of the contribution was 0.3 and 0.7, respectively. However, no widely authorized recommendation is provided with respect to quantifying the contribution. Thus, the allocation should be executed by a team of experts to ensure good quality of results. Finally, the economic cost of each function was calculated, as shown in Figure 9 (on the lowest line). Simultaneously, a customer importance analysis was conducted. Further, by using the results obtained from the analysis of the functions from the viewpoints of economic costs and customer importance, portfolio analysis was conducted, as shown in Figure 10. Depending on the area in the portfolio, the improvement strategies for each function can be generated. For instance, for an area with high economic costs and high customer importance, the improvement strategy is to realize this function by reducing its economic cost, e.g., ‘Design IT introduction clearly’. For an area with high economic costs and low customer importance, the improvement strategy is to exclude this function, e.g., ‘Obtain the customer's consent to the design’. Table 2: The list of the functions. ID F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 F11 F12 F13 F14 F15 F16 F17 F18

Function Grasp the current situation Expose customer needs Assure the achievement of goals Design IT introduction clearly Design innovative and attractive IT system Obtain the customer's consent to the design Enhance the feasibility of the design Create the plan of IT introduction Enhance the appropriateness of IT introduction Set the price of IT introduction properly Enhance cost-benefit performance Planning the design of IT introduction quickly Lower the cost of the design planning Clarify the process to draw up the design plan Execute drawing up the design plan effectively Build organization Customer support Do precise report Table 3: The list of the activities.

ID A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A11 A12 A13 A14 A15 A16 A17

227

Activity Survey the current situation Extract issues and problems Survey technology trends of IT Report the survey result Build organization Model the current business process Assess the current business process with the customer Report the plan on the occasion with president Model the future of the customer's business Extract issues toward the future Describe the future on the occasion with the president Make decision on the concept of IT introduction Put system modeling in execution Assess the effects of IT introduction Draw up the design plan of IT introduction Review the plan of IT introduction with customer Sum up the blueprint of IT introduction

 Sum up the blueprint  of IT introduction

 Review the plan  of IT introduction with customer

 Draw up the design plan  of IT introduction

 Assess the effects of IT introduction

 Put system modeling in execution

 Make decision on the concept  of IT introduction

 Describe the future on the occasion  with the president

 Extract issues toward the future

 Model the future  of the customer's business

 Report the plan on the occasion  with president

 Assess the current business  process with the customer

 Model the current business process

 Build organization

 Report the survey result

 Survey technology trends of IT

Resources cost

 Extract issues and problems

Resource

 Survey the current situation

Activity

(Unit: 1,000YEN)

 Consultant (A)

3,800

0

0

0

1,900

0

0

0

1,900

0

0

1,900

0

0

0

0

1,900

0

 Consultant (B)

5,100

5,100

10,200

0

5,100

10,200

10,200

15,300

5,100

20,400

15,300

5,100

20,400

15,300

10,200

15,300

5,100

15,300

 System engineer

8,900

89,000

71,200

35,600

26,700

53,400 124,600 124,600

26,700 213,600 151,300

26,700 178,000 160,200

53,400 160,200

26,700 133,500

94,100

81,400

35,600

33,700

63,600 134,800 139,900

33,700 234,000 166,600

33,700 198,400 175,500

63,600 175,500

33,700 148,800

Activity cost (YEN)

Figure 8: Activity-resource matrix.

81,400

Survey technology trends of IT

35,600

Report the survey result

33,700

Build organization

63,600

Model the current business process

134,800

Assess the current business process with the customer

139,900

Report the plan on the occasion with president

33,700

Model the future of the customer's business

234,000

29,250

29,250

Extract issues toward the future

166,600

11,900

35,700

Describe the future on the occasion with the president

33,700

Make decision on the concept of IT introduction

198,400

Put system modeling in execution

175,500

Assess the effects of IT introduction

63,600

9,086

Draw up the design plan of IT introduction

175,500

Review the plan of IT introduction with customer

33,700

Sum up the blueprint of IT introduction

148,800

Function cost (YEN)

20,350

15,257

4,814

4,814

5,086

5,086

13,990

13,990

4,814

6,360

6,360

6,360

6,360

13,990

13,990

13,990

13,990

13,990

5,617

11,900

29,250

29,250

29,250

59,500

35,700

11,900

29,250

6,740

33,067

40,938

6,360

33,067

43,875

43,875

27,257

0

9,086

58,500

58,500

58,500

6,740

6,740

49,600

49,600

6,740

Do precise report

6,360

6,740

13,990

13,990

5,617

5,617

29,250

29,250

6,740

6,740

99,200

33,067

43,875

43,875

9,086

9,086

6,740

45,254 277,718 29,247 308,577 65,215 244,436 11,807

49,600 71,551

Figure 9: Function-activity matrix.

228

14,443

33,700

6,740

47,597

6,360

33,700

16,850

6,740

Customer support

5,086

4,814

33,700

13,990

6,783

5,086

19,080

33,700

Build organization

6,783

Execute drawing up the design plan effectively

6,783

Clarify the process to draw up the design plan 6,721

Lower the cost of the design planning

6,783

6,721

Planning the design of IT introduction quickly

6,783

6,721

Enhance cost-benefit performance

6,721

Set the price of IT introduction properly

20,350

Extract issues and problems

6,721

Enhance the appropriateness of IT introduction

6,783

94,100

Create the plan of IT introduction

6,721

Survey the current situation

Enhance the feasibility of the design

20,164

Obtain the customer's consent to the design

33,607

Design innovative and attractive IT system

Design IT introduction clearly

Assure the achievement of goals

Expose customer needs

Activity cost

Activitie

Grasp the current situation

Function

(Unit: 1,000YEN)

6,360

228,375 47,510

49,115

18,717

27,090

75,857 251,236

Customer importance

0.12 0.1

Design IT introduction clearly

0.08

[13]

0.06

[14]

0.04

Obtain the customer's consent to the design

0.02

[15]

0 0

100000

200000

300000

400000

Economic cost [Yen]

Figure 10: Portfolio analysis. 6 CONCLUSION This paper proposed a method to evaluate service from the viewpoints of customers’ demands and economic cost. Concretely, the presented method enables service designers to calculate economic cost of functions of service. The application suggested that the method could support service designers with finding improvements for reducing cost. Future works include developing well established method to quantify the contribution that is used to allocate activity costs to functions. 7 REFERENCES [1] Tomiyama, T., 1997, A Manufacturing Paradigm Toward the 21st Century, Integrated Computer Aided Engineering, 4:159-178. [2] Vargo, S.L. and Lusch, R.F., 2004, Evolving to a New Dominant Logic for Marketing, Journal of Marketing, 68/1:1-17. [3] Oskana, M., 2000, Product Service Systems, AFR report 288, Swedish Environmental Protection Agency. [4] Tukker, A. and Tischner, U., 2006, Product-services as a research field: past, present and future. Reflections from a decade of research, Journal of Cleaner Production, 14/17:1552-1556. [5] Morelli, N., 2003, Product-Service Systems -a Perspective Shift for Designers-, A Case Study -The Design of a Telecentre-, Design Studies, 24:73-99. [6] Shostack, G.L., 1981, How to Design a Service, in Donnelly, J.H. and W.R. George, eds., Marketing of Services, American Marketing Association. [7] Aurich, C., Fuchs, C. and DeVries, F., 2004, An Approach to Life Cycle Oriented Technical Service Design, Annals of the CIRP, 53/1:151-154. [8] Arai, T. and Shimomura, Y., 2004, Proposal of Service CAD System -A Tool for Service Engineering-, Annals of the CIRP, 53/1:397-400. [9] Arai, T. and Shimomura, Y., 2005, Service CAD System -Evaluation and Quantification-. Annals of the CIRP, 54/1:463-466. [10] Sakao, T. and Shimomura, Y., 2007, Service Engineering -A Novel Engineering Discipline for Producers to Increase Value Combining Service and Product-, Journal of Cleaner Production, 15/6:590604. [11] Cooper, R., 1988, The Rise of Activity-Based Costing- Part One: What is an Activity-Based Cost System?, Journal of Cost Management, 2/2:45-54. [12] Cooper, R., 1988, The Rise of Activity-Based Costing- Part Two: When Do I Need an Activity-

[16]

[17] [18] [19]

[20]

[21] [22]

229

Based Cost System?, Journal of Cost Management, 2/3:41-48. Cooper, R., 1990, Elements of Activity-Based Costing, Emerging Practices in Cost Management, Warran Gorham & Lamont. Cooper, R., and Kaplan, R. S., 1988, Measure Cost Right: Make the Right Decisions, Harvard Business Review, 96-102. Johnson, H. T., 1990, Activity Management: Reviewing the Past and Future of Cost Management, Journal of Cost Management, 3/4:4-7. Sakao, T., Watanabe, K. and Shimomura, Y., 2003, A Method to Support Environmentally Conscious Service Design Using Quality Function Deployment (QFD), Proceeding of the Third International Symposium on Environmentally Conscious Design and Inverse Manufacturing (Eco Design 2003), IEEE Computer Society, 567-574. Akao, Y., 1990, Quality Function Deployment, Productivity Press. Warfield, J. N., 1976, Societal Systems - Planning, Policy, and Complexity, Wiley Law Publications. Edwards, E., 1972, Man and machine: Systems for safety, Proceedings of British Airline Pilots Associations Technical Symposium (British Airline Pilots Associations, London), 21-36. Hara, T., Arai, T. and Shimomura, Y., 2008, Integrated Representation of Function, Service Activity, and Product Behavior for Service Development, Proceedings of the 13th Design for Manufacturing and the Life Cycle Conference, The American Society for Mechanical Engineering (ASME). Shostack G. L., 1982, How to Design a Service, European Journal of Marketing, 16/1: 49-63. BPMN Information (http://www.bpmn.org)

Affordability Assessment of Industrial Product-Service System in the Aerospace Defence Industry O.O. Bankole1, R. Roy1, E. Shehab1, P. Wardle2 1 Decision Engineering Centre, Manufacturing Department, Cranfield University, Bedfordshire, MK43 0AL UK 2 BAE Systems Integrated System Technologies, Eastwood House, Glebe Road, Chelmsford, CM1 1QW UK {o.o.bankole, r.roy, e.shehab}@cranfield.ac.uk, [email protected]

Abstract The Industrial Product-Service System (IPS2) takes a whole life cycle view in order to consider the total cost of the IPS2 offering. This paper focuses on the concept of customer affordability which aims to review current practice in industry and with interaction between customer and solution providers to identify factors affecting affordability. It secures a standard definition and proposes a measurement technique called the Affordability Index (AI) within the aerospace defence industry. A preliminary Affordability Capability Audit Tool is developed to give an indication of the confidence level about the AI. It identifies challenges in industry and outlines opportunities for further research scope. Keywords: Product-Service System, Aerospace Defence Industry, Affordability Assessment, Affordability Index, Affordability Capability Tool.

1 INTRODUCTION A Product-Service System (PSS) has been defined as ‘a system of products, services, network partners and supporting infrastructure that is economically feasible, competitive and satisfies customer needs [1]. It offers dematerialised solutions that minimise the environmental impact of consumption’. A PSS consists of products and services which have tangible and intangible elements combined together to deliver value to the customer throughout its life cycle while ensuring economic profitability for the manufacturer. It is important to ensure that the PSS offering is within the customer spending ability, hence the need for an investigation into the affordability assessment of PSS offerings. This paper focuses on current practices of affordability assessment and measurement of PSS offerings within the aerospace and defence industry, to help to decide whether the customer can afford to pay for a capability contract offered by the solution provider. The paper is structured as follows: Section 2 describes the research method and the design of the capability audit tool; section 3 presents the related research in the area of affordability and Industrial Product Service System (IPS2) contracts, or ‘availability contracts’ as these are sometimes known. Section 4 examines current industrial practice and challenges in affordability prediction; section 5 explains the affordability capability audit tool development while Section 6 contains the discussion and conclusion with the limitation of the research and further research direction.

2. RESEARCH METHODOLOGY 2.1 Literature Review and Questionnaire The methodology adopted in carrying out the research, is presented in Figure (1).

Figure 1: Research Method A review of literature in the area of PSS, affordability and the defence and aerospace industry was carried out by consulting relevant journals and academic papers to gain an understanding of the subject areas. Research in affordability is relatively new both in academia and industry, and so the widest possible search was made to gather information from academic and industrial literature, both published and unpublished. The range of sources reviewed includes masters theses, textbooks, conference papers, doctoral dissertations, industry reports, and unpublished working papers. Databases like Compendex, Inspec, and Emerald were used in conducting the search as well as the GoogleTM search engine. This was necessary to gather existing definitions and measurement techniques in literature in order to be able to compare them to those being used in practice. The literature review informed the design of a questionnaire to be used in conducting interviews. Data collection was performed in three organisations within the aerospace and defence industry. Two of these were suppliers of PSS solutions and the third was a major customer. This customer, UK Ministry of Defence (MoD),

CIRP IPS2 Conference 2009

230

has enough knowledge and experience of dealing with many solution providers across the industry. Therefore, data obtained from the customer is robust enough to depict the current practice in affordability. Some examples of questions asked during the interviews are as follows. (i) What is your understanding/definition of affordability? (ii) What factors drive/ affect your affordability? (iii) How does each factor weigh at the bidding stage? (iv) How is affordability predicted at the bidding stage? (v) How do you monitor the affordability of a project over the life cycle? In the supplier organisations interviews were arranged with functional experts involved in each stage of the PSS lifecycle from bids and proposals through to in-service support. This included those responsible for the design of both the product element and support element of the PSS solution. In the customer organisation interviews were arranged with functional experts involved with the appraisal and evaluation of individual of PSS offerings, the deployment of PSS solutions to the end-users, and portfolio planning /review of alternative PSS propositions against long term budgets. Over 30 hours of interviews were conducted in total, typically in sessions of 60, 120, or 150 minutes. Topics in each session were grouped under relevant headings in order to achieve a logical structure. Verbal responses were captured through audio recording and hand-written notes. Where possible, examples of work products were collected. Data from interviews was analysed using MindManager®. Mind maps produced by this tool helped in understanding the current practice in cost estimation and the use of qualitative and quantitative measures in affordability assessment. The understanding of current practice in the organisations interviewed was compared with the observations from the literature review. A summary of outcomes was presented to each organisation for validation, the outcomes being recorded in the form of presentations and deliverable reports. The outcomes included opportunities to improve current estimating practices in the organisations interviewed (e.g. based on ideas from the literature review), and opportunities to improve the qualitative and quantitative measures originally proposed for affordability assessment base (e.g. based on feedback from the organisations). The latter outcomes have the greater relevance to the work presented in this paper. 2.2. Affordability Capability Audit Development Process The aerospace defence business environment differs from others, because it has fewer customers and more contractors. Contractors are invited to bid for a contract which would be awarded to a suitable contractor (prime contractor). Contracts could be awarded for different stages of the Concept Assessment Demonstration Manufacture In-service Disposal (CADMID) cycle (each could last over 10 years) or the whole CADMID. Due to the duration of availability/capability contracts, it is very difficult to make a good assessment of affordability along the CADMID or for some stages of the CADMID (this is further explained in the section 3.2). Also it is very important to know the factors that affect an availability or capability contract. Then the factors are weighted to know

how much impact they would have on affordability. This is a challenging task since the level of knowledge available at the bidding stage is usually low. It would be valuable for the bidding team to assess its capability to determine customer affordability. No method or technique was available in literature, but an excel-based tool was developed by the authors to assess the capability of the bidding team. This is called the Affordability Capability Audit Tool. The aim of the tool is to assess the capability of the bidding team in affordability prediction based on the level of information available about each of the qualitative and quantitative factors affecting affordability. The process followed in the development of the tool is described in figure (2). First, it was necessary to identify a need which the tool would meet. Through interview sessions and workshops, industrial experts agreed that a capability audit tool would be useful for earlier stages of contracting. Then the factors to be included in the tool were identified. After which a method of scoring was defined to be between 1 and 5, 1 being the lowest and 5, the highest. The next stages were to determine the elements that would enable the bidding team to provide the right score for each factor as well as the questions that would enable them assess each element. The expected result of the tool was represented using Microsoft word. Then the tool itself was developed. Lastly, the tool was tested to see if it produced the desired result and this process was iterated until the desired result was achieved.

Figure 2: Affordability Capability Audit Tool development process 3. RELATED RESEARCH 3.1. Affordability Definition Affordability is the ‘degree to which the Whole Life Cycle Cost (WLCC) of an individual project or program is in consonance with the long range investment capability and evolving customer requirement’ [1]. This is the definition developed by the Network of Excellence in Affordability

231

Engineering (NoE in AE) at Cranfield University. This definition is provided for the aerospace and defence industry and it emphasises the need for a correlation between the WLCC of defence projects and the financial ability of the customer not just to pay the acquisition cost when the contract has been awarded to industry, but across the project life which could last for 40 years or more. It would be useful to consider the understanding of affordability in other sectors. The Merriam-Webster Collegiate Dictionary explains the word ‘afford’ as managing ‘to bear without serious detriment’ [2]. Affordability has been described as the ability to bear the cost of something (software sector) [3]; a ‘measure of whether housing can be afforded by certain groups of households’ (construction sector) [4]; the provision of services which can be afforded by customers at different income levels (utility sector) [5]; the ability to procure a system as the need arises, within a budget, operate at a required performance level and maintain and support it within an allocated life-cycle budget (aerospace sector) [6]. It has been defined as the ability to secure a ‘given standard of housing (or different standards) at a price or rent which does not impose, in the eyes of some

Concept

Assessment

Demonstration

provided to the customer; hence it has the following features:  A physical product core (e.g. aero engine) enhanced and customised by a mainly nonphysical service shell (e.g. maintenance, training, operation, disposal)  Relatively higher monetary value and importance of the physical IPS2 core, and  A ‘business to business’ relationship between IPS2 solution providers and their customers [12]. The project life cycle is usually referred to as the CADMID lifecycle [13] as illustrated in figure (3). Within the aerospace defence industry, an IPS2 is typically characterised as an availability contract spanning the Manufacture and In-service phases of the UK MoD’s CADMID. The customer’s motivation in moving to availability/capability contracts is an improved assurance that the required functionality, performance, and availability will be reliably delivered by the solution provider over a contract duration which could be 40 years or more, and that this will be achieved within a cost profile which is affordable and consistent with the original estimates.

Manufacture In-service

Disposal

Figure 3: CADMID cycle of a typical defence availability contract [13] third party (usually the government) an unreasonable burden on household incomes’ (construction sector) [7]. As expressed above, the meaning and measure of affordability varies from one industry to another because their operational models and strategies for cost appropriation differ. However, two elements are commonly considered in these affordability definitions: customer budget or income and the cost of the product/service or IPS2 offering. The standard definition adopted in this paper is the NoE in AE definition since this study focuses on the aerospace defence industry. 3.2. PSS/IPS2 and Availability Contracts A PSS definition provided by [8] describes the concept of ‘tangible products and intangible services, designed and combined so that they are jointly capable of fulfilling specific customer needs’. Also the solution provider or prime contractor relies on a network of suppliers and service in order to deliver an integrated solution to the customers [9]. PSS has also been defined as a ‘system of products, services, networks of “players” and supporting infrastructure that continuously strives to be competitive, satisfy customer needs and have a lower environmental impact than traditional business models’ [10]. The key elements of a PSS are: (i) Product: a tangible commodity manufactured to be sold. It can be used to fulfil the user’s need. (ii) Service: an activity (work) done on a commercial basis for someone which has an economic value. (iii) System: a collection of elements including their relations [11]. The IPS2 concept provides an opportunity for the solution provider to develop innovative offerings by adding complementary services to the products and systems

232

This assurance is achieved by the transfer of risks from the customer to the solution provider, whereas the customer traditionally purchased ‘spares and repairs’ services on an open-ended ‘as required’ basis, they now purchase ‘in-service support’ for a fixed price. The solution provider’s motivation is that the customer will make a longer-term contractual commitment that justifies additional investment in people, processes, and facilities. This additional investment enables a transition from the purchase of individual products to that of services and system solutions which have the potential to reduce the environmental impacts of the customer’s adhoc needs and wants [14]. The main concern is that the fixed price set by the solution provider for an availability contract can be raised sufficiently to cover its increased risk and maintain profitability whilst still being affordable for the customer. This justifies the need to investigate the affordability of availability/capability contracts. It is important to note that ‘capability’ contracting is an objective within the aerospace and defence industry but, to date, relatively few capability contracts have been placed (e.g. on the basis of including more stages of the CADMID cycle, or more ‘defence lines of development’, within the scope of the contract). Currently, most contracts are availability contracts. 4. CURRENT INDUSTRIAL PRACTICE/CHALLENGES 4.1. Affordability Process Interviews were conducted with functional experts in industry from one customer and two contractors within the aerospace defence industry. Findings revealed there is no uniform definition of affordability from both parties (customer and contractors), but both parties agreed that affordability related to a comparison between the customer budget and the WLCC of the project. The definition developed by NOE in AE at Cranfield University

was proposed and adopted as the standard definition of affordability. The process of affordability assessment was described which was captured from the customer and presented in the flow chart in figure (5). Affordability prediction must be done at the bidding stage to inform the negotiations on the scope and price of a contract such that the customer knows if it has the financial strength to bear the burden of the contract, given the value the supplier is able to provide. The process starts with a cost model (activity) being built by the customer which includes provisions made for risks. The estimate could be refined (activity) before being fed back to the Directorate Equipment Programme (DEP) who is responsible to the MoD Finance Director for the equipment plan. The Directorate Equipment Capability (DEC) is the equipment customer, also part of the DEP. DEP pull together the plan while the DEC manage the priorities and programme (activity). The refined estimates are then measured against top level budget (activity). Upon approval, solution providers from industry are invited to tender for the contract, otherwise the estimate needs to be refined (activity). At this stage the contract specification could be adjusted based on functionality, performance or availability (activity). The financial controller is involved in the process of refining and adjusting the estimates. The customer seeks to build flexibility into the contract. Tenders proposed by industry are examined by the commercial team together with the cost estimate, cost implications of risk and the supply chain sustainability, and then an evaluation is made with the Master Data and Assumptions List (MDAL). After this, through life Value For Money (VFM) is assessed through investment appraisal and through life support till the disposal phase. These are also compared to in-house capability and the traditional types of contracts in order to make a good prediction of affordability (activity). In a single bid, the customer would investigate the solution provider’s finances and require a level of detail during the evaluation process, while contrator’s responses are compared in competitive bid. If the tender is suitable, the contract would be approved with negotiations within parameters (activity). When negotiating the contract with the solution provider, bottom up estimates are done to be able to reduce technical risk and reduce overall cost. The process is iterative in order to get the best solution for the customer. For example reduce availability from 99.9% to 99.5% to achieve cost saving. This negotiation would be taken back to the DEP for approval (activity). This leads to a full contract award, otherwise the whole process starts again (activity). The affordability process represented in figure (4) is iterative depending on contract. The earlier stages of process are done internally before an invitation is sent to contractors in industry. This flow chart is more reflective of an individual project. It is useful to note that providing prices to the customer to support this sort of model normally requires a significant level of company effort and appropriate management approval/review.

secures contracts to deliver the capability required by the end-user (a group of service men). 

Build cost model (1)

Refine estimates (2)

Feedback to DEC/DEP (3) (bi-annual budget process) Adjust (10) specification

No

There is a difference between the customer and the end-user. This is due to the fact that the customer is such a large organisation operating in different parts of the world. The procurement arm sources and

Measure against top level budget (4) Yes

Give round figure to (5) industry & invite to tender

Evaluate tender response(s) (6) for affordability against requirement & VFM

Contract negotiation starts (7) No

Yes

Negotiation within parameters (cost and time) (8)

Contract fully awarded (9)

Apart from the challenge of not having a uniform definition, key challenges identified within industry are as follows:  There is no standard way of predicting or measuring affordability. 

The end-user has no view or opinion on affordability, but it can assess customer value in terms of performance. The end-user does not influence the budget allocation on each contract, hence it has no view of affordability. On the other hand, the contractor does not always know the end- user’s view

Figure 4: Affordability prediction process. of customer value apart from the procurement arm’s view which is based on financial judgement. 

233

Factors affecting Affordability include: (i) Change in requirement

(ii) Competition (iii) Price (iv) Whole Life Cycle Cost (v) Budget constraints (vi) Suppliers (vii) Technology innovation (viii) Perceived quality (ix) Impact of regulations (x) Performance related measure

affordability. These impacts both the WLC at the outset of the project and the affordability of extant projects.  Quality (Q) - Usually, the customer focuses on a specific project and the financial commitment involved in that project. The customer wants to ascertain that the solution is delivered at high quality. Hence, customer’s affordability is influenced by its perception and interpretation of quality within a project.



Usually, the solution provider is only allowed visibility of individual projects, not a program (a combination of projects running parallel or consecutively). This means the contractor may to be able to provide solutions that offer most value for money, within a program. A closer working relationship between both parties would allow a better understanding of customer’s need and possible ways of achieving cost savings and good value for money across projects.



Affordability is determined by the WLCC; hence the accuracy of initial cost estimates has a direct impact on customer affordability. A robust estimate would give a better indication of customer affordability. If the estimate is understated or overstated, it would provide a false indication of affordability.



Where projects have a fixed price agreed by the customer and the contractor, the delivery of such projects could be affected by changes in the cost of resources across the project life cycle. Examples of these are labour rates, fuel price, cost of raw materials as well as other factors affecting the supply chain.

An understanding of the affordability process is useful in deriving the factors affecting affordability. 4.2. Affordability Prediction Results from the interviews conducted with both customer and solution providers revealed that there are two major qualitative factors affecting affordability namely: • CATS – Customer Available To Spend based on customer budget. This is the financial ability of the customer at program and project level • WLCC – Whole Life Cycle Cost. This is the cost from concept stage to disposal; cradle to grave Qualitative factors affecting affordability identified through literature review and interaction with the customer and solution providers are refined and represented in figure (5). These factors affect affordability at varying degrees as seen in Table (1); hence they are used in developing the Affordability Index (AI) and the affordability capability audit tool described in Section 5. In order to include these factors as part of the, AI they are assessed and weighted depending on the impact they have on affordability as shown in Table (1). The qualitative factors are explained below:  World Economic Climate (WEC) - The economic climate is influenced by the inflation, interest rate and share prices. Exchange rate fluctuation between two currencies dictates how much one currency is worth in terms of the other. This could have a negative or positive effect on affordability.  Legislation (L) - Changes in UK, EU and International law, regulations, and protocols concerning environmental, safety, social issues can affect

234

Figure 5: Qualitative factors affecting Affordability  Supplier Chain (SC) - Contractors are increasingly dependent on lower tier suppliers to help deliver both products and services for the duration of the availability or capability contract life. It is a major challenge to ensure continuity in the supply chain over the contract life.  Requirement (R) - The change of requirement increases the WLCC of the project because extra effort is required in redesigning the system especially with be-spoke systems and services.  Global Competition (GC) - The rules of competition could drive the cost down. If competitors are offering lower prices, the supplier could be forced to reduce the cost of the service.  Performance-Related Measure (PRM) - In some contracts the customer may not make full payment until the contract has been delivered, hence the level of customer satisfaction with the delivery and performance of capability could impact the customer’s willingness to pay based on equipment performance.  Political Climate (PC) - The aerospace defence industry’s operations are typically affected by the nation’s political climate. Perceived threats from other nations, could affect the government’s willingness to invest in defence projects.  Unknown (U) - This applies to any other factors which arise depending on the nature of the project. These qualitative factors would be quantified based on the expert judgment of the project team. Therefore the weighting of qualitative factors will differ for each IPS2 project. The AI developed by [15], was modified using factors derived from current research shown in Table (1) to reflect the aerospace defence industry. This is presented in equation (1).

An affordability score equal to 1 is just affordable, a score greater than 1 is more affordable while a score less than 1is less affordable. Though this AI is derived from the customer’s perspective, the solution provider can also use this to understand customer affordability and design capability to accordingly.

Affordability Factors (AF)

Step 3 – a presentation of questions to be answered by providing scores between 1 and 5. This step involves users providing scores for each factor element across the CADMID cycle for all factors in order to predict the capability of the team. There are 3 questions for each of the 3 elements for each of the 6 phases of the CADMID cycle for each of the 11 affordability factors (2 quantitative factors and 9 qualitative factors presented in section 4.2). A screenshot of this is shown in figure (6). Three elements were chosen which could give an indication as to the level of information and resources the project team possess in order to be able to deliver the capability required by the project. These are Information (I), Tools (T) and Skills (S). For each element three questions were asked as shown in figure (6) in order to realise the capability of the project team to assess affordability based on the availability of tools, information and skills from a past project. The questions are presented below. Questions Information (i) Do you have information from similar project? (ii) Level of information on current project (iii) Ease of interpretation of the information

Weight (%)

WEC

9

L

11

Q

10

SC

12

R

13

GC

9

PRM

12

PC

13

U

11

Total

100

Table 1 : Weighting scale for qualitative factors Affordability Index (AI)



CATS WLCC

  1   

   

n

 i1



C i  S i   1  Si

1 = No data 2 = Little amount of data 3 = Just enough data 4 = Sufficient level of data 5 = Plenty of data

  n   

Tools (i) (ii) (iii)

WEC*9+L*11+Q*10 +SC*12+R*13+GC*9+PRM*12+PC*13+U*11 100

(1) Where: CATS = what the Customer has Available To Spend/customer budget WLCC= Expected Whole Life Cycle Cost for IPS2 Ci = Estimated Cost incurred in the ith year Si = Expected spending ability of the customer for the ith year i = the years where cost exceeds the expected spending ability of the customer in that year. n = total number of years the cost has exceeded the spending

Skills (i) (ii) (iii)

Do you have available tool(s) from past project? Do you have tools for this project? The ease of use of the tool(s) Do you have a team/individual from similar project? Do you have man power currently available? Level of expertise

The affordability capability audit tool is a Microsoft Excelbased tool consisting of 6 worksheets. A number of steps were followed in the development of the tool.

The first question under each element is an enquiry about the team’s ability to apply information, tools or skills from a previous project to the current project. The second question is asking to about the level of information, tools or skills within the current project. The third question is an enquiry about the ease of accessing the information, tools and the level of skills of the workforce in the current project. Step 4 - Generation and summary of result The total score for each element is generated under the affordability factors across the phases of the CADMID cycle. This is summarised by averaging the score of all three elements at each phase of the CADMID under each factor to provide a single score for each phase of the CADMID under each affordability factor.

Step 1 – an explanation of aim of the audit tool and the approach taken in designing the tool. Step 2 - guidance to users on how to provide answers to the questions within the tool. In order to use the tool, the user would provide a score from 1 to 5 (1 being the lowest and 5 the highest) for each of the questions under each element for all affordability factors.

The scores are presented in a colour coded table similar to a traffic light system. (All scores would be rounded to the lowest whole number). • Sufficient/ plenty of data – any value from 12 to 15. This is represented by a green colour. • Just enough data - any value from 9 to11. This is represented by an amber colour.

5. AFFORDABILITY CAPABILITY AUDIT DEVELOPMENT

235



No data/ little amount of data – any value from 1 to 8. This is represented by a red colour.

Finally, a single score is presented for each affordability factor at each phase of the CADMID cycle. The output of the tool provides two main benefits: • Assess the capability of the bidding team to judge the customer’s affordability of a project.

improved AI for predicting affordability at the bidding stage and across the CADMID cycle was initially validated with industrial partners through interview sessions and workshops. Further validation would be done through industrial case studies. The affordability capability audit tool is at its early stage of development; hence it would be refined and further validated with industrial case study. In conclusion, the following observations were made:

Figure 6: Capability Prediction Sheet •

Highlight gaps in the availability of information at the bidding stage which is required to measure affordability at different stages of the CADMID. Information used in the development of this tool was obtained from interview sessions with both customer and solution providers. The tool can be used by both parties to assess capability at bidding stage. The tool gives an indication of the confidence level about the AI so the customer can take account of risk and uncertainty associated with an IPS2 project. The tool was initially validated with industrial partners through interview sessions and workshops. The workshops included presentations describing the tool and questionnaire sessions which captured the view of respondents. Most respondents agreed that the tool was useful for affordability assessment at the bidding stage. 6. DISCUSSION AND CONCLUSION Affordability is a new research area which has not received enough attention from researchers while the PSS/ IPS2 theme has been evolving in recent years. This paper has provided a description of IPS2 with main focus on availability/capability contracts in the aerospace defence industry. The paper has provided a definition of affordability. Also, it provided a description of the capability audit tool and the AI for affordability. Though the AI is derived from the customer’s perspective, the solution provider can also use this it to understand customer affordability and design capability to accordingly. Most of the work presented in this paper is based on current research being undertaken by the authors. The

236

i. There is a lack of uniform definition of affordability between solution providers and the customer, however, the NoE in AE’s definition has been adopted as a standard definition for industry. ii. Both solution providers and the customer are not formally predicting affordability of a project over its life cycle; however, an Affordability Index is proposed in this paper. iii. The major challenges for affordability prediction has to do with an understanding of customer spend profile. This is because :  Availability of data is low  There is a lack of understanding about uncertainty  There is a lack of understanding of customer value  There is a challenge in quantifying the qualitative factors affecting affordability. The affordability capability audit tool presented in this paper is designed to fill some of the gaps highlighted above. iv. The present paper proposed a methodology to predict affordability of a project at the bidding stage using qualitative and quantitative factors for the defence aerospace industry. v. Based on the literature and interaction with industry, the major qualitative factors (top 4 based on the weighting in Table (1)) affecting affordability: political climate, requirement, supply chain, performance related measure and the major quantitative factors: WLCC and Customer budget were indentified.

The limitations of the paper are outlined below: The first limitation of the research is that it is specific to the aerospace defence sector. Nevertheless, it is possible to adapt ideas from this research in developing AI for other sectors. The questions included in the tool would be refined to provide more detail in order to improve the robustness of the tool. Also the metric developed in this research would be refined as it reaches the later stages of validation. Further research direction includes the understanding of:  The link between customer value and affordability. This would help to understand how a change in customer value would affect affordability.  Use of AI to inform project management and derive metrics for project control. This would be useful in helping to derive performance measurement metrics to monitor and control the performance of a project at different stages of the CADMID cycle.  Affordability research that can inform budget setting. The customer budget is a major factor affecting affordability so the research could help to inform the budget setting process so the budget is robust enough to help improve affordability. ACKNOWLEDGEMENT The authors would like to thank the EPSRC/Cranfield IMRC and the industrial partners within the aerospace and defence industries for the funds and support they have given towards this research.

REFRENCES [1] Ray, A, Baguley, P. and Roy, R., 2006, “Developing a th Framework for Affordability Engineering” The 4 International Conference on Manufacturing Research th th (ICMR 2006), Liverpool John Moores University, 5 – 7 September 2006:11-16. [2] Merriam-Webster Dictionary Online, Available at: http://www.merriam-webster.com/dictionary/Afford (Accessed 18th June, 2008). [3] Bever, B. and Collofello, J., 2002, “An investigation of techniques for addressing software affordability”, Aerospace Conference Proceedings, 2002. IEEE 5:52577 - 5-2585.

[7] Hancock, K. E., 1993, “Can Pay? Won't Pay?' or Economic Principles of Affordability", Urban Studies, 30 (1):127-145. [8] Doultsinou, N., Roy, R., Baxter, D., Gao, J., 2007, “Identification of service knowledge types for technical product-service systems”, 4th International Conference on Digital Enterprise Technology Bath, U.K. 19-21 September2007. [9] Shehab, E. and Roy, R., 2006, “Product-Service Systems - Issues and Challenges”, The 4th International Conference on Manufacturing Research (ICMR 2006), Liverpool John Moores University, 5th – 7th September 2006,:17-22. [10] Goedkoop, M., van Haler, C., te Riele, H., and Rommers, P. Product Service-Systems, ecological and economic basics. Report for Dutch Ministries of Environment (VROM) and Economic Affairs (EZ), 1999. [11] Baines, T., Lightfoot, H., Evans, S., Neely, A., Greenough, R., Peppard, J., Roy, R., Shehab, E., Braganza, A. Tiwari, A., et al, 2007, “The state-of-the art in Product Service Systems”, Innovative Manufacturing Research Centre, Cranfield University, Cranfield, UK, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, 221 (10):1543-1552. [12] Aurich, J., Fuchs, C. and Wagenknecht C., 2006, “Life cycle oriented design of technical Product-Service Systems” Journal of Cleaner Production, 14 (17):14801494. [13] DASA UK Defence Statistics, 2006, “Chapter 1 – Finance”, Available at: http://www.dasa.mod.uk/natstats/ukds/2006/c1/table115.h tml. [14] Mont, O.K., 2002, “Clarifying the concept of productservice system”, Journal of Cleaner Production, 10:237245. [15] Nogal Miguel, S., 2006, “Development for framework for Affordability Engineering Measurement” (MSc thesis), Cranfield University, Cranfield, UK.

[4] The Centre for Transit-Oriented Development and Centre for Neighbourhood Technology, 2007, “The Affordability Index: A New Tool for Measuring the True Affordability of a Housing Choice”, The Urban Markets Initiative, Market Innovation Brief. Available at: http://www.brookings.edu/reports/2006/01_affordability_in dex.aspx (Accessed 20th Nov, 2007). [5] Milne, C., 2000, “Affordability of basic telephone service: an income distribution approach”, Telecommunications Policy 24:907-927. [6] Kroshl, W.M. and Pandolfini, P.P., 2000, “Affordability Analysis for DARPA Programs”, Johns Hopkins APL Technical Digest, 21(3):438 – 447,

237

An Aerospace Component Cost Modelling Study for Value Driven Design J.M.W Cheung1, J.P Scanlan1, S.S Wiseall2 1

Computational Engineering Design, University of Southampton, Southampton, SO17 8BJ, UK 2 Rolls-Royce plc, PO Box 31, Derby, DE24 8BJ, UK {Julie.Cheung, J.P.Scanlan}@soton.ac.uk; [email protected]

Abstract Demand is increasing in aero-engine products for better efficiency and environmental performance whilst keeping the cost low. Unlike performance, the physics behind cost is least understood. This paper presents a proposed unit cost modelling methodology applied to a Rolls-Royce aero-engine fan blade. An objective of the cost model is the allow engineers to understand the breakdown of cost. A value driven design concept is outlined and presents an opportunity to conduct design optimisation. Keywords: Cost Engineering, Cost Modelling, Unit Cost, Value.

1

INTRODUCTION

1.1 Motivation Rolls-Royce plc has formed a new methods group called the Research and Technology Cost Engineering group to develop cost modelling tools and techniques that can be adopted during the engine design phases. The competitive factors underlying the aerospace industry are performance, cost and reliability. A trade-off is required between these factors in order to produce the optimum solution to meet customer requirements. Figure 1 shows the dependency of the factors. For example, by enhancing the performance this can increase the cost for the technology required, but may decrease the reliability due the complexity of a design.

design phase, in particular, at the early stages of design where there are greater opportunities for cost reduction [2]. Understanding cost drivers is one of the goals in the design cycle. There is now a shift to explore processes that incorporate other measures important to the stakeholder, which as a result generates an optimal design solution. Value Driven Design (VDD) presents an opportunity for this [3]. Further explanation of VDD is in section 2.2. This paper introduces an aspect of a research project by discussing a unit cost modelling approach for novel components in Rolls-Royce plc, such as the composite fan blade, and how this could feature in value driven design. 1.2 Research Scope The aim of the research project is to understand, develop and implement a strategy to allow future generations of gas turbines to be designed to meet not only performance and cost targets but to also take into account other stakeholder requirements in the form of a value objective function. The aim is to develop a costing framework which includes:  Cost modelling case studies – novel components and whole engine level.  Investigation of alternative cost modelling techniques e.g. data mining and proximity engineering.  Exploration of other measures important to the stakeholders and how value driven design can be applied. Figure 2 illustrates the connection between topics that will be considered for researching a value driven design process.

Figure 1: Dependency of business factors. [1] Decisions on the design of a product influence its unit cost e.g. the method of manufacture of the design will affect manufacturing costs; the type of material selected and design determines the material cost and weight. Therefore, it is important to control cost throughout the

CIRP IPS2 Conference 2009

238

© Rolls-Royce plc. All rights reserved.

Novel Component Cost Modelling

VDD Whole Engine Cost Modelling

Proximity Engineering

Figure 3: A cost estimation process [9]

Figure 2: Case studies considered for VDD. 2

BACKGROUND

2.1 Cost Engineering Today, there is a comprehensive selection of cost estimation literature for engineering. Curran et al explains that cost estimating provides a forecast of a product’s cost generated from analysing historical data [4]. The European Aerospace Cost Engineering Working Group (EACE) created a list of definitions to describe the work involved in cost engineering [5]. The following is a few which outline some of the Rolls-Royce cost engineering activities:  cost estimating methods and processes development;  cost estimation;  cost control;  design to cost;  development of cost models;  participation in IPPT (Integrated Product/Process Team)  value analysis. The current aspiration is to work towards a more automated cost estimating process that can provide credible cost information [6]. To achieve this, an understanding of the design and manufacturing process is required, which highlight that cost depends on design decisions made about the product [7]. Furthermore, a set of good quality historical data and knowledge to support the cost estimation process will allow quicker responses to the designer (Figure 3). Niazi et al describes the various techniques available for product cost estimation [8]. The granularity of the cost information depends on the type of data used in the cost estimating method or cost model. For instance, intuitive techniques are based on the past experience. This requires an extensive product range in order to generate a good correlation for cost estimation. Analytical techniques delve into more design and manufacturing detail, hence breaking down a product into activities and resources consumed in the development process. Cost engineering takes into consideration design and engineering principles and uses this knowledge to evaluate trade-off studies. Therefore, it is important to treat cost as an independent design parameter [4], which can feature in life-cycle costing and optimisation processes.

2.2 Value Driven Design “Value-driven design is an improved design process that uses requirements flexibility, formal optimization and a mathematical value model to balance performance, cost, schedule, and other measures important to the stakeholders to produce the best outcome possible” [3]. Value driven design is an emerging topic within the concurrent engineering community, as it provides a concept where designers can utilise value models to determine the value of their product designs as a single objective function. Collopy has applied the value modelling approach to aerospace products, whereby profit is used as a metric for ‘value’ of the product to the business within the competitive market [10]. However, in engineering as opposed to ecomonics, a single objective function that represents the design attributes (Figure 4) is much more desirable [11]. The single measure or scoring function would theoretically indicate the ‘goodness’ of the product design, where a high ‘score’ yields a better optimal value-adding product [10] e.g. for a component in an engine, the output would be the value that the component provides to the engine system level. This technique can then support decisionmaking when assessing options in the early stages of design. Investigation of the possibilities of value modelling and value analysis of aero-engine components will be within the scope of this research project.

Figure 4: Scoring function for design.

239

3

COST MODELLING APPROACH

3.1 Novel Component Unit Cost Modelling Novel components, in this case, mean a product that has not been produced before in Rolls-Royce plc. It is a completely new unique component with a new design and manufacturing process. In this unit cost modelling case study, a composite fan blade is considered. It is not unique in the aero-engine industry as General Electric has developed the GEnx and GE90 engines, which uses composite fan blades [12]. Figure 5 shows an assembly of Rolls-Royce plc Trent 1000 titanium fan blades [13].

3.2 Methodology The methodology presented in this case study for novel component cost modelling adopts a similar approach to Shehab’s [16] cost modelling framework whereby the design, manufacturing process and material attributes are modelled. In addition to this a discrete event simulation (DES) factory model is introduced to capture the dynamic operations in the factory. Two software packages are used for modelling the manufacturing cost of the fan blade: Vanguard Studio [17] and ExtendSim [18]. The combination of a static and dynamic model benefits one another to help the user understand and highlight the cost drivers. Figure 5 shows a high level breakdown of the manufacturing cost. Manufacturing Cost Variable - Materials - Labour - Scrap

Figure 5: Traditional Rolls-Royce plc titanium fan blades. An aim for Rolls-Royce plc to deliver light weight and low cost composite fan blades has recently been announced [14]. Today, alternative materials and manufacturing methods have been considered in the aero-engine market due the rising demands from airline businesses for better efficiency and environmental performance [15]. Therefore, an alternative technique to re-use historical data or experience is needed for novel component cost modelling.

- Equipment - Maintenance - Invested Capital

Figure 5: Manufacturing costs. Figure 6 is a diagram of the framework by which to model the manufacturing costs. It is proposed that Vanguard Studio acts as the user interface with ExtendSim running in the background. Therefore, design attributes are entered in Vanguard Studio. The transfer of data between the two packages is further explained in the subsequent sections. Marsh et al linked a factory model to Vanguard Studio to demonstrate the advantages and shortcomings of the system [19]. Vanguard Studio Cost Model The DATUM project [20] introduced Vanguard Studio (formerly Decision Pro) to Rolls-Royce plc as a cost modelling tool. The fixed costs have been captured in the ‘static’ model e.g. machine and tooling costs, amortised costs, raw material cost. Figure 7 shows the type of data entered for the Vanguard cost model. Notice that the equipment and labour times are fed into the factory model.

Figure 6: Cost modelling framework applied to novel components

240

Fixed

Figure 7: Abstract of the Vanguard Studio cost model data structure. discussing the model with experienced manufacturing engineers (ME’s). Hence, it is important for validation that Factory Model the model reflects the ME’s thoughts. The factory model is broken down to manageable Currently, manual optimisation is required via ‘what-if’ hierarchical blocks (Figure 8). In each cell are the experiments to generate the optimal factory layout to meet operations required for that particular process e.g. buffers, annual demand i.e. number of tooling or operators. operator tasks and operations. Example scenarios for the factory model: The model acts as a capacity study, whereby the factory Apply limited resource (operators and equipment) – operation is modelled and tested to evaluate if the annual experienced manufacturing engineers can apply their demand of fan blades is met. Equipment and labour times, knowledge of factory operations in the model. The with uncertainties (distributions) defined in Vanguard model can then be simulated to discover any Studio, are applied to the factory model. The output of the bottlenecks and also highlight where the manufacturing simulation identifies queuing times of work-in-progress; process/logics could fail. bottlenecks and resource availability. When the outputs are Apply unlimited resource – this scenario assumes there analysed, this can help resolve factory logistic issues and is an unlimited financial budget for resource, which is improve efficiency to yield the target annual demand. unlikely in real life. Therefore, this scenario can determine how many equipment, tooling and operators are required to meet the annual demand. Compare automated and manual composite manufacturing processes to highlight the benefits of each process in terms of cost. This is a useful scenario to aid decision-making and the long term cost benefits. 4

FUTURE WORK

4.1 Cost Modelling Study Further development and analysis is required for the composite fan blade cost modelling case study:  The Vanguard Studio cost model will need fine-tuning to accurately represent cost information and RollsRoyce’s accounting system e.g. amortise cost either over number of years or number of parts produced. As both the cost and factory model reaches a maturity level; the goal is to develop a reusable robust model to accommodate design changes. Figure 8: High level composite manufacturing process blocks for factory modelling. An advantage of building a dynamic model is gaining credible manufacturing costs based on the manufacturing process [21]. Therefore, deriving the unit cost is traceable which helps with the verification process - when

The results of the factory model scenarios will be analysed. The effects of these scenarios on the Vanguard Studio cost model will be recorded and sensitivity analysis applied. All models will need to go through a rigorous verification and validation process in order to achieve a mature status.

241

Whole Engine Unit Cost Model Figure 6 shows that a library of cost models can be utilised, which holds Rolls-Royce standard material cost information; standard labour cost rates; equipment and process cost rates. Additionally, a library of engine component cost models exist that can be used to build a whole engine cost model. Therefore, the cost model for the composite fan blade can be included in the library and implemented into the whole engine model to assess the cost impact of a design change. 4.2 Value Modelling The fan blade cost study is a potential application for value modelling – the value it brings to the engine and the business. Collopy states the ideal engine design is one that maximises unit profit. This acts as a parameter in a value model and brought together with an engineering cost model, presents an opportunity for design optimisation [22]. Currently, a deeper understanding is required to define value and how a tool can translate engineering parameters, customer needs and cost into a single ‘goodness’ objective function or score. A methodology will be explored to verify that value driven design is capable of representing design attributes and beneficial to the design and manufacturing community. 5 SUMMARY This paper has discussed the importance of understanding cost at the early stages of design to help design decision making. The cost modelling approach for a novel component was presented along with the scope for a value driven design process. The implication of the cost modelling methodology for a novel component supports design trade-offs and emphasises the importance of unit cost as a parameter in the design optimisation process. For new products, it may not be necessary or possible to deliver detailed cost. Instead, the cost information generated from the cost models reflects the amount of design attribute details available at the particular design stage. Further developments and analysis of the method will reveal the credibility of the cost results, which combined with a value model provides a mean for design optimisation. In future, this approach can be adopted for other aero-engine components. 6 ACKNOWLEDGMENTS This work is part of the author’s Engineering Doctorate (EngD) at the University of Southampton, sponsored by Rolls-Royce plc and the Engineering and Physical Sciences Research Council (EPSRC). 7 REFERENCES [1] Cheung JMW, 2008, Value Driven Design. University of Southampton Engineering Doctorate Conference. National Oceanography Centre, Southampton, UK. [2] Tammineni SV, Rao AR, Scanlan JP, Keane AJ & Reed PAS, 2007, A Hybrid Knowledge Based System for Cost Modelling applied to Aircraft Gas Turbine Design. University of Southampton. [3] Home Page of the AIAA Value-Driven Design (VDD) Program Committee. Available at: . Accessed on: November 2007. [4] Curran R, Raghunathan S & Price M, 2004, Review of Aerospace Engineering Cost Modelling: The Genetic Causal Approach. Progress in Aerospace Sciences; 40: 487-534.

242

[5]

[6]

[7]

[8]

[9]

[10]

[11] [12]

[13]

[14] [15]

[16]

[17]

[18] [19]

[20]

[21]

[22]

Lewis D & Pickerin H, 2001, A Capability Improvement Model for Cost Engineering. The European Aerospace Cost Engineering Working Group. Tammineni SV, Rao AR, Scanlan JP, Reed PAS & Keane AJ, 2008, A Knowledge-Based System for Cost Modelling of Aircraft Gas Turbines. Journal of Engineering Design; DOI: 10.1080/09544820701870805. Weustink IF, Brinke E, Streppel AH & Kals HJJ, 2000, A Generic Framework for Cost Estimation and Cost Control in Product Design. Journal of Materials Processing Technology; 103: 141-148. Niazi A, Dia JS, Balabani S & Seneviratne L, 2006, Product cost estimation: Technique classification and methodology review. Journal of Manufacturing Science and Engineering; 128: 563-575. Tammineni SV, 2007, Designer Driven Cost Modelling. Doctor of Philosophy Thesis, University of Southampton. Collopy P, 2002, Value Modelling for Technology Evaluation. DFM Consulting, http://www.dfmconsulting.com/research.htm. Scanlan JP, 2008, A Brief Statement Inviting VDD Research. University of Southampton GE90 Fan Blade Article. Available at: . Accessed on: July 2008. Image on an Assembly of Fan Blades. Available at: . Accessed on: July 2008. Composite Material Joint Venture Article, July 2008. Aerospace Manufacturing, Vol 3, Issue 20, p7. Overview of Rolls-Royce plc. Available at: . Accessed on: October 2007. Shehab EM & Abdalla HS, 2002, An intelligent knowledge-based system for product cost modelling. International Journal of Advanced Manufacturing Technology; 19: 49-65. Vanguard Studio - a Product of the Vanguard Software Corporation. Available at: . Accessed on: March 2008. Imagine-That-Inc, 2007, ExtendSim User Guide: Imagine That Inc. Marsh R, Cheung WM, Lanham H, Newnes L & Mileham A, 2007, Modelling an Assembly Process using a Close Coupled Generative Cost Model and a Discrete Event Simulation. 4th International Conference on Digital Enterprise Technology Proceedings. Scanlan J, Rao A, Bru C, Hale P & Marsh R, 2005, The DATUM project: a cost estimating environment for the support of aerospace design decision making. Journal of Aircraft; 43: 1022-1028. Potter J, 2000, The Effectiveness and Efficiency of Discrete-Event Simulation for Designing Manufacturing Systems. Doctor of Engineering Thesis, Cranfield University. Collopy P, 2001, Surplus Value in Propulsion System Deisgn Optimization. DFM Consulting, http://www.dfmconsulting.com/research.htm.

Profitability of Industrial Product Service Systems (IPS²) – Estimating Price Floor and Price Ceiling of Innovative Problem Solutions 1

M. Steven1, M. Rese2, T. Soth1, W.-C. Strotmann2, M. Karger2 Chair of Production Management, Ruhr-University Bochum, Germany 2 Chair of Marketing, Ruhr-University Bochum, Bochum, Germany [email protected]

Abstract Companies from industrialised nations are faced with the threat of competition from low-cost countries. We suggest Industrial Product Service Systems (IPS²) as a possible answer. But as the development and production can be quite expensive for the supplier, the question arises how the net benefits of an IPS² for the supplier can be determined to ensure that the IPS² is profitable. We establish a framework for the calculation of both the supplier’s revenues and costs of an IPS². Requirements induced by possible subsequent changes of the IPS² are emphasized. We propose a combination of the Net Present Value Approach and the Real Options Approach as a means of determining the quantified revenues and a combination of Direct Costing, Time-Driven Activity-Based Costing and the Real Options Approach for the calculation of the costs of an IPS² for a supplier over its life cycle. Keywords: Industrial Product Service Systems; Net Present Value Approach; Real Options Approach; Time Driven Activity-Based Costing, Learning Effects, Price Ceiling

1 COMPETITIVE THREATS AND CHALLENGES FOR MARKETERS Companies from established industrialised nations are faced with a multitude of threats, caused especially by companies from developing nations such as India or China. In the past, these threats were primarily based on the common practice of imitating products of competitors from developed, industrialised nations. These imitations exacerbate the amortisation of investments in research and development and can even render them obsolete. Growing capabilities and competencies of such competitors from developing nations pose a further threat, since companies from developed industrial nations are unable to compete with the low labour costs of the aforementioned companies. Highly dynamic markets pose the additional challenge of having to generate sustainable competitive advantages under changing conditions. Focusing on providing products does not suffice to create a viable economic basis for company success [1]. Markets have experienced a shift of focus from products to market requirements and an augmentation of the importance of services. Encompassing this, significant effort is dedicated to an interwoven integration of products and services in order to generate a sustainable competitive edge and prevent out-suppliers from penetrating the customersupplier relationship. Against this background of changing environmental conditions we suggest Industrial Product Service Systems (IPS²) as a possible solution. IPS² are product-service mixes tailored to fit individual customers’ needs. IPS² are stamped by an integrated and mutually determining process of planning, developing, provisioning, and using of goods and services [2]. It is important to note, that in an IPS² context services are no longer merely viewed as an add-on to products, but rather as am equal part of an integrated solution. In this paper we specifically focus on determining the customer value of an IPS²; i.e. the benefit of an IPS² for the supplier. This is done by estimating both the supplier’s

CIRP IPS2 Conference 2009

243

revenues and costs that incur when offering an IPS². Our method is sufficient to account for the specific characteristics of an IPS² which are changes of the IPS² over its life cycle and the combination of products and services. The remainder of this paper is organised as follows: Chapter 2 provides a definition of IPS² and specifies criteria which have to be fulfilled in order to ensure the success of an IPS². In chapter 3 we focus on the determination of the revenues an IPS² generates for the supplier. In chapter 4 we introduce a method to estimate the costs of an IPS². The paper concludes in chapter 5 and gives an outlook onto further fields of research. 2 CONFIGURING AN INITIAL IPS² The goal of offering IPS² is to establish a customersupplier relationship which cannot be easily broken up by out-suppliers. The integrated development of productservice mixes tailored to fit individual customers’ needs can generate entirely new barriers to imitation, allowing a company more long-term competitive advantages [3]. When it comes to the configuration of a tailor-made problem solution for an individual customer, one inherent characteristic of IPS² is of utmost importance: the possibility of partially substituting product-based and services-based components. This allows for various possible ways of executing customer processes, servicebased or product-based. We label these technological possibilities as different mixtures of manual and automatic execution of processes. Furthermore, a second dimension has to be considered. This dimension describes the customer decision towards make or buy of processes. This two-dimensionality, the variability of technology on the one, and the decision of internal or external production on the other hand, generate additional degrees of freedom for customers and suppliers which generate a variety of potential problem solutions that

could be offered to customers. In this connection the characteristics of an IPS² are not static, but can be changed over the lifecycle. This flexibility allows to adjust the characteristics of the IPS² tailored to changing environmental conditions and customer preferences, which increases the benefits of the IPS² for the customer and hence the price or revenues respectively that the supplier can gain from offering the IPS². When calculating the Economic consequences have to be anticipated as best as possible by the supplier and taken into account when choosing which IPS² solution to offer the customer. Each IPS² has to fulfil three basic economic criteria: i) it has to generate a positive value contribution for the individual customer, ii) this value has to be higher than that of the best competitor’s offer and iii) the value creation on the supplier side has to be positive as well [4]. The criteria i) and ii) have to be fulfilled because otherwise the customer would choose either not to invest in an IPS² or to choose an offer from a competitor. Criteria iii) ensures that the IPS² is beneficial for the supplier as well and is decisive for the question if a specific IPS² should actually be offered to the customer. As criteria iii) can only be fulfilled if criteria i) and ii) are fulfilled, we focus on the determination of the supplier’s net benefit of an IPS², which is calculated based on the suppliers revenues and costs of the IPS². 3

ESTIMATION OF SUPPLIER REVENUES

3.1 The Case of non-flexible IPS² The revenues of an IPS² for the supplier equal the price the supplier can charge. Hence, in order to estimate the revenues we have to estimate the price the customer is willing to pay for the purchase and the use of the IPS². To estimate the price ceiling, we have to understand how the customer makes a decision about an investment. A customer will invest in an IPS² only if its value contribution for the customer is positive. The value contribution can be calculated using the Net Present Value approach. Let NPV0 the NPV, P the price the supplier charges for the IPS², Rt and Et the revenues (inpayments) and expenses (outpayments) occurring in period t, and r the rate of return, then the NPV can be calculated as follows: n

NPV0 = − P + ∑ (Rt −E t )⋅(1 + r)−t

(1)

t =1

This means that the NPV0 at time 0 (time of contract) is equal to the discounted value of the net income stream (income R minus expenses E) from the IPS²’s use over periods 1 to t, less the initial payment P0, the purchase price. For the sake of simplicity, it is initially assumed that the supplier has no competitors and the customer decides whether to invest or not to invest in the supplier’s problem solution. The price ceiling is where the NPV equals zero, i.e. the discounted net income stream, also referred to as the project value PV, equals the purchase price: n

P max = PV = ∑ (Rt −E t )⋅(1 + r)−t

( 2)

t =1

If competitors exist, the theoretical upper price limit is determined by the strongest competitor. Let PS be the price of the focal supplier and PC be the price of the competitor, R. The differential advantage of the supplier’s IPS²-solution compared with the competitor’s is defined as: n

[

NPVOS − NPVOC =−(P S −P C )+∑ (RtS −RtC ) − (E tS −E tC ) ]⋅(1+ r ) −t

(3)

t =1

n

[

]

0 = −P S +P C +∑ (RtS −E tS )−(RtC −E tC ) ⋅(1 + r) −t t =1

= −P S +P C +PV S − PV C

(4 )

Solving for PS gives the price at which the customer considers both suppliers equal: P max =P C +PV S + PV C

(5 )

To verbalize this, the focal supplier’s price can be greater than the competitor’s to the extent that the project value for their IPS² is greater [5], [3]. 3.2 The Case of a flexible IPS² As mentioned above, a crucial characteristic of IPS² is the flexibility of the IPS²-configuration over the life cycle, tailored to changing customer preferences. A customer is not bound to an initially chosen configuration, but can choose to flexibly adjust this configuration to changing environmental and structural conditions [6], leading to an increased IPS² price ceiling and hence to higher revenues for the supplier. Flexibility can be taken into consideration combining the NPV-approach with the Real Options approach (ROA). The ROA takes into account flexibility by considering a multi-stage decision process with a decision in t=0 and another decision in t=1 (t=2;…; n). The decider can choose from a set of possible alternatives in t=0, based on all information about future developments and conditions available at that point in time [7]. A decision at t=0 is accompanied by substantial uncertainty, owing to the fact that future developments and conditions are hard to predict [8]. This is not the case for a similar decision in t=1, however. The aforementioned developments are already under way, triggering a substantial reduction of the decision maker’s uncertainty. This degree of flexibility and its consequences for the initial IPS², which has to allow for change options and the inclusion of these into its configuration, can have a substantial impact on the profitability of IPS² for both the customer and the supplier. On the one hand, it is to be expected that expenses on the supplier side rise with growing flexibility, as the preparation of possible changeovers requires the hold-out of the capability to perform the different options. On the other hand, flexibility results in an increase of the value which an IPS² generates on the customer side, because the customer can react to possible changes of the preference drivers, leading to increased income or reduced expenses respectively [9]. Let us illustrate the ROA using the following example: The supplier offers an IPS² to a specific customer. The customer intends to use the IPS² for a life cycle of 6 periods (years). We assume that the required intensity of maintenance depends on the production volume, so that a higher production volume leads to an increased number of required maintenances. As the customer does not have the knowhow and the technical possibilities to perform the maintenance himself, it has to be performed by the supplier. The number of maintenances and hence the maximum production volume is bound by the contract. In t0, the customer can choose between a low frequency and a high frequency of maintenances which would enable a higher production volume. After three years, the customer has the possibility to alter the initially chosen frequency of maintenances which allows him to increase the production volume. This kind of flexibility is of advantage for the customer as he can only predict the production volume under uncertainty.

with superscript S designating the focal supplier and C the strongest competitor. Setting the NPV-difference to zero leads to

244

During the periods one to three, the demand is estimated to remain low with a probability of p- and to increase with a probability of p+. Let us further assume that the customer will first choose the contract with a low maintenance frequency and will only switch to a high maintenance frequency in case of a demand increase, as this will maximise D4. The Project value now can be calculated as an expected value taking into consideration the optimal decisions in t4 in case of a low production opt/opt/+ volume D4 , in case of a high production volume D4 and the probabilities of the occurrence of these developments or decisions respectively:

low maintenance frequency high production volume high maintenance frequency

p+

low maintenance frequency p-

low production volume

p+

3

PV = ∑ (Rtl −E tl )⋅(1 + r)−t + p − ⋅ D4opt/ − + p + ⋅ D4opt/ +

p- high maintenance frequency

t=4

p-

=

change of the production volume

=

decision knot

=

probability of a low production volume

p+ =

(7)

t =1

t=6

probability of a high production volume

Figure 1: Dimensions of IPS² To determine the value of flexibility, one has to start with determining the decisions which a client would make under the different environmental conditions. The result is a sequence of optimal decisions which customers will make to maximize the expected Net Present Value of their IPS². For the determination of the optimal decisions for the customer the rollback method can be applied. This method has been introduced by Magee and is based on the optimization principle of dynamic programming [10], [11]. The initial step of the rollback method is to determine the expected value of the latest possible decisions in the different states. The value of each decision depends on the prospected sum of the positive discounted cash flows which would occur after the decision is made [12], [8]. We refer to this value as the decision value. At each possible decision point the customer will choose the option which leads to the highest decision value [13], [12]. Only these optimal decisions are regarded further in the analysis. Subsequently the values of the second-latest decisions are calculated. This is done by summing up the prospected positive discounted cash flows which occur until the next decision, and adding the values of the optimal subsequent decisions, whereas those decision values have to be multiplied with the probabilities of the states in which the decisions will be made. A successive continuation of this procedure up to the firstmade decision during the investment leads to the decision values of the possible initial specification of the IPS². The decision values of the initial configuration then equal the expected value of the IPS². In our example, one has to start with determining the decision which the customer would make in t4 conditional to the different production volumes which are now known by the customer. As the customer will try to maximise his NPV, he will opt for the maintenance frequency which leads to the highest decision value in t3: 6

6

t =4

t =4

D4opt = max[D 4l , D4h ] = max[ ∑ (Rtl − Etl ), ∑ (Rth − Eth )]

(6)

For the determination of the price ceiling we first have to determine the project value. Let us assume that the demand will be low in t0 when the investment is made.

245

In order to determine if the customer would really opt for a low maintenance frequency in t0, we have to calculate this project value for an initial choice for a high maintenance frequency as well. The customer then will decide for the option leading to the higher project value. To determine the price ceiling, we just have to put this project value into formula (2) in case of no competition or into formula (5) in case of competition. The charged price then equals to the revenue of the IPS² for the supplier. When calculating the price ceiling of a flexible IPS² one has to bear in mind that the supplier can influence the decisions by his price setting. For example, if the supplier would not charge the price for the maintenance in form of a higher purchase price but in form of monthly rates, then the supplier can charge different prices for the different maintenance frequencies. By means of the price setting the supplier can influence the values of the decisions and thus control the decisions. Hence, in this case the decisions of the customer not only influence the price ceiling, but the price also influences the decisions of the customer. The roll back method enables to take into consideration decisions about several characteristics of the IPS², but the number of decision values which have to be calculated increase explosively with the number of options and the number of time points in which decisions can be made, so that even modern computers would be overstrained. Hence, in case of complex decision trees it is necessary to use heuristics or simulation methods to calculate the project value. 4 ESTIMATION OF SUPPLIER COSTS In chapter 3 we have shown how to determine the IPS² price ceiling in order to estimate potential supplier revenues. To enable a sound profitability assessment, we now focus on the costs induced by different IPS². It is essential to correctly determine supplier costs of offering different IPS², as this enables suppliers to set a price for an IPS² which at least covers the costs induced by it [14]. From the supplier’s point of view an IPS² is economically feasible only when the IPS² price ceiling is higher than the IPS² price floor. To calculate costs directly induced by a product, traditional cost systems, such as direct costing [15], [16] can be used. Therefore, these costs do not pose a problem for the IPS² cost calculation. What is challenging, however, is the growing proportion of overhead costs. As soon as 1985 Miller/Vollmann speak of a “Hidden Factory” which cannot be accounted for with traditional costing systems [17]. This is especially the case for services, whose costs tend to be overlooked in the bulk of product overhead costs and which often evolve to become unpredictable cost drivers.

It is therefore necessary to make use of a costing system which is capable of establishing transparency with regard to the origination of overhead costs and allocate these to the cost units fair according to the input involved. Activity Based Costing, which assumes that a multitude of cost generating processes are needed to produce a product or service, seems to best serve this purpose. What is problematic, however, is the fact that Activity Based Costing can mainly be used for repetitive actions with very limited room for decisions [18]. Activity Based Costing is therefore primarily applicable to standardized activities. However, customer integration, which is at the heart of IPS², triggers the need for customized activities. As a consequence, the majority of the processes necessary for customized activities are no longer repetitive as is required to make use of Activity Based Costing. What is needed is a modification of the traditional Activity Based Costing approach, to enable its application to processes with individualized cost patterns. Time-Driven Activity-Based Costing (TDABC), as introduced by Kaplan and Anderson is such a modification [19]. TDABC allocates capacities and costs to sub processes based on their target processing times and their net cost unit capacities. The target processing time is the time needed to conduct a process once and is defined by time per output unit. This shows that time is considered to be the main cost driver. In this context it is important to note that target processing times must not be confused with average processing times as derived from traditional Activity Based Costing. Target processing times contain neither idle time nor additional time. A cost unit’s net capacity can be obtained based on the number of employees and their net working time. The net working time can be obtained based on the gross working time minus vacation, illness and unproductive times. By multiplying target times with the output the sub process net capacity can be derived. By summing up all sub processes belonging to a cost unit the capacity utilization of a focal period can be determined. The difference between cost unit capacity and utilized capacity is the unused capacity of a cost unit. Such a calculation can provide insight into a company’s capacity utilization. In the context of IPS² this insight is crucial when it comes to deciding whether or not to build up additional capacity when accepting additional orders. Making use of time equations is another characteristic of TDABC. A time equation is used to allocate target times to sub processes according to their input involved [20]. Consequences for cost calculations resulting from customer integration can be given explicit consideration in the time equation for each individual customer. To do so, processing times x1 to xn are determined in addition to a basic time frame, which represents the processing times without the effect of customer integration. A basic time frame for conducting a maintenance process of a standardized machine could for example be complemented by additional time if a customer acquires a machine customized to his needs which has higher maintenance requirements. Furthermore, a customer might have highly skilled personnel, which reduces the need for maintenance through preliminary work. Such a situation could also find consideration in the time equation in form of a processing time, which would be deducted from the basic time frame.

By making use of a time equation it is therefore possible to get away from the rigid process costs of traditional process cost calculations. With TDABC process costs can be determined separately for each individual customer and each individual IPS². To apply TDABC to IPS², however, the approach needs to be extended, as constant process costs are not to be expected over the IPS² life cycle. In the context of cost management at an early stage, IPS² costs therefore need to be calculated under explicit consideration of a non-static development of process costs. Against this background we integrate learning effects into the TDABC system. To do so, processes need to be categorized with regard to their innovativeness. Innovativeness substantially influences which part of the experience curve forms the basis of the IPS² calculation. The concept of the experience curve states that costs per unit will decrease, if the accumulated production output increases. Three process types can be distinguished: 1. Customized process related to one IPS² only 2. Customized process related to more than one IPS² 3. Standardized processes The first two categories comprise all processes which need to be changed or re-developed to configure an IPS². Processes of the first category are highly customized to meet an individual customer’s needs and can therefore only be used for one specific IPS². Processes of the second category are also being changed respectively newly developed, but are not as highly customized and can therefore also be used for other IPS². While the learning curve starts with the IPS² configuration for both categories, the learning rate in the second category depends on the output of the IPS² for which the focal processes are required. The third category comprises all existing processes which are used in a newly configured IPS². The learning curve of these processes therefore does not start at zero [21]. Cost development estimations of IPS² processes can be integrated into the time equation of TDABC. This means, that when calculating IPS² costs it is anticipated that processes needed in the future are subject to different costs incurred as is the case when the IPS² is developed. Therewith, process costs can be calculated under consideration of a dynamic cost development over the whole life cycle. Thereafter the respective number and temporal occurrence of the necessary IPS² processes needs to be estimated. After that the IPS² price floor can be determined by summing up the IPS² process costs of the separate periods and a subsequent discounting to the time at which the cost calculations are made. As discussed in chapter 3.2, flexibility plays a major role with regard to IPS². In the following, we therefore focus on the costs associated with offering flexibility. As a means of illustration we employ a scenario. What has to be noted is that the innovative IPS² solutions discussed in this paper are reduced in complexity for our scenario. The manifold IPS² dimensions which challenge the calculation of costs and revenues for IPS² are broken down to the dimension of maintenance. We follow this way of conduct in order to better be able to elaborate on the means of calculating costs and revenues for IPS². Our scenario is as follows:

246

Cm3

Cm1

Cm4

Cm5

Cm6

Ct4

Ct5

Ct6

Cm2

CU3 Ct3

P b= Probabilty to abandon the option

Cm = total costs of manual maintenance of the period

P a= Probabilty to exercise the option

Cu= set-up costs

=decision knot

Ct = total costs of automatic maintenance of the period

Figure 2: Total costs of the same scenario A customer acquires an IPS² with an assumed life cycle of 6 periods. Cost considerations in this scenario are limited to a maintenance contract which is part of an IPS². Furthermore, we assume that the need for maintenance depends on the overall output. The higher the output is, the greater is the need for maintenance. Maintenance can be conducted either via manual or via automatic process execution. Automatic maintenance requires high initial outpayments, whose amortization is dependent on a high maintenance frequency, as the costs for each maintenance are higher for manual than for automatic maintenance. The number of maintenance processes to be conducted by the supplier in period 1 and 2 are fixed by contract beforehand. Owing to a low expected output, these maintenance processes are to be manually conducted to save costs. An increasing demand and therefore an increasing output is expected for period 3, which would result in a higher maintenance frequency. The flexible IPS² component consists of the following switching option: In period 3 the customer can decide to switch from the contractually agreed upon low maintenance frequency to a high maintenance frequency. If the customer in our scenario pays a fixed price for maintenance, which is independent of the actual maintenance frequency, the supplier will choose to switch from manual (service) to automatic (product) maintenance execution. By using the previously described methods it is possible to determine customers’ willingness to pay for such a maintenance contract. The willingness to pay represents the customers’ price ceiling. To determine IPS² profitability it is now necessary to also determine the price floor. To do so, costs which incur if customers’ make use of their switching option must be calculated. The following costs can be identified for our scenario: 1. Costs for materials needed to reconfigure the machine. These are direct costs, which can be calculated with the Direct Costing approach. 2. The number of employees needed to reconfigure the machine. Costs for the configuration processes for which the employees are needed can be determined with TDABC.

247

3. It might also be the case that customer employees need additional training after the reconfiguration. TDABC can serve to calculate the training costs. In a next step the probability that customers will really make use of their switching option needs to be estimated. By doing so the expected value of all costs related to maintenance along the IPS² life cycle can be determined. Figure 2 illustrates the situation at hand. The expected overall costs for the IPS² maintenance including the switching option can be calculated as follows: ⎞ ⎛ t =6 ⋅ (1 + r)− t + pb ⋅ ⎜ ∑ Cmt ⋅ (1 + r)− t ⎟ + t =1 ⎠ ⎝ t =3 t =6 ⎛ −t −t ⎞ pa ⋅ ⎜ Cu ⋅ (1 + r) + ∑ c t ⋅ (1 + r) ⎟ t =3 ⎠ ⎝

Cg =

t =2

∑C

mt

(8)

Costs for manual maintenance (Cmt) are subject to change over the life cycle, owing to the previously described integration of learning effects into cost calculations. Costs for automatic maintenance (Ct) can be assumed to be constant. The summing up and discounting of the costs related to maintenance provide the price floor for the IPS² maintenance contract including the switching option. To calculate the overall IPS² price floor all costs incurring over the IPS² life cycle can be included in the formula. If the IPS² price floor surpasses its price ceiling, the IPS² is economically not feasible and will therefore not be offered to customers without further adjustments and reconfigurations. Hence, it is possible to assess the expected IPS² profitability using the methods introduced in this paper. If various IPS² can serve to solve customers’ problems and are economically feasible, the one with the greatest difference between price floor and price ceiling should be chosen.

5 CONCLUSION To the many challenges companies are facing today, IPS² could constitute a solution. But only if the supplier’s costs are lower than the revenues, an IPS² is really profitable, whereas a supplier should offer the IPS² which leads to the highest difference between revenues and costs. The

subject of this paper was the calculation of an IPS² costs and revenues which takes into consideration the specific characteristics of an IPS², being the possibility to change an IPS² and the service parts. One possible way of determining the revenues is a determination of the price ceiling using a combination of the net present value approach and the real options approach. The mere application of the Net Present Value thereby renders the customer value of an IPS² without any flexibility, leading to a systematic underestimation of the value which the IPS² creates for the customer. Only the combination of the NPV-approach with the Real-OptionsApproach enables a reliable estimation of the true value of the IPS². The costs are accounted by combining the direct costing, the time driven activity based costing and the real options approach. Future research should investigate the optimal IPS²configuration from the supplier’s point of view which leads to sustainable high profits. An application of the methods introduced in this paper using a case study from industry could also be beneficial for the further establishment of IPS².

6 REFERENCES [1] Prahalad, C.K., Ramaswamy, V. 2003, The New Frontier of Experience Innovation, MIT Sloan Management Review, 12-18. [2]

Meyer, H., Uhlmann, E., Kortmann, D., 2005, Hybride Leistungsbündel –Nutzenorientiertes Produktverständnis durch interferierende Sach- und Dienstleistungen, wt Werkstattstechnik online, 95/6:528-532

[3]

Rese, M., 2006, IPS² Cost Decisions and Price Decisions in Time of Value Based Management, in: Plötner, O.; Spekman, R. (Eds.): Bringing Technology to Market, 61-76.

[4]

Plinke, W., 2000, Grundlagen des Marktprozesses, Kleinaltenkamp, M., Plinke, W. (Eds.), Markt- und Produktmanagement, 3-99.

[5]

Oxenfeld, A.R., 1966: Executive Action of Costs for Price Decision, Industrial Marketing Management, 6/1:83-140.

[6]

Pindyck, R. S., 1991, Irreversibility, Uncertainty, and Investment, Journal of Economic Literature, 29/3:1110-1148.

[6]

[7]

Miller, K.D., Waller, H.G., 2003: Scenarios, Real Options, and Integrated Risk Management, Long Range Planning, 36/1:93-107. McGrath, R. G., MacMillan, I. C., 2000, Assessing Technology Projects Using Real Options Reasoning, Research-Technology Management, 43/4:35-49.

[8]

Alexopoulos, K., Mourtzis, D., Papakostas, N., Chryssolouris, G., 2007, DESYMA: Assessing Flexibility for the Lifecycle of Manufacturing Systems, International Journal of Production Research, 45/7:1683-1694.

[9]

McGrath, R., Ferrier, W.J., Mendelow, A.L, 2004, Real Options as Engines of Choice and Heterogeneity, Academy of Management Review, 29/1:86-101.

[10] Magee, J. F., 1964a, Decision Trees for Decision Making, Harvard Business Review, 42/4:126-138. [11] Magee, J. F., 1964b, How to Use Decision Trees in Capital Investment, Harvard Business Review, 42/5:79-96. [12] Smith, J. E., 2005, Alternative Approaches for Solving Real-Options Problems, Decision Analysis, 2/2:89-102. [13] Brandão, Luiz E., Dyer, James S., Hahn, Warren J., 2005, Using Binomial Trees to Solve Real-Option Valuation Problems, Decision Analysis, 2/2:69–88. [14] Reckenfelderbäumer, M., 2004, Die Wirtschaftlichkeitsanalyse von dienstleistungsorientierten Geschäftsmodellen als Herausforderung für das Controlling, in: Meier, H (Eds.): Dienstleistungsorientierte Geschäftsmodelle im Maschinen- und Anlagenbau, 209-242. [15] Arnstein, Wiliam E., Gilabert, F. 1980, Direct costing. [16] Andries, E. 1979, Direct costing. [17] Miller, J.G., Vollmann, T.E., 1985, The Hidden Factory, Harvard Business Review, 63/5:142-150. [18] Coenenberg, A., Fischer, T. M., 1991, Prozesskostenrechnung Strategische Neuorientierung in der Kostenrechnung, Die Betriebswirtschaft 51/1:21-38. [19] Kaplan, R. S. Anderson, S. R., 2007, Time-Driven Activity-Based Costing. [20] Coners, A., Hardt, G. von der, 2004, Time-Driven Activity-Based Costing, Controlling & Management 48/2:108-118. [21] Pfohl, M., 2002, Prototypgestützte Lebenszyklusrechnung.

248

Life Cycle Cost-Orientated Service Models for Tool and Die Companies

G. Schuh, W. Boos, S. Kozielski Chair of Production Engineering, Laboratory for Machine Tools and Production Engineering RWTH Aachen University, Aachen, Germany [email protected]

Abstract As the success of a company’s service provision is founded in its business model, the latter needs to be redesigned to align strategic and operational objectives. Therefore at RWTH Aachen University a new approach to service models for the European tooling industry links products and services by means of a direct calculation of life cycle costs. Tool and die makers are enabled to offer product-service-systems, which allow expanding the range of service provision by directly addressing the cost-effectiveness of the whole product-service-system. Hence the minimisation of life cycle costs of tools and dies can be used as a new sales pitch for services in this unique industry. Keywords: Business Models; Product-Service-Systems; Tool and die making; Life Cycle Costing

1 INTRODUCTION Tool and die companies play an important role within the production industries of Germany, Japan, the United States and a few other countries as studies show [1, 2, 3]. Their key position is a result of the responsibility for industrial value chains in terms of time, costs and quality (figure 1). product development tool & die making tools & dies

facilities

parts production

parts finishing

final assembly

final product

2

Figure 1: Tool and die making within the industrial value chain. Purchasing departments of original equipment manufacturers (OEM) use electronic platforms more and more frequently. Ford Motor Co. expects between 70 to 80% of purchasing transactions to take place on the World Wide Web within the next years [4]. Online submissions of quotas, mainly known as e-bidding, leave the product price as the only selection-criterion. Therefore the range to differentiate oneself from the competitor is decreasing rapidly. Exclusive differentiations in price have not worked out for German tool and die makers over the last years. Furthermore, the basic criteria for such a differentiation are not given in Germany. Therefore a promising approach for differentiation between competitors is to combine the existing range of physical products with customer-specific services. The latter is an integrated product and service offering that delivers value in use (product-service-system) [5]. In order to do this, the information asymmetry between tool and die makers and their customers at the point of sale has to be resolved; otherwise there will be no chance

CIRP IPS2 Conference 2009

to enforce the price of the product-service-system [6]. To solve this problem, business science offers the theory of signaling [7]. Since investments within the production industries are mainly cost-driven, life cycle costs of product-service-systems prove to be a viable signal. As the success of a company’s service provision is founded in its business model, the latter needs to be redesigned to align strategic and operational objectives. The design of a product-service-system needs to be integrated in the strategic planning and positioning of the company. To consider these challenging demands for tool and die makers, a definition of an integrated service model is developed. The design of this model is made possible by detailed knowledge about the life cycle costs of the product-service-system as a whole.

249

COHERENCE BETWEEN BUSINESS MODEL AND SERVICE MODEL In general, products are primary reason for customer loyalty; however differentiation within markets can only be achieved through customer-specific solutions. These solutions are no longer sold as a physical product, but rather as a service bundle. The extension of existing business models provides a chance for German tool and die companies to sustainably improve their competitive position. Valueadded can be achieved on the basis of Porter’s three generic competitive strategies: overall cost leadership, (product) differentiation and focus [8]. Criteria for differentiation in toolmaking can be: time in terms of adherence to schedule, productivity and life span of the tool – whereof each will allow a price premium, if the information asymmetry can be resolved. Many approaches to business models lead to various aspects which a business model can consist of. Table 1 summarizes different elements of models according to Mueller-Stewens and Fontin [9], Bieger, Rueegg-Stuerm and von Rohr [10], Knyphausen-Aufsess and Meinhardt [11] as well as Hamel [12].

Mueller-Stewens/ Fontin

Bieger/ RueeggStuerm/ von Rohr

KnyphausenHamel Aufsess/ Meinhardt

value proposition

incentive system

product/ market combinations, customer value

core strategy

marketing

communication concept, growth concept

product/ market combinations

interface to the customer

production of goods organization, and services cooperation mechanism, coordination mechanism, configuration of competencies

configuration and execution of the value added

strategic resources, value added network

benefit

profit mechanism

interface to the customer

benefit concept

3 ELEMENTS OF SERVICE MODELS All four elements of a service model are linked and dependent on each other. The detailing of the four elements is highly based on an initial configuration of the product-service-system. The competitive position is to be defined along two dimensions: range of offered tools and range of offered services. Broad formations allow the integration into the customer’s processes. Focused formations offer the possibility to use economies of scale and to extend profound know-how. This leads to four possible configurations of toolmakers offering productservice-systems.

Table 1: Synthesis of approaches to business models. In this context, the main aspect in the definition of a business model should be the capitalization and the benefit mechanisms of a company. This also applies to designing business models for tool and die makers that offer product-service-systems. As stated above, the integration of service offerings into a company’s business model is a crucial factor of success. The latter can be achieved by the implementation of a sub-model into a company’s business model. This service model impacts every element of the specific company’s business model. Based on the approach of MüllerStewens and Lechner, the structure of a service model can therefore be divided into four elements (figure 2): • Value proposition: which value is offered to the customer? • Marketing: attracted?

how

can

appropriate

customers

be

• Benefit: how is the profit mechanism to be designed? • Production of services: how shall the output be generated? [13]

+

Service model • Modelling service offerings • Detailed in configurator

Value proposition

Marketing

Is a new part of

• Specialists are focused not only on the offered tool range, but also on the range of offered services. They are experts in special demands in a niche. • Standard-Suppliers are capable of delivering every possible tool. Usually they only offer repair services. This position demands a wide range of machining capacity. • All-rounders are distinguished by their high flexibility and competence in providing solutions. They are neither focused on specific tooling technologies nor on special service offerings. This position requires profound know-how and a small range of key customers. • Customer-integrators are highly customer-focused. They focus on a special tooling technology and support their tools throughout their entire life-cycle. All offered services are geared to specific customer problems and demands. As described above, a service model consists of four elements. One can speak of a Strategic Fit, if all elements of the service model are harmonized with each other [13, 14, 15]. Beyond this, the individual elements are designed such that they support one another. This fit is of particular importance for the tool and die making industry: the cardinal positions of innovative and standardized services lead to varying needs and development potentialities. A concentration on standardized services makes uniquely defined interfaces to adjoining sectors and speediness essential. Innovative services on the other hand demand a steady learning process as well as a significant involvement in their customers’ processes.

Business model Production

Benefit

• Modelling product offerings • Detailed in business plan

Figure 2: Coherence between service model and business model. Once the service model has been designed completely, incentive systems tie in with market and performance decisions while offering extensive and efficient problem solutions to the customers. Incentive systems set three objectives: 1. Structuring the product. 2. Make the customer aware of versatile services. 3. Differentiate oneself from competitors. The more distant service provision is from the initial product, the more customer-specific it has to be designed. Individual partial performances are bundled together into customer-specific packages, thus being advantageous for both the customer and the supplier. Furthermore, the integration into the company’s strategic planning and design of value-added processes is guaranteed. Therefore service models are of great importance for companies.

3.1 Value Proposition The value proposition defines all services that supply the certain needs of a customer. Thereby services which consist of modular bundles [14] gain in importance: Based on competitive strategies, a service model has to configure reasonable service- and market-combinations. These allow a precise drawing of conclusions on how to achieve competitive advantage. On this basis, the value proposition defines which services are offered to which customers and how differentiation from competitors can be achieved. The latter notably results from the combination of a premium core-product and a unique service portfolio. As services are even harder to copy than physical products, the existing service landscape is extended by customerspecific product-services in an incentive system, such as guaranteed availability, preventive maintenance or process analysis (figure 3). Results from the research ING and development project “SMART STAMP ” show, that further integration of tool and die makers into the set-up of the parts production process is explicitly desired by ING their customers. “SMART STAMP ” is a German research and development project and its basic idea consists of two objectives:

250

Stages of Life Cycle: Development, Optimization

Simultaneous Engineering

Process/ Product Development

Optimization

Process/ Product Optimization Prototypes

Core Competency: Tool Manufacture

Process Qualification Logistics

Spare/ Replacement Parts

Tool Efficiency

Maintenance

Aftercare

Stages of Life Cycle: Operation, Withdrawal

Figure 3: Extension of the service portfolio along the entire life cycle of tools and dies. 3.2 Marketing A ‘strategic triangle’ is formed by the customers, suppliers and competitors of tool and die making companies. It describes a market in which service offerings are exchanged. The dependencies among the three vertices of the strategic triangle are to be defined in this element. Overall, it has to be considered that marketing defines not only the attitude towards the company’s customers, but also towards their competitors and suppliers. Marketing is closely related to the definition of services to be rendered. It sets off activities to identify and address the needs of economically attractive customers. This process should be supported by adequate communication tools and the corporate claim. Especially a distribution channel in shape of a key account management allows a direct communication by personal contact with a distribution staff member or (in case of a small toolmaker) the company owner. Thus the individual needs of a customer can be addressed and holistic customer-specific solutions can be achieved. Furthermore, marketing tools, such as internet or exhibition presence and publications in scientific journals, are used to advertise the company itself and its innovative solutions. 3.3 Benefit What matters in planning benefit is the question concerning the company’s profit mechanism: •

Units, usage and output are determined and a benefits basis built up.

Achievable benefits from transactions are estimated and benefit levers are selected. A trade-off between the two contradictory strategies, of maximized availability and utilization of the entire production machinery, needs to be assessed. It has to be specified what is sold and which service deliveries are put down to the customer’s account. An important aspect is the price policy along with the pricing itself, which results from the concrete design of the profit mechanism. Thereby the different stages of benefit for the company are determined. The establishment of a two-stage tariff/ payment rate leads to an extension of the cash flow and provides a basis for a long-term increase in profit. Calculations reveal that further profit can be realized in the utilization phase of the tool’s life cycle. Afterwards the total price is composed of a basic price and usage•

251

3.4 Production of services In a holistic approach to service models it is therefore essential to examine the creation of value throughout the enterprise as a whole. The manner of the value proposition is defined by the production of services. Based on the configuration of the value added the discrete processes are described. Resources and skills are allocated; make-or-buy-decisions are made. Moreover, partner contributions are determined and coordination mechanisms as well as communication channels are installed among the partners and among the defined processes. Regarding larger tool and die companies with corresponding resources, the separation of process chains into different chains for the classic tool making process itself on the one and a process chain “service” on the other hand is a promising possibility. The process chain “service” usually proceeds without the steps of construction, planning and scheduling. A faster, more flexible reaction to customers’ demands can thus be realized. 4 LIFE CYCLE COSTS: ENABLER FOR SERVICE MODELS 4.1 Classification and aim of life cycle costing Life cycle costing integrates a new point of view into the companies’ accounting procedures. In addition to the activity-dependency (fixed and variable costs) and the cost-unit-dependency, the life-cycle-dependency is introduced as a third cost-application-criterion. Figure 4 shows, how different accounting standards can be classified along the two dimensions object and time.

Company

A data interface between stamping plant, press manufacturer and die maker is defined to enable condition-oriented services.

Object dimension



dependent charges, whereby the toolmaker is paid the service price at a certain point of time during the tool’s development phase. Thus the risk of dependence on a specific market can be reduced for the customer through this usage-dependent financing model. In doing so, the operational risk for the tool is carried by the toolmaker – which is a deliberate result from offering a guaranteed availability of the tool.

Projects, Products

Sensor technology is directly applied to press and die in order to collect process-data and data on the actual condition.

Orders, Processes



Liquidity Calculation Life Cycle Costing Income Statement

Process costing

Short-term

Mid-term

Long-term Time dimension

Figure 4: Classification of life cycle costing. [15] • Process costing is a process simultaneous ratio calculation. Its object-orientation lies on single parameters of the running production processes. It is used to control costs of operational processes and is short-term oriented. • Income statements serve as a permanent control of the production of goods and services. Objects of focus can be orders, projects or whole companies. Because of its periodicity it is short- and middle-term oriented.

4.2 Life Cycle of tools and dies In scientific literature the life cycle of production equipment is split into three sections: the acquisitionphase, the utilisation-phase and the disposal-/ rejectionphase. [18] This approach can also be applied to tools and dies. A characteristic in toolmaking is the cost intensive optimization of tools after the machining process. Additionally, the customer has a vast influence on the optimization process of the toolmaker by determining the rules of acceptance. Based on empirical analysis in the ING research and development project “SMART STAMP ”, the design of the interface between toolmaker and customer in this section of the life cycle is of high relevance. Conflicts may occur if tool purchasers limit their decision model on acquisition prices and special quality requirements at the same time. To further analyse this interface, the life cycle of tools and dies is modelled in four sections (figure 5).

4.3 Life cycle costing of product-service-systems In analogy to accounting standards, a product-servicesystem is regarded as a cost unit to conduct a calculation of its life cycle costs. From the tool’s manufacture to its disposal it binds and consumes resources before it enables parts producers to create value added. As described above, the life cycle of tools and dies can be interpreted as an aggregation of processes. Within these processes, which are phases of the life cycle, resources are consumed by production, use and disposal of the tool as well as by the development and provision of services. This consumption of resources causes different types of costs throughout the entire life cycle. For example, cost types can be materials needed for machine processing which are entered on specified accounts and have to be estimated as a whole. In the end, life cycle costs are calculated by means of an activity-based costing. Within the modelling of life cycle costs, cost elements are defined to enable the conduction of the activity-based costing [19]. A cost element determines what costs are caused by the consumption of resources in a specific part of a process throughout the life cycle (figure 6). The total amount of process costs is determined by the interaction of defined cost drivers. These factors have to be collected and evaluated for each process within the life cycle of the product-service-system.

Construction

Planning

Mechanical Processing

Ressources

Simulation

Assembly

Optimization Tryout

Delivery

Initial Operation

Utilization Parts Production Maintenance Logistics Disposal/ Recycling Provision of Spare Parts

Recycling

Figure 5: Detailed life cycle of tools and dies. The section “Manufacture“ begins with simulation, construction and planning processes to initiate the machine processing and ends with the assembly of all parts. Following this, the aim of optimization is setting up a ready-to-operate condition of the tool. This section is highly intensive in means of resources and time because of inaccuracies in simulation, machining and in the forecast of forces in forming processes. Based on this condition, the tool enables parts producers to create value added in manufacturing. Thereby the functionality of the tool has to be obtained by maintenance work. At the end of parts production, the tool is placed into stock, disposed of and/ or recycled.

Life Cycle

P st roce ru s ct sur e

Machining Manufacture

Dr illi ng

• Life cycle costing is usually long-term oriented. This results from the long time span of the life cycle of examined objects. As mentioned above, the calculation can be focused on products and potential factors as well as whole projects. The main characteristic of the life cycle costing method is its period-spanning vision. The main goal of life cycle costing is an extensive evaluation of all costs which occur during the life cycle of an examined object [16, 17]. Thus it may support decisions on alternative products and/ or services and their combination. The costs of tool usage are mostly determined by the tool’s quality and the general organisational conditions. These conditions can be improved, if toolmakers and their know-how are integrated into parts production processes. A life cycle costing oriented acquisition policy would help avoiding ruining price competition in the tool and die making sector.

An important factor throughout the life cycle of a tool is the change of responsibility from the toolmaker to the parts producer after optimization. This allocation of responsibility may be affected by realisation of new service offerings in service models, e.g. by the implementation of usage-dependent charges in financing. Throughout manufacture, all risks are carried by the toolmaker himself. From the date of a ready-to-operate condition of the tool and its acceptance, the responsibility for the tool normally devolves to the parts producer.

Material

• Liquidity calculation is used for a permanent controlling of the cash holdings of a company. It can be short-, middle- or long-term oriented and is not periodical.

Figure 6: Subdivision of the whole life cycle costs into cost elements. The developed life cycle costing model contains several ratios to draw overall conclusions from and to compare product-service-systems of toolmakers with one another. These ratios are calculated as net present value of all costs that occurred in processes throughout the entire life cycle. By combining different cost types, e.g. maintenance costs related to manufacturing costs, the quality of specific processes in each section of the life cycle can be evaluated. 5

APPROACHES TO PROVE THE ADDED VALUE OF SERVICE MODELS A main barrier for the successful implementation of product-service-systems into the market as well as into the toolmaker’s value proposition is the lack of availability of life cycle data, which is neither adequately gathered nor communicated. Firstly, the gathering of life cycle data allows tool and die makers to improve their own production and service processes and to reach an ideal

252

learning curve. Secondly, the life cycle data also delivers information about the point in time at which certain costs incurred. Besides a guarantee of availability for their tools, the allocated data and information put both toolmaker and parts manufacturer in the position to get to know the performances of the tool in operation better. In this manner, areas can be revealed in which service models are useful and would be generally accepted. The developed life cycle costing model shall help to close the existing gap of information between tool and die companies and part manufacturers. Based on an empirical analysis of life cycle data gathered from several tools in operation general statements can be made, that influence the implementation of service models in tool and die companies. The gathered data enables toolmakers to prove the value-added of their service offerings to their customers. The main restriction for the implementation of service models can be described by regarding the relation between the acquisition price and the life cycle costs of a product-service-system. The life cycle benefit describes the difference in total life cycle costs in comparison between different offerings (figure 7). Life cycle costs Manufacture

Optimization

Utilization

Life cycle benefit

Increase in price

Goal

Status-quo

Time

Figure 7: Life cycle benefit of service models. 6 CONCLUSION As best practices in tool and die making reveal, a promising approach for differentiation over competitors is to enhance the existing range of products by offering customer-specific services. To directly address the customer value in service provision, the creation of services should be based on a detailed life cycle costanalysis of tools and dies. Therefore customers will be integrated into the process of service design by means of an intense dialogue with the tool and die makers. A new approach to service models for the tooling industry links products and services by means of a direct calculation of life cycle costs. Tool and die makers are enabled to offer product-service-systems, which allow expanding the range of service provision by directly addressing the cost-effectiveness of the whole productservice-system. Hence the minimisation of life cycle costs of tools and dies can be used as a new sales pitch for services in this unique industry. A crucial factor of success is to cope with the interdependencies between the organizational business model and its implementation. In the research and ING development project “SMART STAMP ” these factors are considered by two main results. Firstly, the developed life cycle costing model allows calculating not only products but product-service-systems as a whole. Secondly, the strategic aspect of developing productservice-systems is considered by the development of

253

service models and their integration into former business model theory. Combining these methodical results, tool and die companies are able to close the information gap between them and their customers. Differentiation on other levels than price is thus enabled. Further work should concentrate not only on developing product-service-systems, costing models and business models but on developing a reference framework for cooperation between tool and die companies and their customers and suppliers. This cooperation is the main enabler for the life cycle-orientated optimization of products-service-systems in the tooling industry. 7 ACKNOWLEDGMENTS The research and development project “SMART ING STAMP ” is funded by the German Federal Ministry of Education and Research (BMBF) and managed by the Project Management Agency Forschungszentrum Karlsruhe, Production and Manufacturing Technologies Division (PTKA-PFT). 8 REFERENCES [1] Cleveland, M., 2002, A Competitive Assessment of the Die and Mould Building Sector, Grand Rapids, Mi.: www.mbs-2003.org, 2005-08-01. [2] Holmes, J., Rutherford, T., Fitzgibbon, S., 2005, Innovation in the Automotive Tool, Die and Mould Industry, in: Wolfe, D., Lucas, M. (eds.), 2005, Network Structure of an Industrial Cluster, Queens University, Toronto: 119-154. [3] Menezes, J., 2004, European Mould Making: Towards a New Competitive Positioning, th proceedings of the 4 International Colloquium Tool and Die Making for the Future, September 28-29, Aachen. [4] Harbour, R., 2000, The Conflicts Of E-Bidding, Automotive Industries, 2000/4. [5] Roy, R., 2008, Evaluating PSS Business Models for Machine Tool Industry, in: Proceedings of the Transregio 29 International Seminar on PSS, January 21-22, Bochum. [6] Akerlof, G. A.: The Market for "Lemons": Quality Uncertainty and the Market Mechanism. In: Quarterly Journal of Economics. Jg. 84, 1970, Nr. 3, S. 488-500. [7] Spence, A. M.: Job Market Signaling. In: Quarterly Journal of Economics. Jg. 87, 1973, Nr. 3, S. 355374. [8] Porter, M., 1998, Competitive Strategy, The Free Press, New York, NY. [9] Mueller-Stewens, G., Fontin, M., 2003, Die Innovation des Geschaeftsmodells – der unterschaetzte vierte Weg, Frankfurter Allgemeine Zeitung, 2003/172: 18. [10] Bieger, T., Rueegg-Stuerm, J., von Rohr, T., 2002, Strukturen und Ansaetze einer Gestaltung von Beziehungskonfigurationen: Das Konzept Geschaeftsmodell, in: Bieger, T. et al. (ed.), Zukuenftige Geschaeftsmodelle: Konzept und Anwendung in der Netzoekonomie. Springer: Berlin. [11] Zu Knyphausen-Aufsess, D., Meinhardt, Y., 2002, Revisting Strategy: Ein Ansatz zur Systematisierung von Geschaeftsmodellen, in: Bieger, T. et al. (ed.), Zukuenftige Geschaeftsmodelle: Konzept und Anwendung in der Netzoekonomie. Springer: Berlin. [12] Hamel, G., 2000, Leading the revolution, Harvard Business School Pr.: Boston, Ma.

[13] Müller-Stewens, G., Lechner, C., 2005, Strategisches Management, 3. Auflage, SchaefferPoeschel, Stuttgart. [14] Belz, C., Schuh, G., Groos, A., Reinecke, S., 1997, Industrie als Dienstleister, Thexis, St. Gallen. [15] Riezler, S., 1996, Lebenszyklusrechnung. Wiesbaden: Gabler. [16] Hawkins, R., 2004, Looking Beyond The Dot Com Bubble, in: Preisel, B., Bouwman, H., Steinfeld, C. (eds.), 2004, E-Life After The Dot Com Bust, Physica, Heidelberg. [17] Böning, M., 1997, Einsatzmöglichkeiten eines lebenszyklusorientierten Controlling von Produktionsanlagen, in: Hochschulschriften zur Betriebswirtschaftslehre, Vol.. 133, München: VVF. [18] Abele, E. et al., 2006, Beeinflussbarkeit von Lebenszykluskosten durch Wissensaustausch, in: wt Werkstattstechnik online, Vol. 7/8, S. 447-454. [19] N.N., 2004, Dependability management - Part 3-3: Application guide - Life cycle costing (IEC 60300-33: 2004), German version EN 60300-3-3: 2004.

254

Obsolescence Challenges for Product-Service Systems in Aerospace and Defence Industry F. J. Romero Rojo1, R. Roy1, E. Shehab1 and P. J. Wardle2 1 Decision Engineering Centre, Manufacturing Department, School of Applied Sciences, Building 50, Cranfield University, Cranfield, Bedford, MK43 0AL, UK. {f.romerorojo; r.roy; e.shehab}@cranfield.ac.uk 2 BAE Systems Integrated System Technologies, Eastwood House, Glebe Road, Chelmsford, CM1 1QW, UK. [email protected]

Abstract The aerospace and defence industries are moving towards new types of agreement such as availability contracts based on Product-Service System (PSS) business models. Obsolescence has become one of the main problems that will impact on many areas of the system during its life cycle. This paper presents the major challenges to managing obsolescence for availability contracts, identified by means of a comprehensive literature review and several interviews and forums with experts in obsolescence management. It is observed that there is a lack of understanding of the impact of obsolescence on whole life cost. Experts agree that the development of a framework to support estimation, management, and mitigation of these costs is desirable, but the difficulty in forecasting future obsolescence issues constrains industry to a reactive approach rather than proactive. Keywords: Obsolescence Management, Obsolescence Costing, Product Service Systems.

1 INTRODUCTION Traditionally in the aerospace and defence industry, an initial contract for development and manufacture was followed by a separate contract for spares and repairs. More recently, there has been a trend towards availability contracts where industry delivers a complete productservice system (PSS). The typical PSS has progressively increased in scale and complexity (e.g. from the humble photo-copier) through to major infrastructure projects (e.g. private finance initiative hospitals) to large defence projects (e.g. complete sea, air or land platforms) [1]. The challenge for both the solution provider and the customer is that, at the point of signing a contract, they must be confident in their estimates of the whole life costs (WLC) over periods of contracts that stretch 20, 30 or even 40 years into the future. This research is part of the PSS Whole-life Cost Project (PSS-Cost) which is carried out by Cranfield University in collaboration with UK industry in the defence and aerospace sector. The project aims to develop a framework for the estimation of WLC in availability contracts and affordability assessment at the bidding stage in the defence context. This could include the nonrecurring costs of developing, prototyping, integrating, and testing the product, the ongoing costs of maintaining the product (including obsolescence management), delivering and operating services (including staff training, commodities, consumables), and the end-of-life disposal of the product and/or termination of services. The exact composition of a WLC estimate will depend on the category of PSS contracts.

CIRP IPS2 Conference 2009

255

The obsolescence problem has a huge impact on the WLC of PSS in the defence and aerospace industry due to the long periods that the system is required to be in service [2]. Obsolescence may affect many aspects of a system such as: (See Figure 1) [3] •

Availability of replacement electronic components needed to replace those that fail in service



Failure of non-electronic components (e.g. typically mechanical or structural), frequently in unexpected ways [4] owing to wear, fatigue cracking, damage and corrosion as the PSS ages.



Software within the PSS (i.e. operational software including operating systems) and the software development environment needed to maintain it (i.e. hardware platforms, editors, compilers, linkers, loaders, and test rigs).



Documentation and data (e.g. in terms of content, data format, and ongoing availability of IT systems and toolsets needed to access and maintain it).



Procedures and methodologies.



Skills and knowledge.

Figure 1. The Holistic View of Obsolescence (at the System Level) (Adopted from [3]) 2 RELATED RESEARCH In the literature it can be appreciated that many attempts have been made to manage obsolescence [2, 5, 6]. However, the challenge is still to develop a methodology that embraces obsolescence forecasting and the costing of all the possible alternatives to resolve it, delivering the optimal planning for managing the obsolescence problem [5, 7]. The first step is to minimise the impact of obsolescence at component level by development of a methodology able to determine the best dates for any design refresh and the optimum combination of actions required. For example, for electronic components, this has been addressed by Singh, Sandborn and Feldman [2, 8] at the University of Maryland by developing the Mitigation of Obsolescence Cost Analysis (MOCA) tool. All the research described in the literature and carried out in order to minimise the impact of component obsolescence makes an attempt to determine: [9] •

How to anticipate obsolescence;

occurrences

of

component



How to react obsolescence;

occurrences

of

component



to

How to reduce the risks of future component obsolescence; Collaboration within the industry [10]; standardisation [11] of designs and modularisation [12] that may promote the interchangeability of components; and the implementation of proactive actions to determine accurately the cost and impact of obsolescence are the major means of minimising obsolescence risks [6]. Most of the research done so far in obsolescence has been focused on electronics components. Very few studies have considered a holistic approach taking into

account the effects of obsolescence on mechanical components, software, skills of the personnel and processes. As Dowling [3] highlights, although there are tools and techniques developed for dealing with components obsolescence, “there is no defined process in MoD or elsewhere for managing system obsolescence”. Therefore, it is considered that a holistic study of the obsolescence topic will allow the overall impact on a PSS to be determined across the whole life cycle. This research will focus on the development of a framework that allows forecasting and measuring the impact of obsolescence at the system level on cost. It will promote the use of proactive strategies in order to minimise the impact of obsolescence. 3

CONTRACTING MANAGEMENT

FOR

OBSOLESCENCE

3.1 Case Study The continuous evolution of contracting in the defence procurement in the UK, which has been motivated by the MoD’s aim to deliver military capability at optimised cost of ownership, is bringing with it new challenges for ensuring both the affordability of military operations and the profitability of suppliers. Acquisition strategies now include a range of initiatives including spares inclusive, availability based contracting and ultimately, contracting for capability (Figure 2). These system-support contracting strategies can range from the provision of traditional fourth line repair and overhaul to usage based service level agreements. This range gives evidence of the recent expansion in the strategic degrees of freedom available to organisations operating in the defence sector.

256

Figure 2. The Evolution of Contracting in the Defence Sector However this business evolution brings with it the potential of increased operational risks for military customers, and issues arising from the commitment to future expenditure over long period of time. There will be increasing trend towards contracting for availability. Nowadays it is regarded as a challenge to be able to cost availability as it is radically different from costing a solution, as it has been done so far. The current trend of contracting is moving in that direction and contracting for availability is the latest stage reached. The essence of availability contracts is that the suppliers are paid for achieving an availability target for the PSS (typically expressed as a percentage, e.g. “available 99.95% of the time”) and not just for the delivery of the product and spares/repairs. This helps to ensure value for money for the customer. The process currently adopted for acquiring equipment, systems or services for the UK armed forces is known as the CADMID cycle (Figure 3). This has six phases: concept; assessment; demonstration; manufacture; inservice, and disposal – all of different lengths, of course, with formal approvals at Initial Gate and Main Gate.

One of the biggest challenges of contracting for availability is costing obsolescence. It can affect not only the in-service phase of the CADMID cycle but any other phase because of the long duration of each stage. At the bid stage there is little information that can form the basis of the obsolescence forecast and the cost estimation related to it. This becomes a big risk for both the customer and the supplier. It is critical to agree in the contract the allocation of responsibilities for managing obsolescence, resolving obsolescence issues and defraying the cost of them. The resolution strategy for each obsolescence issue is decided by the supplier’s project manager. The strategy may also have to be agreed with the customer but this depends on the contract; the allocation of responsibilities between the supplier and the customer varies. The most common strategies are described as follows: (1) The customer is responsible for the cost of resolving any obsolescence issue while the contractor is in charge of managing and resolving it. This has been the traditional way of contracting in the military sector. Customers would like to move away from this contracting style because, from their point of view,

Figure 3. The CADMID Cycle 257

the supplier is not encouraged to find the most costeffective resolution strategy. (2) The supplier is responsible for the management and cost of resolving any obsolescence issue. Some availability contracts are implementing this strategy in order to agree a fix price. In principle the solution will be cheaper than option (1) because the supplier is better placed to manage the issues. However, the risk has transferred from customer to supplier and the supplier price will be driven up to cover the risk budget. (3) Contractor pays for any form fit and function (FFF) replacement while the customer pays for any other obsolescence resolution. (4) Financial threshold. A cost limit is set and the contractor will cover the costs related to solving obsolescence issues up to that limit. From that point onwards, the customer will be in charge of covering the costs. (5) Target cost incentive fee. A target cost is set and if the final cost is lower than it, the contractor will receive a percentage of the cost avoided. This encourages the contractor to manage obsolescence in the most cost effective way. (6) The supplier is responsible for the management and resolution of any obsolescence issue and the cost related to it is shared between the supplier and the customer. All resolution costs are split by a percentage factor between the customer and the contractor (e.g. 70/30, 50/50, 60/40). This is regarded as the best solution as it provides incentives to the supplier to search for the most cost-effective resolution strategy. It aligns the interests of both parties. The resolution strategies are defined by the supplier and approved by the customer. The development of fair clauses for both parties is necessary, taking into account that experts spoken to as part of this research say that it is difficult to estimate the cost of obsolescence for more than eight years ahead. This figure is corroborated in the literature [13] and is based on the huge number of factors that may influence the forecasting of an obsolescence occurrence. 3.2 Approach Adopted The current practice in the aerospace and defence sector has been captured across the Ministry of Defence in UK and the defence aerospace industry. A total of 27 interviewees from the above organisation have participated in workshops or one-to-one interviews. They are mainly Project Managers, Project Engineers and experts in Obsolescence Management, with experience in the area ranging from 3 to 20 years. Another source of information is the documentation provided by these organisations, such as logistics and support plans, obsolescence management plans and examples of cost models. Once the data gathering and analysis were completed, a cross case synthesis was performed by means of an internal workshop involving the members of the research team and compared with the observations from the literature review. The results were validated by Rolls Royce and approved by the other collaborators. 4 OBSOLESCENCE MANAGEMENT The obsolescence problem cannot be avoided [14, 15]. The only way to minimise the impact of obsolescence and mitigate the risk is by planning and managing our response. Most of the organisations covered by the Case Study are using an Obsolescence Management Plan

(OMP) that defines the policy to deal with obsolescence for each specific project. The OMP is developed by the department/expert in obsolescence management in each organisation. The OMP typically calls for a two-stage response: first, to identify obsolescence risks where economically viable; second, to mitigate the impact of residual risks should they arise. 4.1 Risk Assessment This risk assessment is generally based on experience and expert opinion and some organisations have developed formulae on this basis (e.g. extrapolation of experience on past spares-and-repairs contracts to future availability contracts). Based on the results of this assessment it is decided which components (the critical ones) will be monitored in order to proactively tackle any possible obsolescence issue and which components (non-critical ones) can be managed in a reactive way. •

At hardware component level many organisations tend to combine the use of commercial monitoring tools, such as Q-StarTM [17] or the IHS products [18], with manual monitoring or in-house developed tools. This supports the determination of which components are most critical (e.g. on the Pareto principle).



At higher levels of integration including both hardware and software, technology roadmapping is widely used to inform expert opinion. This particularly applies to commercial-off-the-shelf (COTS) product lines, e.g. where a supplier may have a three to five year plan for future development.

4.2 Impact Mitigation In principle, proactive mitigations seek to sustain the FFF replacement of failed system elements over the planned lifetime of the system in the most economic way. The approaches to proactive mitigation identified in case studies include: •

Life-time buys for components which are singlesourced, for materials whose continued supply is at risk, or where commercial factors may come in to play (e.g. a key supplier is at risk of being bought-out by the supplier’s competitor). An understanding of the rate of consumption is necessary in order to establish the quantity of the life-time buy.



Multiple sourcing, e.g. the design of a system should make the best possible use of “industry standard” or “commodity” components, materials, and COTS so that FFF replacements are readily available.



Partnering agreements with suppliers to assure continued availability for single source components, materials or COTS (e.g. subsidies for maintaining production capabilities even if used intermittently). In the limit, the overall budget for proactive mitigations is established as a system-level trade-off between the level of assured availability desired by the customer and the price the customer is prepared to pay. Reactive mitigations are used where the likelihood of obsolescence is believed to be low and / or the cost of proactive mitigation is not economically viable. The benefit of this approach is that costs are only incurred if obsolescence problems materialise. The disadvantages are that it is not possible to achieve an assured level of availability, and the customer and supplier must agree how the mitigation costs are covered if obsolescence does materialise. Reactive mitigation generally involves a degree of redesign at component, assembly, sub-system, or system level. Costs include: -

258



Development engineering costs including re-design, prototypes, integration, verification and validation, safety cases analysis, and re-certification against regulatory requirements.



Production engineering, (re)manufacture, production and fitting of modification kits in the field, updates to documentation and training.

5

CONCLUSIONS

5.1 Key Findings Several challenges in costing obsolescence have been identified in this paper. The challenge of predicting obsolescence more than eight years ahead has been covered in Section (3). Secondly, there is a general lack of standard procedures in the defence industry for the cost estimation of obsolescence. Most of the organisations in the defence sector are estimating this cost at the bidding stage based on experience and expert judgement. This rough estimate is in general inadequate to set the basis for the negotiation of the contract. Thirdly, at the bidding stage it is difficult to predict what obsolescence issues will arise, at what rate, and which resolution strategies will be viable in the future. All of these issues constitute risks and uncertainties in estimating the cost of obsolescence. Finally, the organisations studied in this research used a range of techniques for assessing the risk of obsolescence including prior experience, commercial tools, technology roadmapping but they struggle to combine this information with the understanding of the “health” of their suppliers (supplier assessment carried out by the procurement, commercial and engineering functions), regulation changes (in UK, EC, USA) and market trends (carried out by the sales & marketing and commercial functions) in order to forecast obsolescence events accurately. This is necessary to plan ahead the mitigation strategies that should be put in place. 5.2 Future Research Opportunities The decision between a proactive vs. reactive approach to managing a given obsolescence risk depends on two main factors: •

The customers’ attitude to risk. Do they consider it essential to be assured of a given level of system availability (e.g. in safety critical situations), or are they prepared to accept a certain level of unplanned outages? How can trade-offs between availability and affordability (for the customer), or between availability and profitability (for the supplier), best be modelled and visualised over the whole life of a PSS?



The cost of proactive vs. reactive mitigations. Is it possible to devise some generic guidance, e.g. this was studied by the MoD in 2004 [16] and is being pursued in collaboration with Cranfield University? Further study on each of these factors is necessary to improve obsolescence managing decisions. Furthermore, a need for a framework for the cost estimation of system obsolescence at the bidding stage has been identified. It will allow the use of a flexible process for cost estimation of obsolescence and further integration with the rest of life-cycle costs, taking into account the influence of factors such as the maintenance strategy.

259

6 ACKNOWLEDGMENTS This project was funded by the EPSRC/Cranfield IMRC and the industrial collaborators: BAE Systems (Insyte), Lockheed Martin (Insys), GE Aviation, UK Ministry of Defence (MoD), Rolls-Royce, Association of Proposal Management Professionals (APMP), Society of British Aerospace Companies (SBAC), Galorath International and Cognition Europe. Special thanks to the first three industrial collaborators aforementioned for contributing actively, sharing information and data. 7 REFERENCES [1] Shehab, E. M and Roy, R. (2006), “Product servicesystems: issues and challenges”. Fourth International Conference on Manufacturing research (ICMR 2006). John Moores University, Liverpool, 5th-7th September. [2] Feldman, K. and Sandborn, P. (2007), "Integrating Technology Obsolescence Considerations into Product Design Planning", Proceedings of the ASME 2007 International Design Engineering Conferences & Computers and Information in Engineering Conference, Sept. 2007, Las Vegas, NV. [3] Dowling, T. (2004), "Technology insertion and obsolescence", Journal of Defence Science, vol. 9, no. 3, pp. 151 - 155. [4] Howard, M. A. (2002), "Component obsolescence– It’s not just for electronics anymore", Proc.FAA/DoD/NASA Aging Aircraft Conference, pp. 16–19. [5] Meyer, A., Pretorius, L. and Pretorius, J. H. C. (2003), "A management approach to component obsolescence in the military electronic support environment", South African Journal of Industrial Engineering, vol. 14, no. 2, pp. 121-136. [6] Meyer, A., Pretorius, L., Pretorius, J. H. C., Solutions, O. M. and Africa, S. (2004), "A model to manage electronic component obsolescence for complex or long life systems", Engineering Management Conference, 2004.Proceedings IEEE International, vol. 3, pp. 1303 - 1309. [7] Hitt, E. F. and Schmidt, J. (1998), "Technology obsolescence (TO) impact on future costs", Digital Avionics Systems Conference, 1998.Proceedings., 17th DASC.The AIAA/IEEE/SAE, vol. 1, no. A33, pp. 1-7. [8] Singh, P. and Sandborn, P. (2006), "Obsolescence Driven Design Refresh Planning for SustainmentDominated Systems", The Engineering Economist, vol. 51, no. 2, pp. 115-139. [9] Condra, L. (1999), "Combating Electronic Component Obsolescence by Using Common Processes for Defense and Commercial Aerospace Electronics", IECQ-CMC Avionics Working Group1, NDIA Paper document, September. [10] Torresen, J. and Lovland, T. A. (2007), "Parts Obsolescence Challenges for the Electronics Industry", Design and Diagnostics of Electronic Circuits and Systems, 2007.DDECS'07.IEEE, pp. 14. [11] Behbahani, A. R. (2006), "Achieving AFRL Universal FADEC Vision with Open Architecture Addressing Capability and Obsolescence for Military and Commercial Applications". [12] Tomczykowski, W. J. (2003), "A study on component obsolescence mitigation strategies and their impact on R&M", Proceedings of Annual Conference on Reliability and Maintainability (RAMS), 27-30 Jan. 2003, IEEE, Tampa, FL, USA, pp. 332-8.

[13] Smith, T. (2000), "Future Initiatives for Obsolescence Mitigation Strategies", RTO SCI Symposium on Strategies to Mitigate Obsolescence in Defense Systems Using Commercial Components, 23-25 October 2000, Budapest, Hungary, Defence Evaluation and Research Agency Kent (UK), pp. 30/1-30/13. [14] Sjoberg, E. S. and Harkness, L. L. (1996), "Integrated circuit obsolescence and its impact on technology insertion definition for military avionics systems", Aerospace and Electronics Conference, 1996. NAECON, Proceedings of the IEEE National, vol. 2, pp. 792 - 799.

[15] Sandborn, P. (2007), "Designing for Technology Obsolescence Management", Proceedings of the 2007 Industrial Engineering Research Conference, May 2007, Nashville, TN. [16] MoD (March 2004), Ministry of Defence Component Obsolescence Resolution Cost Metrics Study, QinetiQ and ARINC, UK. (http://www.nocweb.org, accessed on 5-Sep-2008) [17] QinetiQ Technology Corporation (http://www.qtec.us, accessed on 8-Nov-2008). [18] IHS Inc (http://www.ihs.com accessed on 8-Nov2008)

260

Sensitivity Cost-Benefit Analysis to Support Knowledge Capture of Industrial Interests A.M. Paci, M.S. Chiacchio Istituto di Tecnologie Industriali e Automazione, Consiglio Nazionale delle Ricerche, Via dei Taurini 19, Rome, 00185, Italy [email protected]

Abstract In strategic industrial fields, emerging technologies are evolving into strategic enabling technologies for next generation products and services (IPS2). To this end a new approach to market knowledge capture needs early impact assessment of costs and value benefits considering the broad variety of factors of sustainability in medium-long term. This paper presents the framework for new IPS2 and a new methodology for Sensitivity Cost-Benefit Analysis (SCBA). SCBA aims to evaluate new IPS2, based on RTD potential, applying the AHP (Analytic Hierarchy Process) technique. The evaluation results support stakeholders in market knowledge capture about new IPS2 based on new enabling technologies. Keywords: Cost-Benefit, Impact assessment, Production paradigms

1 INTRODUCTION In strategic industrial fields of major European and regional interests, emerging technologies are evolving fast into strategic enabling technologies for the conception and development of next generation products and services. Important initiatives are being undertaken at European and national level to coordinate public and private stakeholders. Multilevel studies promote the diffusion of these emerging pacing technologies for innovative industrial product-service systems (IPS2) and related business models. In this context, the production and manufacturing of new products and services shall refer to market-oriented scenarios meeting and mutually reinforcing competitiveness and sustainability [1]. A new strategic technology represents a possible driver of change, for next generation of products and services to meet the challenge of sustainability, which is composed by economic, environmental, social and technological dimensions [2]. In the past, according to O.K. Mont, the concept of product-service systems included ‘dematerialisation’ as part of the today IPS2 [3]. Looking to new IPS2, ‘dematerialisation’ shall include pacing technologies to shape the ‘entire process’ for next generation products and services as a new framework for IPS2. Today’s IPS2 seek for efficiencies – such as: flexibility, mass-customisation, quality, high added value and cost reduction currently – referred as the new industrial production response paradigm meeting Vision and Strategic Research Agendas for new production [4] [5]. This paper presents the framework for scenarios of new IPS2 together with a new methodology for Cost-Benefit Analysis [6] – named SCBA (Sensitivity Cost-Benefit Analysis). The SCBA aims to evaluate the potential IPS2 in medium-long term market scenarios. The novel framework considers that pacing technologies provide high value features of next generation of IPS2 to

CIRP IPS2 Conference 2009

261

compete in existing markets with existing products and services. In the K-economy, the industrial leadership in mediumlong term of a pacing technology becomes a driver of change. The market exploitation needs knowledge capture for the analysis of prospective IPS2 ideas. The SCBA supports industrial innovation projects by market knowledge capture within scenarios development for an entire IPS2. It focuses on: •

the specific pacing technology potential of IPS2;



the early impact assessment in today markets [7];



time to market for rapid implementation in new markets;



all sustainability related aspects (environmental, social, technological, and economical). The scenarios, built through knowledge management methods and tools [8], involve all stakeholders (producers and buyers) in the design and development of next generation IPS2 and the technology life cycle. The impact assessment outcomes support stakeholders in decision-making and business development of next generation of IPS2, with an overview of market pull and sustainability issues, drivers for change and new relationship among high technologies, producersconsumers, suppliers, regulations and society. 2 INDUSTRY INITIATIVES The European Technology Platforms with the related Joint Technological Initiatives – the most important industry-led initiatives based on R&D priorities for technology sustainable development – cover strategic fields of broad and regional interests. Among these strategic industrial fields, the satellite industry has become a very important sector with a significant impact on the technological, social, economic and environmental aspects. Satellite communications have brought many benefits to society and citizens, in Europe and worldwide.

The space industry makes a vital contribution to the renewed Lisbon agenda for ‘jobs and growth’ and to the i2010 strategy for the European Information Society. Satellite can help build new markets – i.e. audiovisual and media new markets – and applications. Digital TV for example has already been broadcast over pan-European satellite systems for several years, with hundreds of digital TV programmes provided to European consumers. This sector contributes also to the ‘energy and climate change’ and ‘social welfare’ Lisbon objectives with initiatives, such as the wide footprints of satellites that help humanitarian organisations to respond to emergencies or disasters, wherever they occur. Satellite coverage may also be the only way to provide broadband connectivity to very remote areas, in the EU or globally, as shown by the recently announced EU Strategy for Africa. In this context the role of the European Space Agency has developed the strong vision for the space sector. Particularly, Galileo and GMES initiatives demonstrate the multilevel commitment to the space industry [9] [10] [11]. ISI (Integral Satcom Initiative) European Technology Platform [12] was launched in 2006 and was established to bring together for the first time in a unified, industry-led forum all the research and technology aspects related to satellite communications, including mobile, broadband, and broadcasting applications. The purpose is to foster and develop the entire industrial sector, maximise the value of European research and technology development, and contribute to EU and ESA policies [13]. Studies are undertaken to analyse the introduction of new IPS2 and related business models into markets and to assess their impact. Among the impact assessment studies of new technologies for manufacturing roadmaps, aerospace industry demand has been analysed as one of manufacturing sectors examined, in terms of R&D needs within the FP6 Leadership SSA project [14] [15]. This study showed that aerospace sector’s demand for new industrial response paradigms and new business models, together with the need to communicate everywhere and within a range of very different contexts. In recent years, many studies, at European, national and local levels, have been carried out with particular reference to the use of the Galileo system. These studies aim to contribute to the development of value-added services and applications to fully exploit all possible potential of this system into markets. This development of services and applications aims to fulfil market expectations with the development of next generation IPS2 featuring high value services. In particular, the following ones are the four Galileo services: • The Open Service (OS) that provides position and timing performance; it is competitive with other GNSS systems. • The Safety-of-Life Service (SoL) that improves the open service performance through the provision of timely warnings to the user when it fails to meet certain margins of accuracy (integrity). • The Commercial Service (CS) that provides access to two additional signals, to allow for a higher data throughput rate and to enable users to improve accuracy. • The Public Regulated Service (PRS) that provides position and timing to specific users requiring a high continuity of service, with controlled access. This new scenario for the space sector requires innovative approaches to fully exploiting the high-value features of satellites, considering not only technological or economic/financial aspects, but also social and

environmental issues. To this purpose the approach – proposed in the paper – considers a wide range of criteria to analyse the full potential of the new enabling technology. This paper reports on the methodology – developed and applied in an evaluation study – for Galileo system to become the enabling technology for new IPS2 of the space sector. 3 THE SCBA FRAMEWORK The theoretical framework, presented in this paper, aims to propose a new methodology, the SCBA, to evaluate the specific pacing technology that plays a strategic role in the value innovation process for next generation IPS2. The SCBA framework for IPS2 considers that pacing technologies, by replacing key technologies, provide new features of the next generation of IPS2. This framework refers to scenarios development that shows how to move from a starting situation of Leadership in technology towards the development of a new market for successful IPS2 (Figure 1). In order to support the turning of pacing technologies into business, the SCBA enables supporting the evaluation of the potential IPS2 and market knowledge capture – by solving two special problems: • potential competition with similar existing systems (existing markets, existing products, existing data,…); • medium-long term horizon of the development of the next generation of IPS2 and its related difficulties (qualitative data, non-measurable data, strategic pacing technologies). Previous methods to evaluate new product development for IPS2 have been oriented to analyse advanced technology product development and related introduction processes, in order to enhance a company’s competitive advantage. The development in this field includes also the buyer-supplier perspective. The SCBA considers the buyer-supplier perspective and enhances the involvement of all stakeholders (producers and consumers) right from the start – the design phase – of the development of next generation IPS2. In addition, to help organizations make better decisions for the successful next generation of IPS2, the evaluation analysis requires consideration of whole market, integrating competitiveness perspective with the dimensions of sustainability. Sustainability refers to broad variety of criteria. Being part of the analysis, suitable and manageable groups of criteria allow activating pair wise comparison in accordance with an internal principle. With this aim, the SCBA presents a new approach to the traditional Cost-Benefit Analysis (CBA), adding new factors that are less assessable than economic benefits due to the medium-long term horizon of RTD market potential. The new methodology proposes a hierarchical (or treebased) structure of factors constituted by branches enabling the ranking and prioritization of the wide range of criteria at each level of the tree. This structure enables to frame the potential of the pacing technology by analysing new relations among branches to achieve an overall view of sustainability and ranking factors. The SCBA framework, presented in this paper, identifies the AHP (Analytic Hierarchy Process) technique with sensitivity analysis as a suitable method for suitable investigation.

262

Figure 1: Scenarios for new IPS2 framework [4]. The main argument for the choice of AHP – compared to other multi-criteria decision-making techniques, including the multiple regression and multi-attribute utility approach as reported in the literature [16] [17] – lies in: • its applicability to non-measurable criteria; • more detailed and mixed information on pair wise comparison. Its successful application to different decision-making problems – as reported in the literature [18] – and its appropriateness for the framework reported here are the reasons for adopting the AHP technique. The above two features constitute the advantage of this technique’s application regarding pacing technology development in next generation IPS2. In addition, the sensitivity analysis is incorporated into the evaluation methodology to test the stability of the priority ranking. In presence of market changes, stakeholders need the stability of a decision for the development and launch of next generation of IPS2 into new markets. 4 THE NEW APPROACH The shorter life cycle of technologies and the fast introduction of new technologies into IPS2 require a new approach to the evaluation of IPS2 to support stakeholders in the investment decision. In the traditional approach, Cost-Benefit Analysis (CBA) focuses on financial and economic aspects for analyzing prospective IPS2 ideas within an industrial innovation process. It accounts for all (negative and positive) effects of policy measures, allows comparison of the costs with the benefits of the proposal over time and can also be used to rank alternatives in order of their net social gains or losses. But this approach has some disadvantages: • It cannot include impacts for which there exist no quantitative or monetary data. • It presents difficulties in establishing the social discount rate. • Usually it is more expensive and time-consuming than other, less broad, methods. • It may lead to distributional issues being overlooked. The SCBA approach presents new principles of CBA including the assessment of mixed (qualitative and quantitative) evaluation. This new approach responds to the need to evaluate pacing enabling technologies for potential next generation IPS2. This evaluation is based

263

on market knowledge capture, introducing the competitive sustainable perspective and focusing the new ‘science to market’ relationship. Market knowledge capturing for IPS2 regards coexistence or substitution of key with pacing technologies, to respond to competitive needs or to develop new markets looking forward to sustainability issues. This approach evaluates the early impact of other aspects beyond the traditional economic and financial ones. It enables a costs and value benefits analysis for future market scenarios by integrating all production and consumption aspects. The concept of value benefits in the K-economy reinforces the view on present competition and enables future looking industrial innovation strategy. Recently, the importance of other factors, such as socio-political and environmental issues, has been also introduced at EU level, as mutually reinforcing the economic and financial decisions for market development studies. Using appropriate decision-making techniques, this approach allows to build medium-long term assessment of pacing enabling technologies and expected next generation of IPS2: • assessing impacts for which quantitative or monetary data do not exist; • looking at a more relaxed approach towards benefits measurement; • comparing alternatives between IPS that have more or less the same outcome and great value in use; • exploring value benefits of important alternatives with the sensitivity analysis to assess the worth of decision implementation. 4.1 New factors The SCBA approach presents new factors to evaluate the impact of the pacing enabling technology in terms of costs and value benefits and of new relationship among high technologies, producers-consumers, suppliers, regulations and society. The mainly profit seeking factors of competitiveness, just as economic and financial factors, are assessed together with social, political and environmental aspects responding to the Lisbon strategy objectives. The technological factor could be also considered as a key enabler, in particular in the context of strategic industrial fields, such as the space industry. With this aim, the impact of pacing technologies is assessed across four policy dimensions (economic,

social, environmental and technological) that created the competitive sustainable manufacturing scenario. In this context the costs and value benefits are evaluated by distinguishing between general and operational objectives. General ones attend to meet the overall goals of a strategy with global indicators that assess the outcome at a policy level. The evaluation in terms of general objectives is carried on at four dimensions levels. The operational objectives are expressed in terms of outputs, goods and services that the intervention should produce at management level. The evaluation in terms of operational objectives is carried on at sub-levels. 5 THE SCBA Considering all the dimensions of competitive sustainable perspective and the turbulent market context of strategic industrial fields, the proposed theoretical framework requires a multi-criteria technique to support stakeholders in the innovation decision-making process, providing stable inputs for decisions. With this aim, this paper presents a new methodology for Cost-Benefit Analysis – named SCBA (Sensitivity CostBenefit Analysis) – for the evaluation of potential IPS2 in medium-long term market scenarios. The SCBA integrates sensitivity analysis with new principles of CBA. It supports stakeholders in evaluating costs and value benefits of pacing strategic enabling technologies by comparing them with a family of similar key technologies. The sensitivity analysis explores how the strategic and management decision of stakeholders changes in relation to variations in key parameters of existing technologies and in interactions. This technique supports industrial interests in the identification of decisions about value benefits, in order to make the option worth undertaking. SCBA analyses industrial interests for: • Market competitiveness, making a comparison with existing products. The market demand for new features of next generation IPS2 is required for market success and not yet fully exploited by the existing IPS2. • Targeted sustainability impact, which includes social and environmental requirements. The sensitivity analysis shows the outcome of the course of action in a medium-long time horizon. 5.1 Methods and tools Methods and tools for the SCBA are structured to provide inputs coupled with knowledge management tools such as a structured survey made of consultation meetings with stakeholders and questionnaires. In the theoretical framework presented here, a multicriteria method is applied to evaluate the costs and value benefit of pacing enabling technology for next generation of IPS2, considering simultaneously several dimensions of competitive sustainable scenario. The multi-criteria method covers a wide range of techniques that share the aim of combining a range of positive and negative impacts in a single framework to allow easier comparison of scenarios and decisionmaking. This method could be applied in order to consider a large amount of information on a number of different impacts and on different formats. It allows having a mixture of qualitative and quantitative information and of varying degrees of certainty. Its applicability in the examined context presents the following advantages: • simultaneous consideration of the multi-dimensionality of both competition and sustainability;

• evaluation and comparison of different types of data (quantitative and qualitative) in the same framework with varying degrees of certainty; • transparent presentation of the key issues. The theoretical framework, presented in this paper, identifies the AHP (Analytic Hierarchy Process) technique with sensitivity analysis as a suitable method for suitable investigation. As reported in the references, the literature has compared several commonly used multi-criteria techniques. It shows that AHP, multiple regression and multi-attribute utility approach techniques produce similar results, but each one has advantages over the others. The advantages of AHP, represented by the detailed information produced and its applicability to non-measurable criteria, are the reasons for adopting the AHP method in this theoretical framework. The AHP (Analytic Hierarchy Process) is a simple, mathematically-based, multi-criteria decision-making method that allows the presentation of results as a mix of measurable and qualitative criteria. The Analytic Hierarchy Process (AHP) is a structured technique for helping people deal with complex decisions. Based on mathematics and human psychology, it was developed by Thomas L. Saaty in the 1970s and has been extensively studied and refined since then. In the literature, AHP provides a comprehensive and rational framework for structuring a problem, for representing and quantifying its elements, for relating those elements to overall goals, and for evaluating alternative solutions. The AHP converts these evaluations to numerical values that can be processed and compared over the entire range of the problem. A numerical weight or priority is derived for each element of the hierarchy, allowing diverse and often incommensurable elements to be compared to one another in a rational and consistent way. This capability distinguishes the AHP from other decision-making techniques. Although AHP has been the subject of many research papers and the general consensus is that the technique is both technically valid and practically useful, there are critics of the method. In the following, the SCBA methodology is applied to evaluate costs and value benefits focusing the pacing technologies that enable new directions for the European satellite industry. The study is ongoing and the evaluation analysis cannot be disseminated. With the aim of market knowledge capture, knowledge management for innovation has integrated the evaluation analysis with: • consultation meetings with stakeholders (producers and buyers); • questionnaire to survey market key opinion leaders in terms of expectations both due to the present inefficiency of existing IPS2 and related enabling technologies and as future needs in a strategic perspective regarding the new factors change, • pre-structure data and guidelines to target innovation and structured to lead potential consumers to express their needs. At an operative level, in accordance to the application of AHP technique [19], the procedure for processing the stakeholder inputs starts from the building of a tree-based structure and provides the ranking of alternatives; at the end results are tested with sensitivity analysis: Phase 1. Build the appropriate hierarchical structure (Figure 2).

264

Figure 2: Costs and value benefits’ hierarchy structure.

D1 D 2 D 3 D 4 1. Goal: Analyse costs and value benefits of technologies within a competitive sustainable perspective. 2. Criteria. The multiple ranges of criteria are put in order in suitable and manageable groups. • At level 1, there are represented four dimensions (D) of a competitive sustainable perspective: Economic, Social, Environmental and Technological. • From level 2 to level N, the range of criteria (C) is grouped according to the proper competitive sustainable dimension. For example, under Technological dimension there are criteria of: operational aspects, technical requirements, state of the art,.. 3. Alternatives (A). The evaluated items are: Pacing strategic enabling technologies vs. Key similar technologies. Phase 2. Establish priorities and ranking at each level of the hierarchy structure. This analysis is based on stakeholders inputs which are captured through knowledge management tools (questionnaires, consultation,…) and interpreted using the nine-point scale (Figure 3). 1. Determine the importance of each Dm (with m=1,2,3,4) competitive sustainable dimensions with respect to achieving the overall objective. The four dimensions are compared pair wisely, using a nine-point scale. The comparison matrix W1 can be formed to represent the pair wise comparison of four dimensions. The matrix element wij = wi / wj

(1)

represents the weights of dimension Di respect to dimension Dj. The consistency index and consistency ratio need to be checked. At the end of this comparison the local ranking (that in the first level is also global ranking) of dimensions is built.

D1 ⎡ w1 ⎢ w1 ⎢ D 2 ⎢ w2 ⎢ w1 W1 = ⎢ w3 D3 ⎢ w1 ⎢ w4 D 4 ⎢⎣ w1

w1 w3 w2 w3 w3 w3 w4 w3

w1 ⎤ w4 ⎥ w2 ⎥ ⎥ w4 ⎥ w3 ⎥ w4 ⎥ w4 ⎥ ⎥ w4 ⎦

2. Determine the importance of each criterion Cn(m)at each level (from 2 to N) with respect to its upper-level dimension using the nine-point scale. At level 2, four different matrices are built, each one for the m dimensions. The consistency indexes and consistency ratios need to be checked. At the end of this comparison the local priorities for each group of n criteria at second level are constructed. Multiplying the local weight by the global weight of the upper level the global ranking is also obtained.

C1( m ) C 2 ( m ) K C1( m ) ⎡ w1( m ) ⎢ w1( m ) C 2 ( m ) ⎢ w2 ( m ) ⎢ W 1m = ⎢ w1( m ) K ⎢ K ⎢ wn ( m ) ⎢ Cn ( m ) ⎣ w1( m )

w1( m ) w2 ( m ) w2 ( m ) w2 ( m ) K wn ( m ) w2 ( m )

Cn ( m )

w1( m ) ⎤ wn ( m ) ⎥ w2 ( m ) ⎥⎥ K wn ( m ) ⎥ K K ⎥ wn ( m ) ⎥ K wn ( m ) ⎥⎦ K

for each m

3. Obtain the priorities of alternatives with respect to each of the criteria. Then synthesizing the results of steps a) and b), multiplying the local weight by the respectively global weight the global ranking of alternatives is obtained.

A1( n ) A1( n ) ⎡ w1( n ) ⎢ w1( n ) ⎢ w1( n ) A2( n ) ⎢⎢ ⎣ w1( n ) 265

w1 w2 w2 w2 w3 w2 w4 w2

A2 ( n ) w1( n ) ⎤ w1( n ) ⎥ w1( n ) ⎥⎥ w1( n ) ⎥⎦

for each n

Figure 3: The fundamental scale for Pair wise Comparisons based on Scale of Saaty Phase 3. Perform the sensitivity analysis. At the end of this process the sensitivity analysis is performed to test the stability of the priority ranking of alternatives in relation to variations in key parameters. At the end of this procedure, the ranking of alternatives and criteria are obtained and an overall – but also detailed – view of sustainability and ranking factors is achieved. The most beneficial for innovation is the creation of new knowledge that supports stakeholders in the innovation decision-making processes, screening the multidimensions of sustainability as well as the issues of competitiveness. By better analyzing the achieved results at each level, it is possible to capture market inputs for the development of next generation IPS2 and the related benefits in a short, medium and long term perspective. In addition, by analyzing the inputs of several classes of stakeholders, it is possible to define classes of potential users and support the organization on segmentation of market and selection of market clusters. 6 SUMMARY This paper proposes the SCBA methodology for the assessment of costs and value benefits of emerging technologies and for market knowledge capture of new IPS2. This methodology has been produced by the Laboratory of Emerging Production Paradigms (EPPLab) of ITIA-CNR of the Department of Production Systems of the Italian CNR [20] within its strategic research project. The main author of this paper is the head of EPPLab, Dr. Paci who wrote paragraphs 1, 2, 3, 4, 6. Eng. Chiacchio wrote paragraphs 5 within her PhD study on New Production Impact assessment in Economic and Management Engineering at the University of Rome “Tor Vergata”. The proposed SCBA framework deals with emerging pacing technologies – as a driver of change – and their time to market diffusion. It considers emerging technologies as the enabler of high value features of next generation of IPS2 to meet competitive and sustainable issues. SCBA frames new principles of Cost-Benefit Analysis within the AHP method and in the sensitivity analysis. To this aim, SCBA provides elements for a comparison with existing competitive products analysing the producers-buyers market demand for new features not yet fully exploited by the existing IPS2.

Overall, SCBA targets to assess the sustainability impact, which includes social and environmental requirements. The sensitivity analysis shows the outcome of the course of action in a medium-long time horizon in order to make the option worth undertaking. The evaluation results support stakeholders in the market knowledge capture about next generation IPS2 based on new enabling technologies. Therefore, this new framework enables to build marketoriented scenarios for production and manufacturing of new high value products and services, meeting and mutually reinforcing competitiveness and sustainability. 7 REFERENCES [1] Jovane, F., Yoshikawa, H., Alting, L., Boër, C.R., Westkämper, E., Williams, D., Tseng, M., Seliger, G., Paci, A.M., 2008, The incoming global technological and industrial revolution towards Competitive and Sustainable Manufacturing, CIRP General Assembly, Manchester, UK, 24-30 August in publishing in CIRP Annals. [2] Westkämper, E., 2008, Manufuture and Sustainable Manufacturing, in publishing in CIRP Annals. [3] Mont, O.K., 2002, Clarifying the concept of productservice system, Journal of Cleaner Production, 10: 237-245. [4] ManuFuture ETP High Level Expert Group, 2004, ManuFuture: a Vision for 2020. Assuring the future of manufacturing in Europe, European Commission Research DG. [5] Manufuture ETP High Level Group and Support Group, 2006, Strategic Research Agenda, assuring the future of manufacturing in Europe, European Commission Research DG. [6] European Commission, 2008, Guide to Cost-Benefit Analysis of investment projects. [7] European Commission, 2005, Impact Assessment Guidelines. [8] Paci, A.M., Chiacchio, M.S., Lalle, C., 2008, Production Paradigm Ontology (PPO): a response to the need of managing knowledge in high-tech manufacturing, Methods and Tools for Effective Knowledge Life-Cycle-Management, Springer: 227240. [9] http://www.prs-pacific.eu/

266

[10] Navigate Consortium, 2007, ERA STAR Survey Navigate Consortium Lombardia Region Contribution. [11] bavAIRia GMES, Bavarian Ministry of Economy Affairs, Infrastructure, Transport and Technology, 2006, Monitoring for environment and security – Bavaria’s capabilities in GMES. [12] http://www.isi-initiative.eu.org/. [13] Corazza, G.E., Arcidiacono, A., Charrier, J.F., Fogliati, V., Leurquin, C., Piccinni, M., Saggese, E., Tjelta, T., 2006, The Integral Satcom Initiative: A Technology Platform for FP7, ASMS 2006 Conference. [14] Paci, A.M., Transectoral technologies impact on sectors, 2007, Deliverable of the Leadership-SSA Project www.leadership-ssa.net. [15] Paci, A.M., 2006, A collaborative industry-research frame for roadmapping, Production Engineering – Knowledge, Vision, Framework Programmes, Conference Proceedings, E. Chlebus (ed), Wroclaw, Poland, 7-8 December: 5-10. [16] Tavana, M., 2004, A subjective assessment of alternative mission architectures for the human exploration of Mars at NASA using multicriteria decision making, Computers & Operations Research, 31: 1147–1164. [17] Rezaei-Moghaddam, K., Karami, E., 2008, A multiple criteria evaluation of sustainable agricultural development models using AHP, Environ Dev Sustain, 10: 407–426 DOI 10.1007/s10668-0069072. [18] Chen, H.H., Kang, H.Y., Xing, X., Lee, A.H.I., Tong, Y., 2008, Developing new products with knowledge management methods and process development management in a network, Computers in Industry, 59: 242–253. [19] Saaty, T.L., The Analytic Hierarchy Process, 1980, McGraw-Hill, New York. [20] Brochure EPPLab, 2008 http://www.itia.cnr.it/docs/EPPlableafletitaottobre200 7.pdf.

267

Metadata Reference Model for IPS2 Lifecycle Management M. Abramovici, M. Neubach, M. Schulze, C. Spura Ruhr-University Bochum, Institute for Product and Service Engineering Chair of IT in Mechanical Engineering (ITM), Germany [email protected]

Abstract Metadata models are the nucleus of product data and lifecycle management. Current PLM solutions offer generic metadata models for physical products and consider only the product provider’s requirements. They are not sufficient to meet the new requirements and have to be extended in order to provide support for industrial product-service system (IPS²). Based on use case analyses, expert interviews and interviews with participants of the research project the different requirements of IPS2 providers and customers are identified and a proposal for a metadata reference model for IPS2 lifecycle management is introduced. This metadata model provides the foundation not only for basic PLM methods but also for advanced IPS2 management methods like IPS2 change management, IPS2 customer feedback management and IPS2 executive information management. The IPS² metadata reference model could be the basis for further standardisation activities. Keywords: 2 IPS , Lifecycle Management, Metadata Reference Model

1 INTRODUCTION Over the last decade PLM has become the central management approach in engineering. PLM is an integrated approach including a consistent set of models, methods and IT-tools for managing product data, engineering processes and supplications along the product lifecycle [1]. The nucleus of current PLM solutions is the so called metadata model. Metadata models are semantic and conceptual information models which describe the main product related object classes used by all the actors and applications along the product lifecycle. In addition PLM metadata models describe the main properties of the object classes as well as their relationships. PLM metadata models could be seen as a directory of integrated engineering processes and serve as a basis for •

ensuring data consistency between the involved applications and partcipants,



providing a trans-sectoral data search and access,



recording the history (e.g. change history) of the managed product data along the lifecycle,



representing links between models in different application systems. PLM metadata models are always context-specific and have to be developed for a specific company or application area, before introducing a PLM solution. In order to minimise the effort for the development of context-specific PLM metadata models and to guarantee the data capability and interoperability of collaborating partners using different PLM environments, a reference PDM/PLM metadata model has been developed from a standard proposal (STEP PDM). This metadata model serves as a reference for the development of current PLM solutions as well as for the exchange of data between different PLM environments.

CIRP IPS2 Conference 2009

268

The STEP PDM (Standard for the Exchange of Product model data Product Data Management) metadata reference model considers the object classes physical product, related documents, engineering processes of the manufacturer as well as users and user roles. The semantic and knowledge contents of STEP PDM metadata reference models are very poor but sufficient for the current generation of PDM/PLM solutions. In the future, reference domain and product-specific ontologies could contribute to a higher level semantic standardisation and improve the efficiency of PLM solutions. The relationships between application-specific data models and metadata models, between reference context-specific and domain-specific ontology models are summarised in figure 1. Generic

Context-specific (e.g. for a company)

Domain-specific

PLM Metadata Model Reference Metadata Models

Ontologies Application Specific Operational Data Models ... ... ...

Figure 1: Relationship between generic, context-specific and domain-specific models Within the interdisciplinary research project "Engineering of industrial product-service systems (IPS²)", nine German research institutes develop new methods and ITtools for the engineering of future IPS². As the level of complexity of PSS data and processes is much higher as that of physical products an integrated management approach as well as powerful methods and tools for the IPS² lifecycle management (IPS² LM) are mandatory for a successful IPS² implementation. As current PLM solutions do not meet the IPS² lifecycle requirements a new IPS² lifecycle management solution is

developed within the above described project. The prerequisite for this solution is a reference metadata model for the IPS² LM. For this, the current STEP PDM reference metadata model has been extended to IPS² and to the new IPS² LM methods. The following solution describes the requirements and the proposal of the new metadata reference model.

Lifecycle Phases

Object Types

IPS2

Disposal / Optimisation

Services

Manufacturing / Realisation

Physical Products

Development / Planning

(mechatronic, smart, …)

2 REQUIREMENTS OF IPS² METADATA MODELS For the identification of IPS² metadata model requirements use case analyses, expert interviews and interviews with participants of the described research project were conducted. The metadata model has to integrate various operative models from the different groups of participants from provider as well as from customer side involved in the IPS² lifecycle. The metadata model should be able to manage filtered information from different operative data sources along the whole lifecycle of an IPS² (see Fig. 2).

Figure 2: Information sources for IPS² LM metadata model The new IPS² metadata reference model has to cover all the needs of the future LM solutions and to mirror all the types of future IPS². The main development dimensions of a future IPS² lifecycle management are (see Fig. 3): •

the management of new object types like services or IPS², considering product embedded information devices (like RFID tags, sensors, etc.)



the coverage of all the lifecycle processes with the focus on the IPS² use phase



the involvement of new user types with emphasis on customers



the support of new process types like executive decision processes The resulting requirements from these IPS² development directions are described in detail in the next sections. 2.1 Management of new object types The IPS2 provider or provider network has to be guaranteed that it can manage independent products and services as well as integrated product and service packages by the IPS² lifecycle system. In contrast to existing PLM solutions, this system has to manage individual customer-specific IPS2 items which can change 2 during the provision/use phase of an IPS [2]. It is important to consider the fact that traditional PLM-systems only control data about product types since they focus on the product development phase. With the integration of the use phase it is necessary to consider individual product items. So far, the development and delivery of products and services has taken place in separate valueadded processes. Here, the IPS2 provider has to be given the opportunity to dynamically manage integrated overall 2 solutions in the form of an IPS .

Core value-added Processes (e. g. Change Management)

Support Processes

Usage / Service Delivery

PLM Metadata Model Producers / Providers

Suppliers

Executive Decision Processes

Customers – Focus of the classical PLM Approach

Users

Process Types – Vision of IPS 2 Lifecycle Management

Figure 3: Multi dimensional extension of the IPS² LM IPS² are characterised by a systematic combination and a corresponding synchronisation of the individual components, depending on customer requirements. IPS² can be classified by means of the business models function-oriented, availability-oriented and result-oriented [3] and the different characteristics like individuality, connectivity, complexity. The combine components cannot be considered individually due to the interdependency of the product share and the productrelated services. Usually, a product share can be offered discretely without the product-related services. However, this turns out to be very difficult with regard to a productrelated service, if no product is available. Requirements for the extended metadata model are: • management of independent product and service components • management of integrated product and service packages (IPS²) • management of dynamic relations between product and service components 2.2 Support of the whole IPS² lifecycle Requirements Management During the analysis, requirements of customers and providers are gathered which are to be taken into account for the later realization of the IPS2 product [4]. Here, customer surveys, feedback analyses and lead user concepts, for instance, are used in order to determine customer requirements, satisfaction and preferences. This method allows the IPS2 provider to react to changes of requirement definitions in the most flexible way as these can have a significant impact on the planned 2 concept, the feasibility and the costs of the whole IPS [5]. Furthermore, the IPS2 customer has to be provided with the opportunity to dynamically adapt or change the demands made on the IPS2 item while using the product. Hence, the customer can react as flexibly as possible to potential changes in the work environment. Functional description of processes and components 2 In order to support the approach of heterogeneous IPS 2 concept modeling [6] the IPS metadata model should be able to link functional descriptions to all managed product and service components or component assemblies. Resource management for IPS² items Interfaces for managing necessary resources (employees, workplaces, machines) and their calendar of capacity and individual planning are to be integrated. Moreover, the qualification of employees can be saved for

269

optimal resource planning. Therefore, it is possible to organise the resources necessary for the product and service shares of an IPS2 product in a skill-based way (skill management) and to adjust them dynamically to the surrounding conditions [7]. Employees, physical spares as well as the tools required for carrying out service processes, for instance, count as resources. Requirements for the extended metadata model are: • management of dynamic requirements related to product and service components • mapping functional descriptions to components or modules • management of necessary resources related to a specific product or service 2.3 Supported process types Due to the high complexity of IPS², future LM solutions have to cover not only value-added management processes but also management support processes like quality management, technical documentation, customer feedback management and executive decision processes. Customer feedback-management The extended IPS² lifecycle management includes 2 product-related knowledge management. IPS providers can only overcome the obstacles of shortened cycles of innovation and dynamically changing customer requirements, if they become aware of the importance of knowledge as a resource and strategically implement it [8]. The IPS2 provider has to be enabled to generate and save the knowledge gained in any form. This includes 2 customer information, generated through the IPS customer feedback management [9], as well as the automatically created product use information [10] of the IPS2 product. On the one hand, the customer feedback management enables the customer to incorporate requirements and ideas for improvement directly into the product development by evaluation of the future product. Realtime gaining customer-specific demands and wishes parallel to product development processes ensures that customer needs are up-to-date [9]. On the other hand, the customer can incorporate his product experience as well as advantages and disadvantages of the product directly into the PLM system during the product’s use phase. Additionally, usage information by the customers (e.g. complaints or queries) are indirectly extracted from the service and after sales domain of the IPS2 provider and led back to the product development [9]. The totality of all objectively measurable information, which arises during 2 the use phase of the IPS product, counts as automatically generated product usage information. This could be, for instance, sensor data, environment parameters, maintenance information or recorded error lists. The statistical analysis of the saved product usage information should enable the thereby generated and condensed knowledge to be used for the development and conceptual design of new IPS2 products. In addition to this, the data of the environmental IPS2 usage conditions of potential product failures and of different usage profiles could be generated on the basis of the saved knowledge and used in the conceptual design phase of further IPS2 products. Integration of business performance indicators 2 For a successful conception and development of IPS products it is necessary to link PDM data and the

270

corresponding economic information, like costs, values and energy efficiency data already at an early stage [7]. From the point of view of the IPS2 provider this means that, for instance, a prognosis based on a simulation of the economic consequences in case of variation of the offered product and service combinations should be 2 supported [11]. However, it is important for the IPS customer to have the costs related to the purchase of a certain IPS2 product, in terms of an automatic total cost of ownership calculation [11]. This means, it is necessary to assign certain costs and values to individual product and service objects. Requirements for the extended metadata model are: • management of customer initiated feedback information • management of product use information • management of business performance indicators • simulation of economic consequences 3

PROPOSAL FOR A METADATA REFERENCE 2 MODEL FOR IPS LIFECYCLE MANAGEMENT Based on the analysis of the requirements of IPS2 providers and customers and the different types of IPS², a proposal for a reference metadata model for the lifecycle management of IPS2 was developed. The proposed metadata model was developed based on the UML (Unified Modelling Language) object-oriented notation. In the course of the development parts of the STEP reference model “AP 214” [12] were taken into account and product-specific elements have been adapted. Figure 4 shows the top level class structure of the developed IPS² metadata reference model, which is an extract of the complete reference model. In order to make the diagram more clear to the reader obvious classes for data management (Document, File, etc.) and finer granulations of each main class have been omitted. Each IPS2 represented by the main class IPS_System consists of several IPS2_Modules, which are in turn composed of different IPS2_Elements. IPS2_Elements may be Services or Products. In this context 2 IPS _Element is an abstract class, which is never instantiated. The class was only introduced in order to reduce the number of associations needed in the class diagram. The class Product models the physical aspects of an IPS². A product class is a core element in every traditional PDM/PLM-System. It represents physical components, software and via the aggregation of these elements their specific product structure. The class Service models services like planning, maintenance, implementation etc. Figure 5 shows exemplary a finer granulation of the top level class Service. In order to decrease the complexity of the shown IPS² metadata reference model the finer granulations of the other top level classes are omitted. The subdivision of the class Service is based on the classification in [13]. These services are like products part of the comprehensive IPS2 the IPS2 provider sells to the customer. In general, Services need human resources, operational equipment, spare parts and tools in order to be provided. These elements are summarised in the class Resource on the top level of the metadata model.

Figure 4: Top level of the proposed reference metadata model for IPS2 lifecycle management Additionally, Usage_Knowledge, Business_Performance failures. Knowledge discovery methods support the _indicators, Function and Requirement may be associated extraction of implicit knowledge from product use to individual Services and Products. information. This knowledge is also referenced via the class Usage_Knowledge and may be used to derive Usage_Knowledge comprises active feedback which is 2 needed modifications for both existing IPS and customer initiated and passive feedback consisting of 2 requirements for the development of new IPS . product use information gained in the IPS2 use phase. Examples for product use information are sensor data, environment parameters, maintenance events and

Figure 5: Finer granulation of services Requirements (for example as a result of analysing usage component XY” which is in turn associated with the knowledge) may be attached directly to the whole Function “inspection of component XY”. 2 IPS_System, to an IPS _Module or to a single To support the IPS² engineering process it is important to IPS2_Element (Product or Service). The different cover each specific requirement with a set of functions associated requirements may depend on each other or which is realised with certain IPS²_Elements. This reflects even conflict each other, which is underlined by the selfthe capability of substitution of products and services. A association. The initial requirements may change during specific function may be realised by a physical product or the IPS² lifecycle (e.g. use-phase) and have to be updated a service. Therefore relations between the classes in order to satisfy the customer’s demands. In this context, Requirement, Function and IPS²_Element have been set the omitted super-class PLM_Item bundles and provides up. fundamental attributes in order to manage the dynamic of Additionally business performance indicators like costs, requirement engineering resulting from changing and new values and energy efficiency factors can be associated by upcoming requirements. the subclasses of Business_Performance_Indicators to The class Function represents functions or functional each product or service via the inheritance association to behaviours like the inspection of a component XY which IPS2_Element. Of course, individual costs may be may be implemented via a sensor (an instance of the aggregated in order to determine the total costs for an class Product) or a human who inspects the component IPS²_Module or a complete IPS². In this context, the IPS² manually. In the last case, the Resource “human” is metadata reference model offers support for the different associated with a Service “manual inspection of 271

types of IPS² business models. In the case of a functionoriented IPS² only physical product structures are instantiated. Obviously an availability-oriented IPS² consisting of IPS²-modules which comprises certain product and service components may also be represented by instantiating the appropriate classes. 4 OUTLOOK The presented IPS2 metadata reference model within the research project TransRegio29 (TR29) has been designed in close collaboration with the project partners who develop operative IPS2-methods and tools. On the basis of the use cases, the IPS² metadata reference model has been validated according to its applicability. The investigations have shown that all domains of the individual use cases could be represented. The metadata model can therefore be considered to be coherent and hence be used. The proposed IPS² metadata reference model could be the basis for the future standardisation of future PLM metadata models. In the scope of further research activities, the expansion of the existing PLM approach by the development of new concepts and methods in the field of intelligent change management, customer feedback management as well as executive information management is carried out. In addition, the development of a top level ontology is necessary in order to create a semantic level for the description of the data. This level, with the help of semantic relations, will allow the intelligent data administration of the gathered information along the whole product lifecycle. The proposed IPS² metadata reference model will be reviewed and refined by the involved research partners. 5 ACKNOWLEDGEMENTS We express our sincere thanks to the Deutsche Forschungsgemeinschaft (DFG) for financing this research within the Collaborative Research Project SFB/TR29 on Product-Service-Systems – dynamic interdependency of products and services in the production area. 6 REFERENCES [1] Abramovici, M., Schulte, S., 2007, Future Trends in Product Lifecycle Management (PLM), Proceedings of the 17th CIRP Design Conference, Berlin, Germany [2] Främling, K., Harrison, M., Brusey, J., 2006, Globally Unique Product Identifiers - Requirements and Solutions to Product Lifecycle Management, Dolgui, A, Morel, G., Pereira, C.E. (eds.), Proceedings of 12th IFAC Symposium on Information Control Problems in Manufacturing (INCOM), Saint-Etienne, France, 17-19 May: 855-860. [3] Burianek, F., Ihl, C., Bonnemeier, S., Reichwald, R., 2007, Typologisierung hybrider Produkte – Ein

272

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

Ansatz basierend auf der Komplexität der Leistungserbringung, Arbeitsbericht Nr. 01/2007 des Lehrstuhls für Betriebswirtschaftslehre – Information, Organisation und Management der Technischen Universität München Bullinger, H.-J., Schreiner, P., 2003, Service Engineering – Ein Rahmenkonzept für die systematische Entwicklung von Dienstleistungen, Bullinger, Scheer: Service Engineering – Entwicklung und Gestaltung innovativer Dienstleistungen, Springer Verlag, Berlin Heidelberg Bullinger, H.-J., 1992, Forschungsund Entwicklungsmanagement in der deutschen Industrie, Scheer: Simultane Produktentwicklung, München Sadek, T., Müller P., Welp, E. G., Blessing, L., 2007, Integrierte Modellierung von Produkten und Dienstleistungen – Die Konzeptphase im Entwicklungsprozess hybrider Leistungsbündel, The 18th Symposium Design For X, Neukirchen, Deutschland, 11-12 October Scheer, A-W., Boczanski, M., Muth, M., Schmitz, WG., Segelbacher, U., 2006, Prozessorientiertes Product Lifecycle Management, Springer Verlag, Berlin Heidelberg Kleinaltenkamp, M., Frauendorf, J., 2003, Wissensmanagement im Service Engineering, In: Bullinger, Scheer: Service Engineering – Entwicklung und Gestaltung innovativer Dienstleistungen, Springer Verlag, Berlin Heidelberg: 371-391. Schulte, S., 2007, Integration von Kundenfeedback in die Produktentwicklung zur Optimierung der Kundenzufriedenheit, Dissertation, ITM, RuhrUniversität Bochum, Shaker Verlag, Aachen Abramovici, M., Neubach, M., Fathi, M., Holland, A., 2008, PLM-basiertes Integrationskonzept für die Rückführung von Produktnutzungsinformationen in die Produktentwicklung, wt Werkstattstechnik online, Springer-VDI-Verlag, Düsseldorf: 561-567. Becker, J., Beverungen, D., Knackstedt, R., Müller, O., 2008, Konzeption einer Modellierungssprache zur tool-unterstützten Modellierung, Konfiguration und Bewertung hybrider Leistungsbündel, ERCIS, Proceedings of the GI-Tagung Modellierung, Workshop Dienstleistungsmodellierung. Berlin: 4562. ISO 10303, 2000, Industrial Automation Systems – Product Data Representation and Exchange: Overview and Fundamental Principles, International Organization for Standardization, Geneva, Switzerland Kortmann, D., 2007, Dienstleistungsgestaltung innerhalb hybrider Leistungsbündel, Dissertation, LPS, Ruhr-Universität Bochum, Shaker Verlag, Aachen

Remanufacturing on a Framework for Integrated Technology and Product-System Lifecycle Management (ITPSLM) 1

A. Guelere Filho, D.C.A. Pigosso, A.R. Ometto, H. Rozenfeld University of São Paulo, São Carlos School of Engineering, São Carlos, Brazil [email protected]

Abstract In PSS context the product lifecycle needs to be managed in a holistic and structured way by using product life-cycle management (PLM) approach. The remanufacturing is an important strategy in this process, since the parts of a product that is being taken back at the end of its service can be remanufactured and new products send back to the market, fulfilling the needed functions. This paper aims to present a framework for ITPSLM composed by three business process (innovation management, technology development, productservice system development) and two support process (configuration management and business process management) and its relation to remanufacturing. Keywords: Remanufacturing; Product-Service Systems; Eco-Innovation; Product Life-Cycle Management

1 INTRODUCTION The traditional approach to environmental management has evolved from pollution control, the end-of-pipe approach, to preventive or cleaner production approaches. The latter is defined as the continuous redesign of industrial process and products to prevent pollution and waste generation at their source and minimize risks to humans and the environment. This approach was applied initially to industrial processes (hence cleaner technologies) and then, to be more inclusive, to the industrial products themselves (hence cleaner products). Recently, it has become clear that such interventions have to be more radical than just the redesigning of existing products in order to catalyze a transition towards a sustainable society. Long-term economic growth in combination with a reduced pressure on the environment asks for changes in our production and consumption systems and for commitment of all actors in society, since consumption and production of products throughout its lifecycle is at the origin of most pollution and resources depletion that our society causes. Economic growth means increasing the perceived value created in society, as expressed in financial terms. It is linked to perceived value creation and not necessarily to material or product streams circulating in the economy. High-perceived value would be achieved by combining an optimal mix of products and services. Furthermore, the image of the product, service and their provider can have a major influence. This results into a value increase of all products in circulation. On average services are to be preferred over products as a means to perceived value creation. The combination of products and services can exceed the traditional functionality of products, in terms of quality and cost performance. The following two items presents the literature review carried out in Product-Service Systems (PSS), Product Life-Cycle Management (PLM) and Remanufacturing. In sequence, it is presented our proposal of a framework for Product Life-Cycle Management (PLM) in the Context of

CIRP IPS2 Conference 2009

273

Product-Service Systems (PSS) and with focus on remanufacturing. 2 PRODUCT-SERVICE SYSTEMS (PSS) The Product-Service System (PSS) concept is a possible and promising business strategy potentially capable of helping achieve the leap, which is needed to move to a more sustainable society. PSS can be seen as strategic innovations which companies may choose in order to separate resource consumption from its traditional link to profit and standard of living improvements; to find new profit centers, to compete and generate value and social quality while decreasing (directly or indirectly) total resource consumption. A Product-Service system is a marketable set of products and services, jointly capable of fulfilling a client's need [1]. It is the result of an innovation strategy, shifting the business focus from designing and selling physical products only, to selling a system of products and services which are jointly capable of fulfilling specific client demands [2]. It is also a result of rethinking of the product value chain and ways of delivering utility to customers that will have a smaller environmental impact than separate products and services outside the system [3]. According to [4], a PSS is a system of products, services, networks of actors and supporting infrastructure that continuously strives to be competitive, satisfy customer needs and have a lower environmental impact than traditional business models. PSS can prove beneficial to the environment in combination to creating (new) business. Key-factors of success are similar in many cases, e.g.:   

Creating value for clients, by adding quality and comfort. Customizing offers or the delivery of the offer to clients. Creating new functions or making smart or unique combinations of functions.

  



Decreasing the threshold of a large initial or total investment sum by sharing, leasing, and hiring. Decreasing environmental load. Often this will bring additional and perceived Eco-benefits. Increase the quality of the contacts with clients.

  

The generic eco drivers for the adoption of PSS are [2]:         

Image improvement. Practicing producer's responsibility. Covenants with authorities. Health and safety management. Environmental costs reduction. Legislation. Published product tests. Non Governmental Organizations and societal pressure. Green purchasing by authorities or consumers.

The successful application of a product-service system requires that manufacturers and service providers extend their involvement and responsibility from products to phases in the life cycle (figure 1), that are usually outside the traditional buyer-seller relationship, such as take back, recovery of products and materials, reuse and refurbishment and remanufacturing [3]. Inputs Principal flow (direct) Secondary Flow (reverse)

Raw Material’s Extraction

Primary Industry

Recycling Manufacturing Remanufacturing

Use

Reuse Residues

Product’s Discard Residues Inputs

Treatment and Final Disposal

Figure 1: Material Life-Cycle of products. Producers are playing the role of the coordinating actor in the product chain. Usual responsibilities for products are extended through an increased or deepened responsibility for service, including the responsibility for educating customers and increasing their awareness about efficient product use and for proper organization of the take back arrangements and systems for reuse, remanufacturing and recycling. Compared to Ecodesign cases, the transition towards PSS will often imply an important process of change for the company towards new thinking on how the company should create value, produce, distribute and approach her clients. 3 PRODUCT LIFE-CYCLE MANAGEMENT (PLM) Product lifecycle management (PLM) can be defined as a business approach for managing a company’s product throughout its life cycle. The PLM is a key element for companies in creating sustainable value and a competitiveness factor in a market where customers and consumers are interested in the environmental impacts of the products they consume. The legislation is another important factor in this scenario.



The sustainable value creation includes the product, process, marketing or organization innovation. It embraces the following aspects: Increase of the environmental, social and economical performance of an existing product. Improvements of products development process. Assessment of environmental impacts of products/manufacturing processes in order to minimize its environmental burden. Definition of a sustainable supply chain (by the definition of new partners, governance rules, adopted technologies, etc.).

The integrated product life-cycle management perspective is influencing the way organizations plan its business, take strategic decisions, develop products and manufacturing process, manage operations, deal with suppliers and consumers, and plan the end-of-life of its products [5]. A short definition for Product Lifecycle Management states that PLM is a concept for the integrated management of product-related information throughout the entire product lifecycle [6]. This concept is rather broad, but it is in accordance with the CIMDATA [7]: ‘PLM is a strategic business approach that applies a consistent set of business solutions in support of the collaborative creation, management, dissemination, and use of product definition information across the extended enterprise from concept to end of life—integrating people, processes, business systems, and information’. Stark [8] in his book states that PLM represents an approach for effectively managing a company’s product throughout its life cycle. He presents an extensive list of needs, justifications, strategies and implementation activities for PLM. Indeed, he emphasizes Product Data Management (PDM) when discussing software. One can come to the understanding that PDM is the heart of PLM, which might be true if the PLM solutions available on the market are analyzed. Nevertheless, project management functionalities are also presented in most existing PLM applications. 4 REMANUFACTURING Remanufacturing is an end-of-life strategy that reduces the use of raw materials and energy necessary to manufacture new products. Economically, remanufacture is an interesting strategy due to preserving the product’s value added during the design and manufacturing processes. As for the environment, the importance of remanufacture lies on extending the product’s lifetime, by diverting them into a second life, given that if the products last longer through remanufacturing, less material can be used to meet customer needs [9]. The remanufacture process can be defined as a product recovery strategy focused product restoration and reconditioning of its parts, in order to rebuild it according to its original design. Remanufacture is an effective manner to maintain products in a closed-loop and to guarantee products end-of-life proper management. Remanufacturing helps reducing environmental and economical costs of manufacturing and final disposal costs of products and components [10, 11]. Remanufacturing as an industrial process includes several stages, among them product disassembly, cleaning and identification of parts, parts recovery, testing and product reassembly [10, 11, 12]. Kerr et al. [13] considers remanufacturing the most efficient way to maintain products in a closed loop. Through remanufacturing, products can be restored to a like new condition, with the same quality and function as new products. Thus,

274

remanufacturing of end-of-life products and components reduces environmental and economical costs both in new product manufacturing as at its final disposal. According to Kerr et al. [13], sustainable production and consumption will only be possible with closed systems. With remanufacturing, a much smaller fraction of the endof-life resources needs to be recovered through recycling. In addition, intelligent remanufacturing systems provide the opportunity for product upgrades, thereby extending product life and incorporating less environmentally harmful technology. There are several Ecodesign’ methods and tools that focus on end-of-life alternatives, such as remanufacturing, and can be successfully applied in order to obtain more sustainable products. The end-of-life ecodesign methods focused on closing the loop of materials generally include more than one end-of-life strategy. Since a product complexity varies substantially, some components, systems or sub-systems are keener to be recycled, reused or remanufactured than others. The proper end-of-life management system must be able to deal with these characteristics of products parts and adjust the disassembly and parts separation processes to obtain higher profits and better environmental performance and results, optimizing the closure of materials loop. Six examples of Ecodesign methods and tools oriented to the end-of-life of products are listed below: Environmental Design Industrial Template (EDIT): This software acknowledges that economics and product design are the factors establishing the end-of-life of products. The software allows the user to analyze the effects of product’s design at its end-of-life. The tools concept is the generation of a product disassembly sequence that optimizes profit generation in a way that end-of-life treatments can be evaluated. This tool allows the designer to define how and with which material the product will be made, choose parts and processes considering some environmental and economic information, access and modify the available data base, and simulate end-of-life results. The user must supply information on materials (weight, toxicity, disposal costs, and information on recycling – material recycling cost and energy used in the process). If parts can be reused or remanufactured, user must inform its costs as well as retail prices. Thus, given the products design, EDIT is able to simulate an optimum disassembly sequence with larger economic value. As a result, the designer knows to which extend the product can be reused, remanufactured, recycled or disposed, plus time and energy spent on the disassembly process [14]. Environmental Design Support Tool (EDST): Evaluates products design on terms of its environmental sustainability, i.e. material selection, recyclability and disassembly analysis. According to the writers, disassembly is the first step in evaluating a product’s environmental performance by this tool, and it provides the time needed on disassembly, number of distinct components and other information. The tool generates an index number to help evaluating the disassembly process (larger numbers imply more difficult disassembly). Material evaluation is made through and index composed by weight of material, total amount of different material used and total amount of dangerous and recyclable materials used. This index is obtained through a questionnaire and its guidelines. Recyclability evaluation is focused on residue management and pollution control. Material endof-life alternatives are reuse, remanufacture, recycling, incineration and final disposal, according to the kind of material used [15].

275

Green Design Advisor: evaluates products through eight metrics: number of materials, mass, amount of recycled material, toxicity, energy use, disassembly time and endof-life disassembly cost. The evaluation is made in two steps: definition of an appropriate data model that includes all relevant data in determining products environmental impact, and environmental point’s calculus. Thus, the method can identify the weaknesses of the product and indicates the direction for improvement [16]. Method to Assess the Adaptability of Products (MAAP): the main purpose of this method is to evaluate the product’s conformity at assembly, maintenance, repair, upgrade and remanufacture processes, as locating potential design improvement areas. Conformity is represented by metrics (adaptation). The closer the value is to one, better the design adaptation, and the closer to zero, the worst it is. Methods clearness is guaranteed by submetrics (remanufacturing, maintenance, repair and upgrading), which establish together the adaptation metric, as well as parts, connectors and space. The submetrics are divided in subcriteria: parts (components and removal direction), connectors (number of different components in each group, number of different components, number of connectors and tools), space (visibility, reach, identification and direction to disassembly), remanufacture (disassembly maintenance, assembly and architecture structure), repair (disassembly, assembly and architecture structure repair) and upgrade (functional and interface decoupling). The evaluation is made through the ratio between the ideal and the real values of a specific parameter. Based on the metrics result, the tool provides guidelines to product improvement in terms of its adaptation [17]. Product Life Cycle Planning (LCP): this tool considers detailed environmental requirement through life cycle perspective, and the environmental aspect is integrated to quality and cost aspects at initial design phases. The study focus on the following life cycle options: “product upgrade”, “product maintenance”, “lifetime extension”, “product or component reuse”, and “material recycling”. At the first stage, a medium or long term production plan and a product plan are established according to business requirements and products lifetime. At the next stage, product specifications and its life cycle are established. The conciliation between costumer and environmental requirements (in accordance with the company’s strategy) lead to an adjustment of values for quality and environmental characteristics. LCP is supported by a tool software, the LCPlanner, based on an Microsoft Excel macro. LCPlanner automatically produces several matrix and analysis graphics using entry data supplied by the designer and data from LCA and QFD databases [18]. There are several Ecodesign tools and methods available aiming at end-of-life treatment and alternatives. In order to put it to practice, Ecodesign’ methods and tools must be integrated to the initial phases of product development processes in a systematic way. Remanufacturing may become more feasible on the same measure as these Ecodesign tools and methods are made available in a systematic way to designers. 5

INTEGRATED TECHNOLOGY AND PRODUCTSYSTEM LIFECYCLE MANAGEMENT (ITPSLM) FRAMEWORK Some proposals of a framework for PLM in the context of PSS already exist, as the Braunschweig Framework of Total Life Cycle Management [19], a systemic and lifecycle oriented framework for a life cycle phase comprehensive point of view on products and the corresponding processes. In the same way, Aurich et al [20] presented in its paper a novel concept for PSS-LCM

that considers customer-oriented PSS planning, integrated PSS development, knowledge based PSS control and life cycle-oriented process management. Similar framework approaches has also been presented by other authors. This paper presents a framework for ITPSLM (Integrated Technology and Product-System Lifecycle Management – figure 2) composed by three business process (Innovation Management, Technology Development and ProductService System Development) and two support process (Configuration Management – ECM and Business Process Management - BPM). A business process (BP) represents a collection of activities that produces a result (product or service) for a specific group of customers. This definition does not equate to generally defining that a process simply transforms inputs in outputs. A BP aggregates many activities that are normally spread out in different organizational entities (e.g., departments or divisions). Activities of same nature can also be grouped in knowledge areas known as process areas, such as project management, requirement management, cost management and so on. The framework presented on figure 2 is divided in three macro-phases that embrace all the lifecycle of a product or product-service system: pre-development, development and post-development. Innovation Management embraces Innovation Strategic Planning and other innovation management activities (listed on figure 2). The portfolio management, that is present on the gates, can be highlighted. The Innovation Strategic Planning begins with the corporation innovation strategic planning, where the market, technology, products and services trends prospection is done considering a horizon of some years, depending on the company sector. This planning can be integrated to enterprise or business units strategic planning. The strategic view of this BP is also explored on technology and product development projects, by the application of portfolio planning techniques, for example.

On the context of Technology Development, the analysis of technology, current product-services and new ideas take part. The prioritized projects are the input for two development funnels: technology and product-service (the two BP of ITPSLM). It is important to address, however, that there are differences between these two funnels. The technologies must be available and established to be used on the product development process, after the conceptual design. The Product-Service System Development is a new concept of New Product Development (NPD) that considers the context of product-service systems. The Product-Service System Development has a broader scope, since the product continues being designed on the “traditional” way. The associated services represent additional activities and new requirements to be considered; According to [21] new product development (NPD, also frequently referred to just as product development) is ‘The overall process of strategy, organization, concept generation, product and marketing plan creation and evaluation, and commercialization of a new product’. Clark & Fujimoto [22] states that ‘Product development process is the resulting process when market information is transformed into information and necessary sources to manufacture a product with the aim of commercializing it’. For Pugh [23] ‘Product development process is the necessary systematic activity from the identification of the market/customer needs until the product sale, an activity that includes product, processes, people and organization’. It is not a new notion that developing products has become one of the key processes for competitiveness in manufacturing. A structured and systematized NPD leads to shorter time-to-market, project repeatability, lean development, and knowledge reuse. NPD systematization provides a standard for project management, also known as a reference model. It defines a common language for all stakeholders involved with NPD and represents a body of knowledge.

276

Figure 2: Business Process of Product Life-Cycle Management (PLM)

One of the well-known factors of the NPD is that the degree of uncertainty in the beginning of the process is very high, decreasing over the time. The decisions in the beginning of the development cycle are responsible for 70% of the cost of the final product [24]. Regarding the product related environmental impacts, considering environmental requirements at the beginning of the product development phase can reduce environment impacts by an estimated 70% [25]. Hence, taking environmental aspects into consideration during the NPD phase plays an essential role in reducing product lifecycle-related environmental impacts. It is important to notice that the NPD is the business process where lies the opportunities for innovation in function fulfillment. The Product-Service System Development also emphasizes the need of a constant and iterative Life Cycle Assessment (LCA) in order to define the requirements and incorporate the sustainable aspects into the PSS; The two final phases (Life Monitoring and Market TakeBack) are responsible for closing the loop of materials on the product-service lifecycle, a primordial condition to reduce the environmental impact of products. All the previous phases and business process must consider reuse, remanufacturing and recycling as strategies for the products’ end-of-life. Reuse is the process of collecting used materials, products, or components from the field, and distributing or selling them as used. Thus, although the ultimate value of the product is also reduced from its original value, no additional processing is required. The process of remanufacturing consists of collecting a used product or component from the field, assessing its condition, and replacing worn, broken, or obsolete parts with new or refurbished parts. In this case, the identity and functionality of the original product is retained. The resulting (remanufactured) product is then inspected and tested, with the goal of meeting or exceeding the quality standards of brand new products. Recycling is the process of collecting used products, components, and/or materials from the field, disassembling them (when necessary), separating them into

277

categories of like materials (e.g. specific plastic types, glass, etc.), and processing into recycled products, components, and/or materials. In this case, the identity and functionality of the original materials are lost. The success of recycling depends on: (1) whether or not there is a market for the recycled materials; and (2) the quality of the recycled materials (since most recycling processes actually reduce the value of the material from its original value, as the material itself has degraded) [26]. Clearly, the choice of which practice is best for an organization will depend on the organization and product characteristics [27]. To successfully implement remanufacturable products, they should have been designed for this purpose previously. Thus, the initial phases of the product-service system development process must consider the aspects of disassembly opportunities, facilities and reverse logistics. Some of the desirable characteristics for remanufacturable products are easy disassembling, easy cleaning, wear resistant, easy to reassemble and valuable when remanufactured. The Configuration Management (ECM) is related to engineering changes across the whole product life cycle. ECM support process is responsible for manage all the information related to a product. Two distinct terms can be defined in the configuration management: the Configuration Item (CI), that is the collection of all objects that form a product or part of it, and the Configuration Document (CD), that are all the information that characterize the CIs. The configuration management manages the CIs through the management of CDs and CIs meta-data. The configuration management organize and control, in a uniform and systematic way, the descriptions envolving its identification, control and audit. Business Process Management (BPM) is a field of knowledge at the intersection between management and information technology, encompassing methods, techniques and tools to design, enact, control, and analyze operational business processes involving humans, organizations,

applications, documents and other sources of information [28]. BPM deals with the continuous improvements of the entire PLM. It is a structured and systematic approach to analyze, improve, control and manage process aiming to improve the quality of products and services. Hence, in light of the PSS concept, BPM will be responsible for the shifting from NPD to IPSLM. 6 SUMMARY Sustainability demands the balance between economical, social and environmental aspects. Most part of the existing environment impacts is product-related. Thus, a significant decrease in product-related environmental impacts is prerequisite to sustainability. The PSS emerges as a promising approach for a more sustainable society, since it makes possible, between other things, the remanufacturing as an end-of-life strategy. The concept of Product Service System implies a shift in business thinking from selling products to providing service solutions to customer needs. Therefore, the reuse, remanufacturing and recycling of products become more feasible and can be implemented more easily. The introduction of new ownership patterns, such as leasing focused on extending the products lifetime, increases manufacturers interest in designing for durability and enabling the reuse and remanufacturing of products and cores prior to recycling.

[5] [6] [7]

[8]

[9]

[10]

[11]

[12]

The PLM framework in the context of PSS proposed in this paper is unprecedented since it: 

Links all the business process throughout the product life-cycle.  Includes the conceptual and methodological framework of ecodesign.  Understands the life-cycle of a product from the raw material extraction to the products’ end-of-life.  Integrates sustainability into the business view in order to aggregate value to the product.  Includes the product monitoring and take-back, a non-modeled business process, aiming to close the loop of materials in a cycle economy. The grounding of a strategy for integrating environmental issues into the Product Life-cycle Management process must consider the products complexity and diversity and the fast evolution of knowledge on products and services design. It is important to notice, however, that no environmental improvement will be achieved unless new products are competitive and can fully replace low environmental performance products. Thus, product’s functionality, performance, aesthetics, quality and cost must be compatible with environmental requirements. 7 ACKNOWLEDGMENTS The authors would like to acknowledge to the financial and infrastructure support of Institute Factory of Millenium (IFM), Nucleous of Advanced Manufacturing (NUMA) and FAPESP. 8 REFERENCES [1] Goedkoop, M.J., van Halen, C.J.G., te Riele, H.R.M., Rommens, P.J.M., 1999, Product Service systems Ecological and Economic Basics. [2] UNEP, Product Service Systems and Sustainability: Opportunities for Sustainable Sollution. France [3] Mont, O., 2000, Product-Service Systems, The International Institute of Industrial Environmental Economics, Stockholm, Sweden. [4] Mont, O., 2004, Product-Service Systems: Panacea or myth?, The International Institute for Industrial

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

278

Environmental Economis, Lund: Lund University, pp. 146. Rowledge L.R., Knowledge Management for Sustainable Value Creation, EKOS International. Saaksvuori, A., Immonen, A., 2004, Product lifecycle management, Springer, Berlin. CIMdata, 2002, Product Lifecycle Management: Empowering the Future of Business, www.cimdata.com. Stark, J., 2005, Product lifecycle management: 21st century paradigm for product realization, Springer, London. Zwolinski, P.; Lopez-Ontiveros, M.A.; Brissaud, D., 2006, Integrated design of remanufacturable products based on product profiles. Journal of Cleaner Production vol. 14 1333-1345. White, C.D.; Masanet, E.; Rosen, C.M.; Beckman, S.L., 2003, Product recovery with some byte: an overview of management challenges and environmental consequences in reverse manufacturing for the computer industry. Journal of Cleaner Production 11, 445-458. Kerr, W.; Ryan, C., 2003, Eco-efficiency gains from remanufacturing: A case study of photocopier remanufacturing at Fuji Xerox Australia. Journal of Cleaner Production 9, 75–81. Sundin, E.; Bras, B., 2005, Making functional sales environmentally and economically beneficial through product remanufacturing. In: Journal of Cleaner Production 13, 913-925. Kerr W, Ryan C., 2003, Eco-efficiency gains from remanufacturing: A case study of photocopier remanufacturing at Fuji Xerox Australia. Journal of Cleaner Production, 9: 75–81 Spicer, A.; Wang, M, 1997, Environmental Design Industrial Template (EDIT) a software tool for analysis of product retirement. In Journal of Cleaner Production. Yu, S. Y.; Zhang, H-C.; Ertas, A., 1999, Environmental Conscious Design an Introduction to EDST. In Journal of Integrated Design and Process Science. Sun, J.; Han, B.; Ekwaro-Osire, S.; Zhang, H-C., 2003, Design for Environment: Methodologies, Tools and Implementation. In: Journal of Integrated Design and Process Science. Willems, B.; Seliger, G.; Duflou, J.; Basdere, B., 2003, Contribution to Design for Adaptation: Method to Assess the Adaptability of Products (MAAP) In: Proceedings of EmDesIgnW: Third International Symposium on Environmentally Conscious Design and Inverse. KOBAYASHI, H., 2005, Strategic evolution of ecoproducts: a product life cycle planning methodology. In Design. Herrmann, C., Bergmann, L., Thiede S., Halubek, P., 2007. Total Life Cycle Management – An Integrated Approach Towards Sustainability. 3rd International Conference on Life Cycle Management. University of Zurich at Irchel, Zurich. Aurich, J. C.; Schweitzer, E.; Fuchs, C.: Life Cycle Management of Industrial Product-Service Systems. In: Takata, S.; Umeda, Y. (Hrsg.): Advances in Life Cycle Engineering for Sustainable Manufacturing Businesses, London: Springer, 2007, 171-176. PDMA Glossary for New Product Development. Avaiable at:

. Accessed on: Feb. 15th 2007. [22] Clark, K.B., Fujimoto, T., 2001, Product development performance: strategy, organization and management in the world auto industry, Harvard Business School Press. [23] Pugh, S., 1990, Total design: integrated methods for successful product engineering, Addison Wesley. [24] Boothroyd, P., Dewhurst, W., 1994, Product Design and Manufacture for Assembly. Marcel Dekker.

279

[25] Graedel, E., Allenby, R., 1995, Industrial ecology, Prentice Hall, New Jersey. [26] Beamon, B.M., 1999, Designing the Green Supply Chain; Logistics Information Management. [27] Sarkis, J., 2003. A strategic decision framework for green supply chain management, Journal of Cleaner Production. [28] Van der Aalst, W.M.P., ter Hofstede, A.H.M., Weske, M., 2003, Business Process Management: A Survey, Business Process Management, Proceedings of the First International Conference, Springer Verlag.

Commercializing Sustainable Innovations in the Market through Entrepreneurship D. Keskin, H. Brezet, J.C. Diehl Industrial Design Engineering, Delft University of Technology, Landbergstraat 15, Delft, 2628 CE, The Netherlands [email protected]

Abstract The system innovations for sustainability require changes at multiple-domains (social, cultural, institutional and technological) and multiple-levels (micro-, meso-, and macro-levels) of the socio-technological system. From a sustainable production and consumption view, systems innovations can be enhanced through a PSS strategy at organisational level. This can be achieved through new entrepreneurial entries because of their potential in commercialising sustainable innovations and consequently bringing the necessary institutional change that favours such innovations. This paper aims at investigating how sustainability-driven entrepreneurs enhance the commercialisation of sustainable PSS and the consequences of this process and the potential role of design in it. Keywords: Sustainability, (Product-Service) System Innovations, Entrepreneurship

1 INTRODUCTION Increasing demand for resources coupled with increasing world population suggest that we need radical changes, also often referred to as system innovations, to create larger jumps in societal systems’ efficiency and achieve the goals of sustainability. A system innovation means a shift from one socio-technological system to another (e.g. transport, energy, food) and requires changes on different domains: social, cultural, institutional and technological [3]. The multi-level concept divides the sociotechnological system into micro-, meso-, and macrolevels, i.e. niches, regimes and socio-technical landscape, respectively. The changes occur either in a bottom-up fashion or a top-down fashion, i.e. “… breakouts at the micro-level find fertile soil at the macro-level, or a break through at the macro-level can be accompanied by suitable initiatives at the micro-level” [14]. From this perspective, the focus of this paper will be the organizations that execute innovation experiments at the micro-level and how they may influence the other elements of the societal system and bring the required radical change at meso-level and consequently at macrolevel, which are more resistant to changes. From a sustainable production and consumption view, the concept of product-service systems (PSS) has been suggested as a promising strategy for organisations operating on the micro-level of socio-technical systems. The elements of PSS consist of a system of products, services, network of actors and supporting infrastructure [11] that corresponds with the multiple dimensions of the socio-technological system: social, cultural, institutional and technological. The elements of PSS should be designed and continuously adjusted to each other aiming at sustainable system innovation and optimisation [11]. Besides the elements of PSS and sustainability criteria, the institutional environment (i.e. meso-level) in which the organisations operate is an important consideration since it affects the success and institutionalisation of the PSS. The institutional environment is characterised by a set of norms, expectations, procedures, standards and routines [6]. There is a need for the change of the institutional environment that favours sustainable solutions and their acceptance by the society. According to van den Hoed [6], there are five sources of institutional change: (1)

CIRP IPS2 Conference 2009

280

shocks; (2) market changes; (3) new entrepreneurial entries; and (4) institutional entrepreneurship; and (5) new technologies. In the overall picture of the recent 25 years history of sustainable production and consumption, only new entries and new technologies have contributed to relevant changes, while shocks (not to manage anyway) and market changes almost have not occurred. Therefore, from the perspective of radical change, there is a need to focus here on the role of entrepreneurship in relation to sustainable PSS innovation. In this study we are particularly interested in the role of the combination of new entrepreneurship and design in the development of successful radical new product-service systems. In this framework, this paper aims to explore the following main and sub-research questions: How can sustainability-driven entrepreneurs enhance the introduction and success of sustainable PSS? - What is the role of design in this process? - What are the consequences of this process regarding institutional environment, infrastructure and user practices? In order to do so we first discuss and define PSS, entrepreneurship and the role of design. Next, three case studies will be presented to explore and analyse the impact of entrepreneurship and design on the success of sustainable PSS. 2 PSS IN THE CONTEXT OF TRANSITIONS As previously indicated in the introduction, system innovations require changes on different domains: social, cultural, institutional and technological - comprising elements such as technology, regulations, user practices and markets, cultural meanings, infrastructure, etc. [3]. From a demand side perspective, technology plays an important role in fulfilling and realizing functionalities in user contexts which are made up of users, their competences, preferences, cultural values and interpretations; and shaped by a variety of existing products, infrastructures and regulations. On the supply side, technology is produced, distributed and tuned with existing user contexts, requiring aspects like technological knowledge, machines, various actors, skilled labour,

capital, natural resources, distribution networks and regulations (Figure 1). The term ‘socio-technological system’ represents the co-dependence and interrelatedness of demand and supply aspects [3].

Figure 1: Different domains of the socio-technological system In this framework, the concept of product-service systems (PSS) is a promising strategy for organisations operating on the micro-level of socio-technical systems. The elements of PSS consists of a system of products, services, network of actors and supporting infrastructure that corresponds with the multiple domains of the sociotechnological system: social, cultural, institutional and technological - involving new markets, user practices, regulations, infrastructures and cultural meanings (Figure 2). The elements of PSS should be designed and continuously adjusted to each other aiming at system innovation and optimisation [11].

Figure 2: The correspondence of PSS elements with the multiple domains of socio-technological system The significant role of infrastructure should be underlined here since it has a direct effect on individual consumption patterns and environmental impacts. Infrastructure is codependent to products which influence the shaping of infrastructure. Because products have shorter innovation cycles, the design and development of products are easier than infrastructure. However, product improvements through redesign provide maximum factor four improvements. Moreover, it is a shared opinion that factor 10, 20 or higher improvements are necessary to keep within the limits of the environmental impact of the year 1990 in the year 2025. This requires improvements beyond product level including infrastructure [11].

2.1 Defining PSS Based on the aforementioned PSS elements and sustainability criteria, a PSS is defined as [11]: “… a system of products, services, networks of actors and supporting infrastructure that continuously strives to be competitive, satisfy customer needs and have a lower environmental impact than traditional business models.” Therefore, PSS is not merely selling physical goods or services but designing a combination of products and services where the focus is given on environmental concerns, economical feasibility of the systems and social issues [19]. In a PSS strategy, the concept of product is not just the result of traditional production processes but rather the result of a system of physical products and services which are mutually combined to satisfy a specific client demand. The central value of products is exchanged with the value of utilization where the customers pay for performance [10]. In such a scenario, different types of relationships have to be established so that the system will be more favourable to customers than the traditional production system. At the customer side (or demand side), consumption is a satisfaction-based process and tangible products are not the only way of providing this to customers. In other words, customers are not searching only for products or services but rather for a system of products and services that satisfy their needs and desires [20]. Therefore, a PSS should be designed in a way that will be more desirable to customers than tangible products alone [15]. The added value of a product previously came from the production processes that transform raw materials into products. But today this is changing and the added value come from all the non-material aspects of a product, which are technological improvements, product image, brand name and aesthetic design [10]. At the business side (or supply side), therefore, companies are moving away from mass production to mass customization and using more and more services to compete and differentiate in the market [10]. This means that the companies should better understand its customers, which requires a tight relationship of customer and the company. Such an approach, furthermore, brings different kinds of partnerships with other producers and suppliers, public bodies and non-profit organizations for an integrated solution to satisfy customer needs [20]. A “sustainable” PSS strategy forces the industry to focus on a system thinking [8] with the aim of optimizing the interests of all the bodies involved in the PSS and also improving resource usage and environmental quality. Besides the elements of PSS and sustainability criteria, the institutional environment (i.e. meso-level) in which the organisations operate is an important consideration since it affects the success and institutionalisation of the PSS. Institutions do not only comprise regulations and legislations but also the societal norms, ethical rules and established patterns and lifestyles. Therefore, they play a significant role on the way the business is structured and consequently the consumption patterns of the society [11]. On the other hand, there is a need for the change of the institutional environment that favours sustainable solutions and their acceptance by the society. Entrepreneurship is one of the sources of institutional change since new entries can alter power structure within the institutional context providing opportunities for a new dialogue and discussion. This, consequently, may lead to a renegotiation of established institutions. The reason why new entrepreneurial entries are likely to be more innovative than establishes firms is that they are less constrained by the vested interests and developed

281

routines and consequently propose new practices [6]. Furthermore, fast-decision making, flexibility and allocating resources external network increase the likelihood of developing and implementing new products and services [1]. Besides entrepreneurship as a change agent, it plays a significant role in commercialising the PSS. The concept of PSS has been deeply studied at the academic level and knowledge has been created regarding its characteristics, potential benefits, barriers, design tools and methods. However, this knowledge has barely found applications in practice [22]. From this perspective, entrepreneurship is being suggested to play a critical role in enhancing the introduction of successful PSS. 2.2 Entrepreneurship Although there is a variety of definitions of entrepreneurship in literature; entrepreneurs, in general, are defined as individuals who conceive new business opportunities and take on risks required to convert those ideas into commercial reality [16]. As economist Joseph Schumpeter [17] described, they are often agents of ‘creative destruction’, which transform old-ways of doing through a dynamic pattern of innovative upstarts that unseat established firms. Hart and Milstein [5] argue that economy is driven by firms that are able to capitalize on the “new combinations” and entrepreneurs creating new processes, products, and markets tend to be the key actors in this process of change - such as in transition from coal-age technologies to oil-age technologies which are now giving way to information-age technologies. The emerging challenge of global sustainability is a catalyst for a new round of creative destruction which offers unprecedented opportunities to entrepreneurs with the foresight to capitalize on it [5]. The field of sustainable entrepreneurship is interdisciplinary by nature. From the entrepreneurship perspective, it focuses on the activities of individual entrepreneurs and the impacts they have on the wider socio-economic system. From the sustainability perspective, it focuses on the development of sustainability of whole societies and ecosystems. Therefore, sustainable entrepreneurship links the microlevel entrepreneurship with macro-level sustainable development through the organisations that operate at meso-level of the societal system. The significant role of organisations in this link is that they function as an essential tool of the entrepreneurs whilst they constitute a huge part of the institutional landscape of the society [12]. The real sustainability gains will be made by harnessing the innovative potential of entrepreneurship to resolve environmental challenges with innovative PSS solutions. 2.3 The role of design Entrepreneurs can contribute in several ways to sustainable (product-service) system innovation [11]. The four elements of PSS mentioned previously can serve to design of new, significantly more sustainable PSS. For instance, an entrepreneur can choose to develop a service, using a relative lower number of existing products or can develop a product based upon a new technology, like solar energy. But also a new organization or more efficient use of existing infrastructures can contribute to the new system. The role of design in this context mainly to adapt, improve and innovate physical products as enablers of new PSS systems. In this sense they are the part of the creative –destructive- approach of usually new entrepreneurs, able to map potential future consumer needs, to value the potential of new product technologies

282

and to develop novel products that stimulate radical changes in production and consumption. 3 RESEARCH METHODOLOGY Building upon the work of Berchicci on sustainable entrepreneurship [1], in this study we are particularly interested in the role of the combination of new entrepreneurship and design in the development of successful radical new product-service systems. For this purpose a case study research was carried out. This case study research is of an explorative nature and primarily expects to derive descriptive and explanatory findings from the cases through company profiles and interviews. The subject matter of the case study is young entrepreneurial firms in process of commercializing sustainable PSS innovations. The three companies which have been investigated are: (1) Enviu (www.enviu.org) (2) Tuk Tuk Company (www.tuktukcompany.nl) (3) Evening Breeze (www.evening-breeze.com). Most of the data are derived from personal interviews with the representatives of the company with in addition internet pages and company documents. An attempt is made to classify the three cases according to the PSSframework as developed by Mont [11] and to draw some first preliminary conclusions. Each case will be classified in the following categories: Activity of company (project development, competition, incubation, business model development, venture planning), type of product(s) (redesign, new design), type of service (support services, point of sale services, various concepts of product use, maintenance services, and end-of-life services), network of actors (existing chain, new chain) and infrastructure (existing, new). 4

CASE STUDY

4.1 Enviu Enviu is an international non-profit organization by and for young professionals, students and young entrepreneurs with mainly a business, economic or design background. Enviu has a relative small staff of 11 professionals, however each year more than 2.000 volunteers are participating in the program. Via their work, study and in free time they are involved in projects aimed at delivering sustainable solutions to the market. The mission of Enviu is to cooperate on developing sustainable solutions to environmental issues. This is done on the basis of requests (i.e., demand), which means that an issue must be recognised locally, and local stakeholders must have expressed their desire to solve it [2]. Beside the position of Enviu in relation to sustainability, stakeholder dialogue (SHD) (Figure 3) between the different layers of society is an important consideration during the development of projects [2]. The added value of SHD is the strengthening capacity of and between local stakeholder groups, realisation of the stakeholder dialogue, facilitation and stimulation of the processes to initiate more sustainable practices with the emphasis on a business approach with a social face and delivering applied socio-economic and business knowledge and access to knowledge networks [2].

implementation into the business, promotion, food and beverages. In the Netherlands the concept has diffused to other clubs such as the 058Podium in the north of the country. Here, a new organization was created, under the name Entertain&Sustain aiming at integrating pleasure and sustainability issues among the youth in a natural way [21].

Figure 3: The vision and operating procedures of Enviu The core activities of Enviu include executing specific projects that contribute to sustainable solutions, developing activities that create awareness and commitment among young people and collecting and developing knowledge about different fields related to sustainable development. The most important activities in projects consist of supporting partners in capacity strengthening, applied research and to facilitate the realisation and execution of stakeholder dialogue. This approach is also reflected in the type of projects initiated: The Sustainable Dance Club (SDC) The SDC is a spin-off company of Enviu together with Döll that aims to green the clubbing scene. Central element in this club has been the design of the “energy flux floor” (Figure 4), enabling dancing people to generate electrical energy by moving the floor elements [13]. The floor has two versions: permanent integrated into an existing club, and mobile for hiring.

Figure 3: Energy flux floor Besides the floor, other elements have been added to the sustainable dance club concept, from water and waste reduction measures to improved bar logistics, personalized new drinking devices and energy reduction via sound equipment redesign [4, 7, 18]. Furthermore, the design of a mini-sustainable dance club emerged that can be used at events like festivals and expos. In all subprojects new business chains have been created and improved and totally new designed products have played an important role, where Enviu is the main initiator and facilitator. Besides the existing and new products, SDC as a company offers services to clubs that want to adopt the system, such as: sustainability scan (calculating the carbon footprint of clubs and events), consultancy for the possible improvements regarding environmental impact, economical and organisational aspects of the company (ie. clubs), workshops demonstrating sustainability

The Hybrid TukTuk Competition The Hybrid TukTuk project is an international competition for the redesign of the existing tuk tuk taxis for the lower end of the market in India. In large cities across Asia, a million auto-rickshaws serve as one of the most important means of transportation every day. At the same time, these auto-rickshaws cause considerable air pollution and a large amount of CO2 emission. The auto-rickshaw drivers are part of the poorer groups of society and earn on average $3 to $4 a day. Through this project Enviu aims to actively involve young entrepreneurial people in the development and introduction of a Hybrid Tuktuk upgrade kit through their study, work and free time. The guiding concept is to have different prototypes or test model upgrade kits which make existing tuk tuks environmentally sound in an effective and affordable manner and to simultaneously improve the socioeconomic situation of the tuk tuk community. The project goals are: - To improve the air quality in large Asian cities through the reduction of the CO2 emission of auto-rickshaws by 40% to 60% - To improve the economic situation, and with that the social position, of the auto-rickshaw drivers and their families by lower fuel costs and more efficient motors This competition will take off in 2008, with the involvement of a number of international teams of industrial design and automotive engineers. Existing tuk tuk platforms, additional equipment and financial support will be provided by sponsors from different backgrounds (http://hybridtuktuk.com). The Rotterdam Innovation Lab In collaboration with financial organizations, the Rotterdam Harbour Authority, Hogeschool Rotterdam and YES!Delft (the incubator of Delft University of Technology), Enviu is creating an incubator for sustainable entrepreneurship and new ventures for the sustainable future of Rotterdam as a harbour city. Particularly the aim is to develop the right conditions to bring sustainable solutions to the marketplace, and do it on a profitable way. 4.2 Tuk Tuk Company Tuk tuk is increasingly recognized as an alternative to the taxi for the inner city transport in Europe. In parallel to the hybrid tuk tuk project of Enviu, Tuk Tuk Company (TTC) is currently engaged in the improvement of the existing Thai tuk tuk and the development of a new sustainable tuk tuk. The company started in 2006 to challenge the existing taxi business since the taxis are currently not very affordable in the Netherlands and as a result most of them are waiting for customers in front of train stations and taxi stops. Tuk tuks were imported from Thailand and adapted to the European standards. Currently 55 tuk tuks are driving on the roads of Amsterdam, The Hague and Rotterdam. TTC has seen the opportunity and the need to make the tuk tuks more sustainable which lead to: 1) the

283

improvement of the existing tuk tuks and, 2) the development of a new sustainable electric tuk tuk. Redesign of the existing tuk tuk After having tested the tuk tuks in Dutch cities, TTC determined that Thai tuk tuks lacked reliability and comfort levels of European standards and customer expectations, which lead to the re-design of the existing tuk tuks (Figure 4). The company set three goals for the re-design: sustainability, weather independency and maintenance reduction [9]. The tuk tuks were built with LPG tanks in order to make them as green as possible for now. Furthermore, Thai tuk tuks do not have a front break. RDW (Rijksdienst Wegverkeer – the Dutch agency which controls the licenses for Dutch roads) demanded a front break which led to the adaptation of the front wheel construction.

Figure 4: Redesigned tuk tuk New design of an electric tuk tuk The new tuk tuks (Figure 5) in development will be electrical, fully emission free and will replace the old tuk tuks. The E-tuk (E stands for electric) is being currently developed in cooperation with the Delft University of Technology and the University of Arnhem. TTC aims to maintain the heritage of the old tuk tuk and make it suitable for the European city environment. Tuk tuks have a very low radius in personal transport. In other words, they do only short drives of 1-2 kilometres, whereas other types of personal transport such as cars and taxis have bigger radius which makes them dependent on the infrastructure (i.e. for charging) or force them go hybrid to enlarge the radius. Small radius of transport of tuk tuks makes it independent of the infrastructure and allows different charging options (i.e. fast or slow). In addition, for a small company like TTC with relatively small cash flow, it’s difficult to find resources to finance the innovations. The company acknowledges the financial support of the Dutch government for the development of E-tuk.

Figure 5: E-tuk

284

Business model TTC has a franchise scheme, which fosters scaling up of the system. The company delivers tuk tuks to entrepreneurs/drivers who franchisees the concept, making themselves responsible for their earning by selling singular drives on the streets. They are also responsible for the maintenance of the vulnerable tuk tuks which decreases the maintenance costs for the company. The company also offers drivers services such as PR and local marketing. TTC hopes to start including microfinance for drivers into their business model. Because the tuk tuks are rented for the cost price, they bring no profit. The main income of the company is the advertisements at the roof of the tuk tuks, however this often fluctuates. For that reason, TTC plans to collaborate with an advertising company to create a win-win situation for TTC through the network of the advertising agency and for the agency through the shares of the advertisement incomes. Apart from the advertisements the company profits from events. When a specific group that needs to be transported at a specific time from A to B and back, TTC coordinates everything. They have a wide range of customers for events such as companies, event agencies, hotels/tourists, and other companies in the leisure industry (for example canal companies in the Netherlands which organize canal tours for tourists). Barriers The local governments of Amsterdam, The Hague and Rotterdam do not offer TTC parking places for tuk tuks as long as they are not electric. Currently this is the main challenge for the company which prevents them building their customer base. The drivers stay ‘stand-still’ anywhere on the streets but not in parking position. For this reason, TTC has recently signed a cooperation agreement with the Rotterdam Central Taxi (RTC). The tuk tuks are being offered to public and tourist in the centre of Rotterdam. Social aspect In addition to its commitment with the environment, the company has also the ambition to offer a social service to the community with their GoodWerk (in English: GoodWork) project. The project is being developed for refugees and disabled people to allow self-employment through entrepreneurship. The company helps refugees and disabled people to start their own business with tuk tuks or offer employment allowing them to use tuk tuks. TTC is subsidized by DWI (Dienst Werk en Inkomen – service for employment and income) for each refugee that is being trained. 4.3 Evening Breeze Evening Breeze is a spin-off venture which has been started in 2006 at the Design for Sustainability Programme of the Delft University of Technology. The company determined - together with some other environmental experts - that air conditioning accounts for 80% of the energy use in tropical hotel rooms, which results in high energy bills for the resort management and an environmental load for the community. Moreover the experimented comfort of the current air conditioning systems is disappointing. A research revealed noise and draft to be the biggest complaints. To overcome these challenges, Evening Breeze designed and developed the Evening Breeze bed (Figure 6) - an air conditioned canopy bed which provides the sleeper with a

comfortable sleeping environment while reducing the 3 cooled space from the entire room, approximately 80 m , 3 to the bed, a mere 8 m . The company is currently in process of the first demonstration projects in Mozambique, Caribbean, and South Africa.

Business Model The industry value chain of the company is shown in Figure 8. Evening Breeze is the link between suppliers and the local installation partners; however the resorts and end users are important for the company not only for marketing activities but also for their feedback on the design of the Evening Breeze bed.

Figure 8: Industry value chain

Figure 6: The Evening Breeze bed at Sefapane, South Africa. The Evening Breeze Bed The system comprises of a low capacity high efficiency cooling unit. According to the user’s preferences the air is cooled, dehumidified and filtered and gently spread over the sleepers through a porous section in the ceiling of the canopy bed (Figure 7). A matching mosquito netting protects the sleeper against unwanted intruders and enhance the air circulation.

Figure 7: The Evening Breeze bed climate system A huge greenhouse gas emission reduction is realised with the Evening Breeze bed. The annual 5 MWh/year saving potential exceeds the electricity consumption of a West European household and causes 3 tons of CO2 emission. Moreover harmful emissions are reduced by minimizing transport and cooperating with local suppliers. The application of environmentally friendly materials minimises the ecological footprint. The R410A coolant contains no chlorine that reduces the damage to the ozone layer and allows more efficient operation, due to the higher heat transfer. Commercialisation Demo beds were tested at some of the beach resorts located on the island of Bonaire, Dutch Antilles. Subsequently some resorts in South Africa were equipped with an Evening Breeze bed. Reactions of the first users were positive and provided the input for the optimisation phase. The first resort to completely adopt Evening Breeze beds will open in the beginning of 2009 in Mozambique. The beds for this project are locally manufactured in Zanzibar.

The company strives for excellent partners in sustainability and teams up with bed manufacturers to deliver the beds to its customers. For the African and Caribbean market local manufacturers are preferred, for the American and European market branded manufacturers are chosen. Although there are some design guidelines for the bed, there are no standard beds, i.e. the local companies are responsible for their own design with different materials and style. The mosquito nettings are produced by the South African based Kiwinet and the Thai based TanaNetting. Both companies excel in their sustainable business model. Kiwinet mainly employs previously disadvantaged women and TanaNetting is the preferred supplier of the WHO. The installation partners are responsible for the service and maintenance of the Evening Breeze bed and receive marketing and technical support from Evening Breeze. The company itself is divided into two departments: the commercial department focused on the installation partners, resorts and end users; the technical department focused on the suppliers and responsible for the design and assembly. Besides the suppliers, installation partners and resorts, Evening Breeze currently works with an advisory board which supports the company with knowledge on sustainable tourism, market insights and the local network. The first three years Evening Breeze has a negative cash flow and consequently needs financing. Currently, financial support comes from different channels, such as government subsidies (both Dutch and local governments), local NGOs, banks, informal investors who also couch the company in terms of organisational matters and networking. Apart from that, it receives knowledge support from different consultancies, as well as the Delft University of Technology. 5 DISCUSSION The findings of this exploratory case study have been summarised in the following tables (1,2,3) for each case. The cases have been classified in the PSS-framework of Mont [11], involving the design of products and services and the consequences of this on the infrastructure and network of actors. Enviu innovates in different ways. In the SDC project (= new company, in which Enviu, Watt and Döll cooperate) the role of design is both creating total new club products/experiences and integrating existing best practices. The Hybrid Tuk Tuk project is an international competition, in which redesign is the key, and is preceding future green entrepreneurship, if the results are successful. The Climate Innovation Lab is a special incubator project, focussing on selecting green new ventures by students for the Rotterdam Climate Program.

285

The activities of Enviu cover a mix of new design, redesign, setting up new organizations, improving services and using/adapting existing infrastructure. The key is that not only Enviu is a green entrepreneur, building on best practices and total new design, but that this is also the point of departure for the facilitation (birth and growth) of succeeding new green ventures. Tuk Tuk Company has innovative projects regarding the different aspects of sustainability. By redesigning the existing tuk tuk, the company aims to improve it in terms of its environmental performance, reliability and comfort for European customers. Likewise, the new E-tuk project aims to improve the environmental performance of the tuk tuk by integrating a new technology and preceding green entrepreneurship. In both projects design has a crucial role of making use of the old and new technologies, understanding customer needs and contextual conditions of both company and design environment, and bringing future scenarios of use. Additionally, E-tuk project requires design of new facilities for charging the tuk tuks which requires changes in infrastructure. The business model of the company enables scaling up of the project since this way individual entrepreneurs are

offered different support services and encouraged for selfemployment. Similar to franchising concept is the GoedWerk project which has a social aspect as well. Apart from redesigned and newly designed products with better environmental performance, the projects of TTC create new partnerships within local and international chains that create economical and social benefits. Evening Breeze is a young company in the process of implementing its first demonstration projects in Mozambique, Caribbean, and South Africa. This results in new business chains through collaborations with different local stakeholders including air-conditioner installation companies, manufacturing companies, eco-resorts, as well as different suppliers. The innovation of Evening Breeze bed is mainly in the use phase of the product, which reduces energy consumption up to 80%. The technology used in the product is not new, but it is a combination of existing technologies adjusted for the bed. The company sees the opportunity of leasing the cooling system to increase its efficiency by allowing different resorts to use it at different parts of the year, however the necessary capital for leasing do not exist due to the company’s size and age.

Design

Consequences

Project

Activity

Product(s) type

Service

Infrastructure

Network of actors

Sustainable Dance Club

Combination of old & new companies in one new company (SDC)

Radical innovation (dance floor) and product adaptation and existing products

To clubs: Sustainability scan, consultancy, workshops To the end users: Adapted drinking and bar service

Adapted building

Existing and new chains, volunteers, strong links to universities

Hybrid Tuk Tuk

International competition to stimulate sustainable design and entrepreneurship

Redesign of existing tuk tuk with new technology

To the participants of the competition: Couching To the end users: Training and maintenance

Existing

Existing and new chains

Rotterdam Innovation Lab

Incubation of start-ups

Old and newly designed products

Fostering sustainable innovation from young entrepreneurs

Adapted building

New organisation and cooperation

Table 1: Findings of the exploratory case study for Enviu Design

Consequences

Project

Activity

Product(s) type

Service

Infrastructure

Network of actors

Redesign of existing tuk tuk

Project development

Redesign of existing Thai tuk tuk based on European standards

Delivery of an alternative public transport for short distances within the city for a wide range of customers

Existing

Existing and new chains, strong links with universities

E-tuk

Project development

New design of an electrical tuk tuk

Delivery of an alternative public transport for short distances within the city for a wide range of customers

Existing and adapted (charging stations)

Existing and new chains, strong links with universities

Franchise scheme

Business model development

Redesigned and new products

PR, local marketing and training for entrepreneurs

Existing

New contracts with multiple actors

GoedWerk

Project development with a social mission

Redesigned and new products

PR, local marketing and training for refugees and disabled people

Existing

Existing and new chains

Table 2: Findings of the exploratory case study for Tuk Tuk Company

286

Design

Consequences

Project

Activity

Product(s) type

Service

Infrastructure

Network of actors

Evening Breeze bed

Project development and new venture planning

Redesign of existing cooling system and bed with an innovative use phase

Delivery and maintenance of the system

Existing

New chain, strong links with universities

Table 3: Findings of the exploratory case study for Evening Breeze 6 CONCLUSIONS This paper aimed at investigating how sustainabilitydriven entrepreneurs enhance the commercialisation of sustainable PSS and the consequences of this process and the potential role of design in it. For that reason, an explorative investigation has been made on three companies which are in the process of commercialising such innovations. The cases have been classified in a PSS-framework involving variables such as products, services, network of actors and infrastructural elements. The results show that all four variables might lead to a sustainable product-service system in different ways. For instance, as in the case of Sustainable Dance Club and Climate Innovation Lab, existing products (either used as they are or redesigned) in combination with new/adapted services may be developed as new business concepts which bring institutional and infrastructural innovation into the system. Complete new design of products making use of new technologies (E-tuk), new organisations/partnerships (Rotterdam Innovation Lab, GoedWerk), making use of existing business models (the franchise scheme of TTC) or bringing new ways of use through existing products (Evening Breeze bed) are all different opportunities for companies in creating sustainable innovations. In conclusion, this paper suggests that the required systems innovations for sustainability can be enhanced through a PSS strategy at an organisational level. This can be achieved through new entrepreneurial entries because of their potential in creating new solutions, building up the required scientific and business wise networks, commercialising sustainable innovations and consequently bringing the necessary institutional change that favours such innovations. 7 ACKNOWLEDGEMENTS The authors would like to thank A. Idil Gaziulusoy and Oriol Pascual for their valuable comments and contributions.

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17] 8 REFERENCES [1] Berchicci, L., 2005, The Green Entrepreneur’s Challenge, P.hD. Dissertation, Faculty of Industrial Design Engineering, Delft University of Technology. [2] Brouwer, J.S., Overeem, I., 2005, Enviu and sustainable development of natural areas, Enviu Paper series No. 1 [3] Elzen, B., Geels, F. W., Green, K., 2004, System innovation and the transition to sustainability: theory, evidence and policy, [4] Garcia, B.F., 2008, Design of a Sustainable Bar for the Sustainable Dance Club, Graduation Report, Faculty of Industrial Design Engineering, Delft University of Technology. [5] Hart, S., and Milstein, M., 1999, Global Sustainability and the Creative Destruction of Industries, Sloan Management Review, Fall 1999, 41, 1; ABI/INFORM Global, pp. 23-33.

[18]

[19] [20] [21]

[22]

287

Hoed, R. van den, 2004, Driving Fuel Cell Vehicles, P.hD. Dissertation, Faculty of Industrial Design Engineering, Delft University of Technology. Hogenkamp, N., 2008, Walls with ears for Sustainable Dance Club, Graduation Report, Faculty of Industrial Design Engineering, Delft University of Technology. Lamvik, T., 2002, Key Elements of Product Service Systems, Going Green Care Innovation, Vienna, November 2002. Linden M. van der, 2008, A tuk tuk for the western European city environment, Graduation Report, Faculty of Industrial Design Engineering, Deft University of Technology. Mont, O., 2002, Clarifying the Concept of ProductService Systems, Journal of Cleaner Production 10 (3) 237-245. Mont, O., 2004, Product-Service Systems: Panacea or Myth? P.hD. Dissertation. International Institute for Industrial Environmental Economics. Lund University. Parrish, B., 2008, Sustainability-Driven Entrepreneurship: A Literature Review, SRI Papers (Online) ISSN 1753-1330, No. 9. Randag A. 2007, Flux Floor - Energy conversing Dance Floor for the Sustainable Dance Club, Graduation Report, Faculty of Industrial Design Engineering, Deft University of Technology. Rotmans, J., Kemp, R., 2003, Managing Societal Transitions: Dilemmas and Uncertainties: The Dutch energy case-study, OECD Workshop Ryan, C., 2000, Design for Environment: Dematerializing Consumption Through Servicesubstitution is a Design Challenge, Journal of Industrial Ecology, Vol. 4, Issue 1. Schaper, M., 2002, The Essence of Ecopreneurship, Special Issue on Environmental Entrepreneurship, Greener Management International,No.38,pp.26-30. Schumpeter, J.A. (1934) The Theory of Economic Development (Cambridge, MA: Harvard University Press). Stroomer, E., 2007, New sustainable Concepts for the Clubbing Scene. Graduation Report. Faculty of Industrial Design Engineering. Delft University of Technology. Tischner U. and C. Vezzoli, 2004, UNEP Eco-design manual, PSS Module, UNEP (to be published). UNEP, 2003, Designing Sustainable ProductService Systems for All. UNEP, Paris. Versteeg, A., 2008, Entertain & Sustain – How to integrate Sustainability into Pop Culture? Graduation report. Faculty of Industrial Design Engineering. Delft University of Technology, 2008. Vezzoli, C., Ceschin, F., Kemp, R., 2008, Designing transition paths for the diffusion of sustainable system innovations: A new potential role for design in transition management? Changing the Change, Torino, 2008.

Environmental Impacts of Rental Service with Reconditioning – A Case Study R. Khumboon, S. Kara, S. Manmek, B. Kayis Life Cycle Engineering & Management Research Group, University of New South Wales, Sydney, Australia [email protected]

Abstract This paper presents a combination of selling service and life time extension strategy which has previously been proposed by others but with no supporting environmental impact assessment. It aims at evaluating rental service with reconditioning using a photocopier as a case study. Life Cycle Assessment was employed to provide a quantitative comparison of rental service with reconditioning and traditional product selling. The findings indicate that the reconditioning is potentially a promising way for improving the environmental performance of the rental service. However, it is premature to draw the general conclusion with further environmental impact measurement being required since the approach is very dependent on specific environmental impact factors. Keywords: Product service systems, Rental service, Reconditioning, Life cycle assessment

1 INTRODUCTION Over the past years, the global trend of environmental awareness is widely recognised. A growing number of businesses have consequently started providing integrated solutions of products and services. The concept of the Product Service System (PSS) has emerged as one of the elements of such solutions. A PSS has been defined as a system of products, services, supporting networks and infrastructure that is designed to be: competitive, satisfy customer needs and have a lower environmental impact than traditional business models [1]. It aims to provide companies with higher value adding process and to meet specific customer requirements, with reduced overall environmental impacts [2], [3].

environmental impacts may come from the transportation during the service period. Some kind of rental service such as the machine tool or photocopier requires the maintenance during the rental period which involves transportation between service provider and customer site. In an attempt to reduce the environmental impacts which occur during the life cycle, reconditioning is one of the strategies employed to reduce the end-of-life waste. The reconditioning is defined as the process of rebuilding or replacing major components of a product and returning it with a normal working condition [8]. Reconditioning can also be considered as a potential to extending the product life time and reducing the amount of material consumed [9].

Among types of PSS, the rental service is one of the possible strategies for a product-oriented company to move towards service-oriented company. Its underlying principle is that the remainder of the product ownership stays with manufacturers and provides the rightfulness for the usage of a product over a given period of time to the customer [4]. Thus the customer pays only for the use of the product when needed and need not be concerned about maintenance, repair or disposal of the product. In terms of the environmental perspective, the service provider who has the effective control of the product can extend its life span and improve the product utilisation [5].

Although the reconditioning was referred as a strategy for the improvement of the environmental performance, there have been limited quantitative measurements performed for this strategy. This paper aims at investigating the advantages and disadvantages of reconditioning in the rental service. The methodology starts by identifying the sources of environmental impacts in the general rental service. Subsequently, the case study of a photocopier rental service with reconditioning is evaluated and the Life Cycle Assessment is employed to quantify the environmental performance of the rental service.

However, the transition to the rental service can not automatically lead to the reduction in environmental impacts [6]. It requires further consideration of the environmental aspects for the whole life cycle. From the life cycle perspective, a product will be analysed in each phase along its life cycle, from material acquisition, component manufacturing, product assembly, customer use, and disposal. In general, the main source of the environmental impacts of the product life cycle will be from the material consumption or energy use [7]. However, in regard to the rental service, another source of

2 THE RENTAL SERVICE AND ENVIRONMENTAL ISSUES 2.1 Product Service System and Rental Service The Product Service System (PSS) approach was launched in Europe in the mid- 1990’s and focuses on delivering value to the customer throughout the product life cycle in an economically profitable, environmentally efficient and socially responsible manner [10]. An underlying principle of the PSS approach is the transfer of product ownership and responsibility to manufacturers,

CIRP IPS2 Conference 2009

288

and it provides the value of function or utility to the customer. There are three main categories of PSS; 1. Product oriented PSS, 2. Use oriented PSS, and 3.Result oriented PSS [11]. Rental service fits the second type of PSS i.e. use oriented PSS. In this category, the product ownership remains with the manufacturer who provides the usage of a product over a given period of time to the customer. Thus the customer pays only for the use of the product when needed and need not be concerned about maintenance, repair or disposal of the product. Although there are many rental services that exist in current businesses, not all of them consider the principles of PSS because the PSS oriented rental system takes environment into account in order to reduce the environmental impact. 2.2 Rental Service and Environmental Factors It is necessary to understand the environmental impacts of the rental system in order to find the appropriate method to improve their environmental performance. There are three generic types of services which effect environmental impacts [12]. 1. Type Alpha Services: In this type a service is provided in a fixed location and the customer travels to the service. 2. Type Beta Services: The service goes to the customer. In this type of service, a service provider performs the service at a customer location. 3. Type Gamma Services: Remote Provisioning. This group of service are provided without either the customer travelling to the service or their service travelling to the customer. Rather, the service is provided by electronic means such as internet, telephone. Most of rental services are drawn into the Alpha and Beta categories. Examples of product rental in Alpha type include laundrettes, computer cafes and video rentals. On the other hand, manufacturing machines, photocopiers, TVs, refrigerators and furniture rentals are examples of the Beta group. From the environmental perspective, Type Alpha and Beta service are relevant to the transportation in the rental service. There are two types of transportation which is the crucial source of environmental impacts. There are the transportation of delivering service and the transportation of maintenance. The following describes the environmental impacts of both types of transportation. 1. Transportation of delivering service. Transportation of service between service provider and customer is one of the main drivers of environmental impacts. The distance, frequency, and type of transportation are the factors which make the difference of environmental impacts of each rental service. The term of frequency of delivering service in the rental service means the frequency of change in customers. This frequency is usually determined by the term of the rental periods which are short and long terms. In short term rental periods, the product is allotted to many customers in sequential periods which make transportation high due to number of changes in customers. Therefore, in some situations, the short term rental makes more environmental impacts than long term rental. 2. Transportation for maintenance In general, technical maintenance is usually part of both Alpha and Beta types of services. The maintenance in Alpha type can be done in-house at the service provider’s site. However, the Beta service type requires service provider travel to do maintenance at the customer’s place.

In some circumstances, the environmental impacts of the rental service in the Beta type are higher than those in the Alpha type. There are two types of maintenance, preventive and corrective. The corrective maintenance includes activities pertaining to fixing the product which become out of order or broken. Preventive maintenance on the other hand, relates to actions of keeping equipment working and/or extends the life of the equipment. Some rental contracts require both preventive and corrective maintenance while some contracts need only corrective maintenance. Although the maintenance depends on the rental contract, it also relies on the characteristic of the products. Some complex machines such as airplanes or machine tools require experts or technicians to do both preventive and corrective maintenance. While some common machines such as TVs, furniture, or refrigerators only require corrective maintenance. The product which requires both maintenance activities generates the environmental impacts from the transportation between the service provider and customer site. 2.3 Rental Service and Reconditioning In order to reduce the environmental impact, the life cycle is an effective way to reduce environmental impacts during the product’s lifetime. Among numerous methods, strategies of repair, reconditioning, remanufacturing, and recycling of products are proposed to increase the product’s life. Figure 1 provides an illustration of ways for closing the material loop. Loop 2 in the figure is concerned with the recycling of materials, while loop 1 relates to the extension of the product life time. When the service provider controls the entire product life cycle, they can close loops and achieve reductions in waste. Loop 1 includes product reuse, repair, recondition, and upgrading hence is preferred because less energy is needed and the product life is extended. Therefore, the rental service with life cycle option with loop 1 is promoted to improve the environmental performance of service economy [13].

Figure 1: Closing the material loops of material recycling and product life extension [13]. Although the business models of product rental service with life time extension strategies are proposed [14], [15], [16], [17], there is no environmental assessment in this business model. This study aims at investigating the advantages and disadvantages of the reconditioning in the rental service. This is achieved by comparing of the rental service with reconditioning and traditional product selling. The photocopier rental service with reconditioning in Thailand was selected as the case study in this paper. Firstly, the photocopier rental service is classified in the Beta type of service where the service provider performs the service and both preventive and corrective maintenances are provided at the customer’s location. Secondly, reconditioning is preferred as a life time extension strategy. The next section explains the

289

methodology for the environmental assessment of the case study. 3 METHODOLOGY The photocopier case study was used as a case study to investigate the environmental impacts incurred during the life cycle of the rental service with reconditioning as shown in Figure 2. This environmental assessment was also emphasised on the resource efficiency and the environmental impacts during the reconditioning and maintenance phases. The Life Cycle Assessment (LCA) was used as a tool to quantify the environmental impacts of the service system using the Eco-Indicator 99 H/A method from the SimaPro 7.1 software. The environmental performance indicator is expressed in terms of a single score value which has a unit of points. In order to evaluate the efficiency of the reconditioning, the rental service with reconditioning scenario was also compared with the traditional selling product scenario. The functional unit of the two scenarios was the life cycle of a photocopier usage during 9 years as shown in Figure 3. The first scenario is the rental service with reconditioning, after the first use at USA of 3 years. The used photocopier is shipped to Thailand and reconditioned. Subsequently, the reconditioned photocopier is rented to a customer for the next 6 years under a rental contract of 2 years as shown in Figure 3. The second was the traditional selling product which the customer replaces anew photocopiers every three year during 9 years period. The expected outcome was to quantify the difference in material consumption of the two scenarios which reveals the efficiency of the reconditioning process. Moreover, the difference in the environmental impacts of the transportation involved during the maintenance of both scenarios can be

290

compared and analysed. As a consequence, the background of case study of photocopier rental service and the input data are given in the following section in order to provide the basis of this analysis. 3.1 Background of the company The company which provides the photocopier rental service in Thailand was selected as a case study in this research. In general, this rental service company is an organisation which has more than 5000 customers around Thailand. It imports second-hand photocopiers from USA and reconditions them in Thailand before providing the rental service to the customer. A rental contract of a copier is approximately 2-3 years. During the rental contract, the preventive maintenance and corrective maintenances are done to support the customer. The life cycle of photocopiers in this case study is presented in Figure 2 in order to explicitly illustrate different phases which can be briefly described as follows. Phase 1: First use at USA The customer at USA buys brand new photocopiers from manufactures. However, the copiers are normally replaced at three year intervals to avoid excessive breakdown of the equipment and high cost maintenance. After the end of first life, the photocopier will be sold in second-hand markets. Phase 2: Shipping The second-hand photocopiers from USA are shipped to the photocopier rental service in Thailand by shipping which takes approximately four weeks. Phase 3: Reconditioning The second-hand photocopiers are reconditioned in the

factory. In addition to this, the reconditioning process involves with an intensive inspection of the copiers where the major components that fail or have bad conditions are rebuilt or replaced. For instance, the covers of copiers are disassembled and repainted. Subsequently, the cleaning process and quality tests are made to the copiers in order to ensure that the copiers meet the standard requirements. Phase 4: Rental service After recondition process, the photocopiers are ready for providing the rental service to the customer. However, they are normally replaced at six year intervals to avoid high cost maintenance. 3.2 Input data The life cycle input data of the case study for the four phases as stated previously was collected on the basis of the company’s historical record. The input data of phase 1 includes the materials, manufacturing and transportation for the distributing and maintenance of a photocopier and the electricity consumption for the 3 years. Then, the input data of phase 2 was merely on the estimation of the distance in terms of tkm (tonne kilometre) of water transportation for the shipment from the USA to Thailand. Subsequently, the input data of phase 3 was focused on the materials for the replaced parts and the electricity consumption of the remanufacturing of the photocopier. Lastly, the input data of phase 4 included the materials of the spare parts, electricity during the usage of 6 years and the travel distance for the transportation involved in all trips of the preventive and corrective maintenance during the 2 years rental contract. The next section presents the comparison of the LCA results between the two scenarios.

selling product which is 43.07% higher than the material consumption by the rental service scenario with the figure of 78.78 points. Secondly, the environmental impacts of the transportation for the rental service as shown in Figure 4 produces 49.47 points whilst, the traditional selling product scenario generates only 1.29 points. This result shows although, the rental service with the reconditioning gains the environmental benefit from the material reduction as discussed previously; but its environmental impacts from the transportation are significantly high. Thirdly, Figure 5 illustrates the environmental impacts of transportation in the rental service with reconditioning scenario. It can be clearly seen that the transportation during the rental period has the largest environmental impacts of 31.43 points. The second highest contribution was the 17.29 points of the copier shipment from the USA to Thailand. Moreover, 0.42 points of the environmental impacts of transportation during the first life cycle at USA is substantially less that the rental service in Thailand which has a figure of 31.43 points. Therefore, the substantial amount of the environmental impacts for the rental service scenario was found from the transportation activities during the rental period in Thailand. This was due to the increasing of the transportation involved in the preventive and corrective maintenance during the rental service. Single Score (points) 40 31.43 30

17.29

20

4 LIFE CYCLE ASSESSMENT RESULTS 4.1 The results from LCA analysis

10

The results from LCA analysis on the case study are concluded into following four key points. Firstly, Figure 4 presents the single score of the two mentioned scenarios in terms of material consumption, transportation, energy consumption during the usage life cycle stage and the total single score for the entire life cycles. As it can be seen, the total single score results show that the rental service contributes 25% less environmental impacts than the traditional selling product. This was due to the significantly high environmental impacts from the material consumption of the traditional Single Score (points)

First life cycle at Shipping from Reconditioning at Rental service at USA USA to Thailand Thailand Thailand

Finally, the comparison of environmental impacts between preventive and corrective maintenance is presented in Figure 6. The environmental impacts of transportation in the corrective maintenance are 14.39 points whereas 17.05 points were produced by the preventive maintenance. Similarly, the environmental Single Score (points)

250

219.3

25

182.91 165.2

20

150

17.05 14.39

15

100

Life cycle phases

Figure 5: Environmental impacts of transportation during the life cycle phases of the rental service.

300

200

0.33

0.42 0

10.82

78.78 49.47

50

36.97 35.1

10 6.16

1.29

5

0

Material

Transportation

Rental service with reconditioning

Energy

Total 0

Traditional selling product

Figure 4: Environmental impacts of traditional product selling and rental service with reconditioning in terms of material, transportation, and energy.

Preventive Maintenance

Transportation

Corrective Maintenance

Material consumption

Figure 6: Environmental impacts of transportation of the preventive and corrective maintenance in rental service.

291

impacts of the corrective maintenance for the material consumption aspect are 6.16 points while 10.82 points was found in the preventive maintenance. Therefore, Environmental impacts from transportation of the corrective maintenance contribute almost as high as those from the preventive maintenance. 5 DISCUSSION The reconditioning in the case study of photocopier rental service presents the improvement in environmental performance in terms of a reduction in material consumption. The rental service with reconditioning generates only 43.07% of the environmental impact of traditional product selling. Therefore, reconditioning provides benefits in terms of material consumption for the rental service in the case study. The case study also shows that the environmental impact from transportation in the rental service are significant, compared with those in traditional product selling. Reconditioning resulted in a reduction of 54.1 points while the increase in transportation due to maintenance service resulted in an increase of 31.01 points. Therefore, even with the impact of transportation in service, reconditioning still has the potential for improving the environmental performance of the rental service. However, the result presents that the environmental impacts from shipping the photocopier from USA to Thailand are considerable due to the long distance of transportation. The huge environmental impacts negate all improvements from the reconditioning. Although there is a possibility of the searching for new suppliers from nearby countries, other factors such as the availability, condition of the products and policies should be taken into account prior to such making decisions. Another important aspect from this case study is that it is not enough to rest the environmental improvement only with the reconditioning. The optimization of all environmental aspects in every life cycle phase is also significant. For example, the product in the case study is in the product group which consumes energy during the use phase. Thus, the energy saving technology is also important for the reduction in environmental impacts during the use phase. It implies that reconditioning can not extend to the product’s life cycle indefinitely. It should be limited by upcoming new energy saving technologies. The product replacement with new technologies or the component upgrading is more suitable in such situations. In addition, factors of deterioration of components which affects the increase in energy consumption over time should be taken into account. A final point of discussion centres around maintenance during the rental service. This case study shows that the environmental burdens of corrective maintenance are almost as high as those from preventive maintenance. Although the maintenance is quoted in the rental agreement, the corrective maintenance can be reduced. One possible way to decrease the environmental impacts from corrective maintenance is the effective management of preventive maintenance. The principles of preventive maintenance such as the components replacement before the failure should be applied. Furthermore, preventive maintenance scheduling should consider incorporating the transportation of many machines at the same time. 6 CONCLUSION The rental service is one of the various types of the Product Service System. Combining rental service and reconditioning is proposed as a means for improving 292

environmental impacts. This paper presented the environmental impacts of the photocopier rental service with reconditioning. The results show that the reconditioning is a promising way of improving the environmental performance of the rental service, despite the high environmental impact caused by the transportation from maintenance. However, it is too early to draw the general conclusions that the reconditioning has potential to reduce the environmental impacts. This is because there are other factors that cause environmental impacts such as product type, component characteristics and usage of customer. Therefore, it requires more examples and further research to generalise such conclusions. Another finding in this paper is that the transportation for maintenance between service provider and customer is a significant source of environmental impacts. The rental services such as machine tools and photocopiers require both preventive and corrective maintenance from the service provider. The distance, frequency, and the type of transportation are main factors of the environmental burdens in the maintenance service. Therefore, it is important that all of the environmental factors in every phase of the product life cycle should be considered in order to improve the environmental impacts of the rental service. 7 ACKNOWLEDGMENTS The authors would like to express our sincere gratitude to staffs at the company who support the valuable data for this study. The warmest thanks also go to our colleagues at the Life Cycle Engineering & Management Research Group at UNSW, Australia for their helpful suggestions. Lastly, special recognition must be given to the UNSW and Dhurakij Pundit University, Thailand for their financial assistance. 8 REFERENCES [1] UNEP, 1997, Sustainable Service Design, United Nations Environment Programme, New York: UN Publications. [2]

Mont, O., 2000, Product Service-Systems, Final Report, IIIEE, Lund University.

[3]

Wong, M., 2004, Implementation of Innovative Product Service Systems in the Consumer Goods Industry, PhD Thesis, Cambridge University.

[4]

Tukker, A., and Tischner, U., 2006, New Business for Old Europe, Sheffield, UK, Greenleaf Publishers. Lamvik, T., 2001, Improving Environmental Performance of Industrial Products through Product Service Systems, Norwegian University of Science and Technology, Mont, O., 2004, Product Service-Systems: Panacea or Myth? PhD Thesis, Lund University, Sweden. Kaebernick, H. and Soriano, V., 2000, An Approach to Simplified Environmental Assessment by Classification of Products, Proceedings of the 7th International Seminar on Life Cycle Engineering, The University of Tokyo. King, A.M., Burgess, S.C, 2005, the Development of a Remanufacturing Platform Design: A Strategic Response to the Directive on Waste Electrical and Electronic Equipment, Proceedings of the MECH E

[5]

[6] [7]

[8]

Part B, Journal 623-632. [9]

of

Engineering

Manufacture:

Parkinson H J, 2003, Thompson G, Analysis and Taxonomy of Remanufacturing Industry Practice,

Proceedings of the MECH E Part E, Journal of Process Mechanical Engineering, 217(3): 243-256. [10] Mont, O., 2002, Clarifying the Concept of Product Service Systems, Journal of Cleaner Production, 10(3): 237-245. [11] Tukker, A., 2004, Eight Types of Product Service System: Eight Ways to Sustainability? Experiences from SusProNet, Journal of Business Strategy and the Environment, 13(4): 246-260. [12] Graedel, T E., 1998, Life-Cycle Assessment in the Service Industries, Journal of Industrial Ecology, 1(4), 57-70. [13]

Stahel, W R., 1997, the Service Economy: ‘Wealth without Resource Consumption’? Philosophical Transaction: Mathematical, Physical and Engineering Sciences, 355(1728): 1309-1319.

[14] Mont, O, Dalhammar, C, Jacobsson, N, 2002, A New Business Model for Baby Prams Based on Leasing and product remanufacturing, Journal of Cleaner Prod., 10(3): 237- 245. [15] Sundin, E., and Bras, B., 2005 Making Functional Sales, Environmentally and Economically Beneficial through Product Remanufacturing, Journal of Cleaner Production, 10(3): 913- 925. [16] Besch, K, 2005, Product Service Systems for Office Furniture: Barriers and Opportunities on the European Market, Journal of Cleaner Production, 10(3): 1083-1094. [17] Williams, A., 2007, Product Service Systems in the Automobiles Industry: contribution to System Innovation? Journal of Cleaner Production, 15(11): 1093-1103.

293

The Practical Challenges of Servitized Manufacture T. Baines, H Lightfoot Cranfield Innovative Manufacturing Research Centre, Cranfield University, U K Abstract Servitization is now widely recognised as the process of creating value by adding services to products. Since this term was first coined in the late 1980s it has been studied by a range of authors who have specifically sought to understand the methods and mechanisms of service-led competitive strategies for manufacturers. This paper reports on the experiences of a large company as they have moved towards servitized manufacture. This has been based on an extensive series of interviews with key personnel. The results of the study and implications for research are all reported. Keywords: Manufacturing, Service, Case-study

1 INTRODUCTION Servitization is now widely recognised as the innovation of an organisation’s capabilities and processes, to better create mutual value, through a shift from selling product to selling Product-Service Systems [1,2]. Such a strategy is now widely advocated as a means by which western manufactures can face-up to the challenges of competitors in lower cost economies. However, few researchers have documented the associated consequences to the organisational design of the host manufacturer as they seek to pursue such a strategy. Therefore, this paper describes the experiences of a typical UK based manufacturer as they have adopted a servitization strategy. The paper describes case based research that has sought to gain an in-depth and multi-disciplinary understanding of the implications of a servitization strategy. The targeted company (which we refer to under the pseudonym of ServCase) is a large manufacturer that, through the successes of integrated products and services, now generates a large portion of revenue from product-centric service contracts (i.e.: services that are tightly coupled to the product offering). Our research with ServCase has helped us to appreciate how servitization necessitates companies to make modifications ranging from the language they use to interact with customers, though to their organisation design. Such experiences, along with background to this topic and our research design, are all presented in this paper. 2 THE CHALLENGE OF SERVITIZATION 2.1 Defining servitization and Product-Service Systems The first use of the term servitization came in 1988 from Vandemerwe and Rada in their article in the European Management Journal titled ‘Adding Value by Adding Services’[3]. Here they defined servitization as the increased offering of fuller market packages or ‘bundles’ of customer focussed combinations of goods, services, support, self-service and knowledge in order to add value to core corporate offerings. Rightly or wrongly the terms service and product (goods) are intrinsically linked to discussions about servitization. The term ‘product’ is generally well understood by manufacturers. As Goedkoop [4] defines, a product is a tangible commodity manufactured to be sold, and quite simplistically is

CIRP IPS2 Conference 2009

294

capable of ‘falling on your toe’ and of fulfilling a user’s needs. Invariably, in the world of manufacture, it is usually considered to be a material artefact (e.g.: car, boat, plane). Conversely, we will consider that services are an economic activity that does not result in ownership of a tangible asset. The concept of a Product Service-System (PSS) is a special case of servitization. A PSS can be thought of as a market proposition that extends the traditional functionality of a product by incorporating additional services. Here the emphasis is on the ‘sale of use’ rather than the ‘sale of product’. The customer pays for using an asset, rather than its purchase, and so benefits from a restructuring of the risks, responsibilities and costs traditionally associated with ownership. Similarly, the supplier/manufacturer can improve their competitiveness as these ‘solutions’ may be clearly differentiated from product based offerings while, simultaneously, retaining asset ownership which can enhance utilisation, reliability, design and protection. 2.2 Previous servitization research Since this term was first coined servitization has been studied by a range of authors (e.g Wise & Baumgartner, [5]; Oliva & Kallenberg, [6]; Slack, [7]) who have specifically sought to understand the methods and implications of service-led competitive strategies for manufacturers. In addition, and somewhat independently, during this same period there has been a growth in research on the related topics of Product-Service Systems (PSS), Service-Science (SS) and Integrated Vehicle Health Management (IVHM). This increasing body of research indicates a growing interest in this topic by academia, business and government. One reason for this is the belief that a move towards servitized manufacture is a means to create additional value adding capabilities for traditional manufactures. Furthermore, that such services are distinctive, long-lived, and easier to defend from competition based in lower cost economies. Indeed, many governments see such moves downstream as key to competitiveness (see Hewitt, [8]). As a consequence, more and more western manufacturers are seeking an ever increasing percentage of their revenues from services [5]. However, there is some concern that servitized manufacturers could be in greater danger of bankruptcy and make lower returns in the longer-term [9].

Nevertheless, it is difficult to argue against a careful adoption of some services in certain situations. To succeed with servitization a manufacturer is likely to need some new and alternative organisational principles structures and processes [6]. These may be different to those associated with traditionally product manufacture. For example, it may be insufficient to simply attempt to replicate the Lean principles of Toyota. This is an area of some contention amongst scholars as the adoption of Lean is often seen as the solution to tackling the poorer performers in the services sector. While this may be appropriate in some instances, authors such as Chase [10] argue strongly for reversing the trend of applying operational management based concepts in the services environment. They suggest that there is a subtle mix of organisational structures that are appropriate to a servitized manufacturer that are distinct and different to those associated with, either a more traditional product manufacturer, or a pure service provider. However, researchers have yet to fully understand the nature of these structures and their associated issues. 3. RESEARCH DESIGN The aim of the research presented in this paper has been to gain a deeper understanding of the issues that arise when a servitization strategy is followed in real-life. Specifically, we have set out to investigate a “servitized organisation” that designs, builds and delivers integrated product and services, and to identify the challenges they are encountering in the pursuit of such a strategy. Our method has been to carryout an in-depth and multidisciplinary case-study analysis. The choice of case

company was critical to this study, as we sought to investigate a manufacturer who has a track-record of achieving business success through providing a portfolio of product related services. Therefore, our case study organisation is a UK based OEM that designs and manufactures high value capital equipment for the power, defence and aerospace markets. For reasons of confidentiality and in order to give us greater freedom to discuss our results and findings, we refer to the company as ‘Serve Case’. The company, which operates globally and today generates over 50% of revenues from the provision of services that are closely coupled to its products. Whilst ServeCase continues on its servitization journey, it is sufficiently advanced to provide a basis for exploring the characteristics of an operations strategy in this evolving context. This case study has taken place from June-November in 2007. During this time we have worked with seasoned researchers from across the disciplines (engineering, manufacturing, management). Our approach has been first to develop a history map with ServCase to understand how they have arrived at their servitization strategy. Then working in pairs, researchers have conducted interviews with key personnel from across the organisation (e.g. marketing, customer support, engineering, manufacturing operations, and supply chain) and captured their views on how the organisation operates and the issues they are facing. Each interview was recorded and then transcribed. These were then analysed using mind-mapping techniques to identify the common issues arising across the organisation. It is these issues that are the principal findings of this study, and these are presented in the following section. 4. SUMMARY OF KEY FINDINGS 4.1 Emergence of Servitization at ServCase

ServCase provides capital equipment products, and often offers these with a broad range of services that ensure asset availability via a risk and revenue sharing contract. While the origin of this business dates back to the early 1990s it really only took shape in the early 2000s. This market proposition emerged in response to customers who sought to offset their repair and overhaul costs and responsibilities for products. Similarly, ServCase sought to prevent component suppliers attacking their lucrative aftermarket. To counteract these significant threats, ServCase’s initiated a series of Joint Ventures (JVs). Since then, their services business has grown to such an extent that over 50% of company revenues are now derived from such contracts. The changes issues that have arisen as a consequence of this transition from traditional manufacturer are summarised as follows. 4.2 Language is particular and peculiar One of the most striking differences at ServCase is the everyday language used by the employees in the delivery of services. Whereas with a conventional manufacturer personnel use (and fully understand) nomenclature such as product, part and component they may only loosely understand the term service. As a noun, the word ‘services’ usually refer to the offering (e.g.: maintenance, repair, insurance) and a single offering is a service. However, as a verb, service can also be used to refer to a level of performance (e.g.: the company provided good service). This is only one example of may words and phrases whose semantics take on particular and specific meanings. This distinction appears strongest amongst personnel who deal most closely with customers of services. Future challenges are, therefore, to make such language pervasive throughout the organisation. 4.3 Value dimensions are special One reason that language changes, is that a servitized product offering is different. At ServCase the nature of the boundary with the customer changes from being bias towards transactional to that of relationship. Traditional manufacture tends to take a linear view of product production (by the manufacturer) which is then sold (a transaction) to the customer for their use (consumption). However, when ServCase deliver an integrated product and service there tends to be a series of ‘touch points’ between the host business and customer. For example, initial contract negotiation may be lengthy; monitoring of the asset in use may be carried out by the business; this may lead to servicing of the product by the business; and finally the host may take-back the product at end-of-life. While the product itself may still be sold to the customer (as is the case with ServCase) the associated services are more closely associated with a relationship business model. Hence, servitization tends to combine both transactional and relationship business models and, most importantly, revenue, profits and cash flow arise mainly from the relationship aspects of this model. The metrics used to define value offered to the customer vary to reflect the changing business model. This is particularly apparent in the measures employed to assess performance. Conventional manufacture will frequently focus on QCD (Cost, Quality and Delivery) associated with product families. Here, quality conformance will typically be assessed in terms of reject components; cost will be comprise of labour, materials and overheads; and delivery performance will tend to be assessed in terms of due-date performance. With services at ServCase, value becomes more associated with asset use, rather than sale or repair, then the appropriate measures can be subtly different. Quality becomes associated with reliability of services in the field; cost can include

295

penalties incurred through asset down-time; and leadtime may become more critical. Here, a future challenge is to precisely define, distinguish and communicate the key performance measures 4.4 Products and design process are different As the value proposition changes, then product designs at ServCase have also altered to reflect the balance of value gained through asset use rather than more simply artefact ownership. As mentioned above, ServCase sell their product, and offer complementary services to assure asset availability. As significant revenue is generated through services, their products incorporate a facility for remotely sensing performance in the field. Here, extra cost is added to product manufacture which can not be recouped at point-of-sale, but rather relies on the customer taking-up the service contracts offered. This is typical of the many product features that are introduced to aid maintenance and servicing to support asset availability in the field. Design process also differ. Traditionally, product designs are conceptualised remotely, prototyped and refined, and then put into practice. With services, prototyping tends to take place through application. Here, one danger is, as ServCase have found, for engineers to attempt to apply conventional product design processes. Understanding more about how these processes differ is a considerable future challenge. 4.5 Integrating service and product delivery systems is challenging As with product designs, the organisational design required to support the value proposition also changes. The conventional view of materials flowing into a factory, through production, to be consumed by the customer is does not adequately describe ServCase. While a small portion of this somewhat uni-directional material flow does occur, there is also super-imposed a complex service delivery system that monitors and supports the asset in use. This system transcends the traditional internal / external barriers of the host business; instead calling on partners and suppliers to affect the delivery of the required service. This delivery system is directly impacted by the relational component of the business model and associated performance measures (as outlined in section 4.3). These requirements are so particular, that ServCase has decoupled this delivery mechanism from their more conventional production system. However, they recognise that as business pressures increase, sharing of resources and knowledge, are likely to necessitate these systems to be more tightly integrated. How to achieve this is a topic of some debate within the organisation. 4.6 Transformation issues are both particular and pervasive ServCase illustrates a manufacturer that, in the adoption of a servitization strategy, is encountering changes to language, value, along with designs of product and organisation. Throughout this case, time and time again, we have been made aware that one of the biggest challenges that ServCase are facing is transformation. Sections 4.2 – 4.4 above summarise how across the organisation and its broader supply chain ServCase has changed, and continues, to change. Against each of these strands ServCase is defining new design paradigms, and each of these introduce particular challenges to the mind-sets of customers, employees, and suppliers. For example, educating employees in the language of service, changing process to better suit the nature of service design, and adopting integrated product and service delivery systems. Understanding the specific

296

transformational issues, and how to overcome these, is a principal future challenge.

5. CONCLUDING REMARKS ServCase is one example of a UK company that has adopted a servitization strategy. Our work here has given us a much clearer understanding of the particular issues that are arising as ServCase attempts to deliver integrated products and services successfully. In brief, these are: 

Language used in service is particular and peculiar.



Value dimensions are special and biased towards relationships rather transaction.



Products and design process are different and better enable service support.



Integrating service and product delivery systems is challenging.



Transformation issues are both particular and pervasive throughout customers, employees, partners and suppliers. There is little to suggest that these issues are particular to the ServCase business or sector. However, for completeness, our future work will now look externally to this organisation to carryout a complementary investigation of the suppliers, partners and customers of ServCase In conducting such an investigation we look forward to further developing our understanding of the challenges faced through servitization, and reporting these in future papers. 6. ACKNOWLEDGEMENTS We wish to acknowledge the support of the Engineering and Physical Sciences Research Council for their support in carrying out this work. Also, we are grateful to our colleagues Professors Evans and Neely for their work in setting-up and managing the case study on which this paper is based. 7. REFERENCES 1. Baines, T et al (2007) ‘State-of-the-art in Product Service-Systems’ Proc. IMechE Part B, Vol 221: Journal of Engineering Manufacture, 1543-1552. 2. Aurich J et al, 2007, Advances in Lifecycle Engineering for Sustainable Manufacturing Businesses, Proceedings of the 14th CIRP Conference on Lifecycle Engineering, Tokyo, Japan 3. Vandermerwe S & Rada J,,1998,‘Servitization of Business: Adding Value by Adding Services’ European Management Journal. Volume 6, No 4, 314324 4. Goedkoop M et al, 1999, Product Service-Systems, Ecological and Economic Basics, Report for Dutch Ministries of Environment (VROM) and Economic Affairs (EZ) , 5. Wise R & Baumgartner P. September-October1999, ‘Go downstream: The New Profit Imperative in Manufacturing’ Harvard Business Review,77,5 133141. 6. Oliva R & Kallenberg R. 2003 ‘Managing the Transition from Products to Services’International Journal of Service Industry Management 14, 2, 160172 7. Slack N.. ‘Operations Strategy: 2005, ‘Will it ever realise its potential’ Gestao & Producao 12, 3, 323332

8. Hewitt P, Secretary of State for Trade and Industry, 2002,4, The Government’s Manufacturig Strategy 9. Neely A, 2007, ‘Servitization of Manufacturing’ 14th EurOMA Conf. Ankara, Turkey

10. Chase R & Garvin D,July-August 1989 ‘The Service Factory’ , Harvard Business Review, 67,4,61-69.

297

Challenges for Industrial Product/Service Systems: Experiences from a learning network of large companies E. Sundin1, G. Ölundh Sandström2, M. Lindahl1, A. Öhrwall Rönnbäck1, T. Sakao1 and T. C. Larsson3 1 Department of Management and Engineering, Linköping University, Sweden 2 Department of Machine Design, School of Industrial Technology and Management, Royal Institute of Technology, Sweden 3 Division of Functional Product Development, University of Technology, Luleå, Sweden [email protected]

Abstract In Sweden, there are a growing number of manufacturers that are using the approach of industrial product/service systems. This paper explores how manufacturers and university researchers have started a workshop series where important and topical product/service system issues are elucidated. The companies face many challenges in order to achieve a good product/service system business. Many challenges are related to changing different peoples’ mindset within the company and/or with external companies and customers. Having a learning network approach of dealing with these challenges has been perceived as a good manner of tackling the questions raised within the product/service system providing companies. Keywords: PSS, Integrated Product Service Engineering (IPSE), IPS², Learning networks

1 TRANSITION TO PRODUCT/SERVICE SYSTEMS Manufacturers today regard service activities as increasingly important; as a result, a number of them are shifting their focus from “product seller” towards “service provider” [1]. Sakao et al. [2] conclude that the main drivers for this changed focus are customer connection, customer demand and increased competition. The transition to product/service systems places new and demanding requirements on product and service development and production, along with new requirements for companies in the way they relate to and build up relationships with customers. Previous authors’ research shows that existing product/service offerings are developed by the companies’ marketing departments and based on existing products optimized for traditional sale [2-4]. With product/service offerings, the skill to combine different types of products and services into a desired function becomes more crucial. In order to be able to deliver, companies need to continually develop their value chains and the competence of their staff [5, 6]. This also implies that creating increased value for the customer is more in focus [7, 8]. The offering of product and service offerings is not a new concept per se, although it is a new concept for several manufacturing companies. Research shows that the transition from being focused on selling products to becoming a more service-oriented company is a process filled with possibilities and difficulties, to include organizational and financial issues and the ability to manage a new relationship with the customer [3]. In parallel to these trends, concepts such as Functional Sales [9] are readily found, not only as theory but also in practice in several industries. Other related concepts include Total Care Products (Functional Products) [10, 11], which comprises combinations of hardware and support services, Product/Service Systems (PSS) [12], Integrated Solutions [13], Service/Product Engineering (SPE) [14], and Industrial Product/Service Systems (IPS²)

CIRP IPS2 Conference 2009

298

[15]. In common to these concepts, service activity is beginning to be increasingly incorporated into the design space, an area which has been traditionally dominated by physical products in manufacturing industries. In this paper, we will further on use the term PSS when describing above concepts. 2 OBJECTIVE The objective of this paper is to present how large manufacturing companies continuously can improve their PSS business through learning networks. How the learning network was organized, and how the companies’ current challenges relate to PSS, are also described. 3 RESEARCH METHODOLOGY The authors of this paper have more than eight years’ experience in PSS research, in which small, medium and large companies have been involved. The authors’ PSS research builds on research performed in different research disciplines, such as: environmental research, product development, remanufacturing and business models for industrialized services. In the research performed over the years, a plethora of research methods have been used. However, qualitative research interviews [16] have been the most frequent method for collection of primary data. We have interviewed various functions, e.g. those of CEOs, designers, and service developers, but also of customers. The learning network described in this paper is not really a traditional research project, but rather a way for researchers to contribute to and support companies’ continuous improvement of their work with developing and offering PSS. In this research, the role of the researchers is in this case toned down; rather, the themes of the network meetings are mainly decided by the companies,

and the companies take turns hosting the meetings. The researchers’ role is to act as discussion partners and to provide knowledge. From a research perspective, it is very useful to follow the discussions and learn more deeply about the challenges the companies are facing and exploring, or to obtain greater depth in the research topics. So far a network start-up meeting and two learning network workshops have been held according to Table 1 below: Table 1. Participants at the learning network workshops Company participants

University participants

Start-up meeting

12

6

First Workshop

8

4

Second Workshop

9

5

Workshop

The workshops in the network have been well documented, by recording and taking detailed notes of discussions and copies of the companies’ presentation slides. The methodology used for the learning network is further described in the following sections. 4

modeling development work [23]. The fundamental ideas for a learning network are partly based on the model of “the cycle of experimental learning” by Kolb [24]. That model promotes the practitioners’ need to disconnect from an action-experiencing loop by getting support to learn from their experiences through self-reflection, but also by discussing their practice in a more conceptual way and getting input from others. All together, this gives them the possibility to act according to new insights. In a learning network, company representatives as well as researchers are participants, as illustrated in Figure 1. The company representatives bring in their own experiences and share their knowledge with other company representatives, but also with researchers. The researchers then contribute with reflections, and can provide input based on their knowledge.

MGMT

4.2 Principals of a learning network One way of increasing the interaction between researchers and companies is to engage both parties in a learning network. By doing so, researchers get access to and in depth knowledge of companies’ current business logic and the challenges they are facing. One of the main benefits of a learning network approach is that it supports organizational change and is useful for

Company 2

Company 1 Research group

Company 3 MGMT

A LEARNING NETWORK APPROACH

4.1 Theoretical issues are good but… Much research in the PSS area has been more focused on theoretical issues, and on making conceptual proposals for how PSS can and should work (see e.g. [17]), than practical ones. For example, numerous papers have focused on the theoretical and potential benefits of PSS, e.g. from an environmental, marketing, political economic or societal perspective (see e.g. [1, 18]). However, researchers need to have a strong understanding of practical experiences of PSS, something that requires close interaction with the companies themselves. In parallel, and already described in this paper’s introduction, researchers have developed a large number of concepts and definitions of concepts meaning more or less the same thing. This is in parallel with the number of concepts used in industry, e.g. leasing, rental contracts, and pay-per-service-units (for more see Lindahl et al. [19], which taken together have blurred and obstructed the dialogue, especially with the industry. The authors’ experience is that it is important to be selfcritical; i.e., what is important is that which brings the research forward and develops the PSS area. Is that to focus on finding the one and only concept with the one and totally perfect definition and to define all theoretical potential benefits with PSS? Our conclusion is not; instead, we have tried to mainly base our research on a close co-operation with companies that already have or are in the process of starting to sell PSS. This has been done in a number of projects, e.g. [1, 20-22].

MGMT

Company n MGMT

Figure 1. An illustration of a Learning Network [23]. According to Ritzén et al. [23], the goals for companies to participate in the Learning Networks are to: 1) gain theoretical insights and new knowledge 2) share experiences 3) have time and space for reflection 4) gain motivation to overcome organizational barriers for change 5) get support in process management Learning networks can be used in different ways, enabling researchers to be more or less active in the company’s change process. In a learning network that the IPSE research group performed with SMEs, the purpose was twofold – both to develop the IPSE methodology and to develop the PSS mindset among the companies. In that project, the learning network approach was part of the IPSE methodology for supporting SMEs in understanding and developing integrated product and service offerings. In the project, a workshop series was developed with focus both on understanding a new business model and learning how that affects the company, including product development and the design of the products. The series consisted of 5 workshops developed by the researchers based on the needs of the companies. The workshop development was an interactive process, and since the companies were divided into three networks, the researcher had time to reflect and make improvements and changes [20]. During the workshops, the researchers participated by discussing different themes with the companies and guiding them in how they could develop integrated products and service offerings. Participating researchers have also been responsible for the documentation.

299

At the workshops, knowledge has been shared and transferred between the participating companies and the researchers. The positive experiences from the learning networks with small companies, along with the expressed need from larger companies, led to a formation of a learning network for larger companies. 5

LEARNING NETWORK FOR LARGE COMPANIES FOR CONTINUOUS IMPROVEMENTS In our contact with representatives from large, primarily Swedish companies offering PSS, or companies at the starting gate for doing so, they have often expressed that they are feeling quite “lonely”. Some reasons expressed for this are e.g.: 

Since PSS is often a quite new phenomenon within their companies, the number of colleagues is often quite small since they are just in the beginning of building up their PSS capacity. In some cases, they have no colleagues to discuss their problems and solutions with.



PSS is often treated as an “odd duck” by the traditional parts of the organization. One reason for this is that the concept is new and differs from the traditional focus, around which the company’s structure often is built upon.



Since the concept is often new, they also lack history and examples to use when tying to persuade others within the company. Furthermore, since PSS is quite new, an important argument is to be able to point out that there are a number of other companies, in various branches, that also have invested in the PSS area.



Finally, but not least and related to above, when they are at their companies, they are often very focused on their work and simply lack time for reflection. To meet others with the same type of work stimulates reflection. Representatives from larger companies have wished for or asked for a forum were they could meet other companies in order to learn from each other, and to build up a network of colleagues since they don’t have this within their own organizations. They have further argued that traditional one or two-day seminars with speakers were seldom enough and too speaker oriented, e.g. not really based on the actual needs of the companies This network was imitated by the authors of this paper and based on the IPSE project’s state-of-the-art analysis in combination with that explained in 4.3, namely a need for meeting representatives from other PSS companies. This was seen as a compliment to the work with SME companies. To strengthen the network, the IPSE project’s researchers also invited other Swedish PSS researchers to form the group responsible for the PSS network for larger companies developing PSS offerings. The first activity was to summon a start-up meeting to which the researchers invited interesting and interested companies. The learning network for the large companies had the purpose of supporting continuous improvements. 5.1 Network Start-up At the first start-up meeting on the 29th of November 2007, there were 8 companies represented by 1-2 persons from each company. A number of representatives from other companies were genuinely interested in participating in the network, but for various reasons could not participate. The persons from the companies were, for example, aftersales managers and business developers. In addition, the companies all had a strong wish to develop their service development and introduce new PSS offerings.

300

Also, six researchers participated that had different expertise such as eco-design, economics, and development of services and products; however, they all had the mutual interest of the development of products and services. The aim of the first meeting was to get a deeper understanding about participating companies’ expectations, needs and requirements for the network. The meeting was led by a research funding member with experience and interests in the PSS design research area. After a short introduction from the companies, the main focus was on discussions about the network objective and future workshop themes. The following paragraphs describe the results of these discussions. Incentives for participating and aim of network The companies’ aim with service development was expressed as obtaining increased profitability by supporting their customers to increase their value and to reach their goals. All of the participating companies agreed that the perspective of customer value is central in service development. However, the experience of service development varied among the company members, and the field is still somewhat new to them. One of the incentives for participating in the network was to get the opportunity to learn from each others’ experiences, thereby avoiding drawbacks and facilitating success. The companies all had challenges to attend to in order to achieve a well-functioning PSS. The network meetings were a place where they could reflect, get inspiration and learn. A crucial aim for the companies was to obtain knowledge transfer between the companies; this could, for example, include discovering how PSS problems have been tackled by other companies. Furthermore, the problems could be discussed among the participants, and solutions could be found which would make the companies more competitive in their respective markets. The network would be a learning network where knowledge is transferred and new knowledge is developed. The researchers would contribute with a general view of the problem solving, while the companies could provide company-specific data. Besides the knowledge transfers, the network would be an arena for developing new research ideas and developments for new PSSs. PSS Issues and tentative workshop themes At the meeting, future meeting themes were discussed by firstly mapping out the questions the participants wanted to have answered. The questions also represent challenges that the companies were currently facing. 

How different markets are differentiated regarding prerequisites and possibilities for PSS, i.e. both geographical and product markets?



How can and should product and service development be integrated to increase internal and external efficiency?



How can one earn money on an already installed product base – and how then should offerings and business models be developed?



How can one change the customers view on the value of competence, and get paid for that competence?



What should the actual product/service portfolio look like?



How should the delivery systems for services and offerings be developed?



How can the internal organization be brought on board – and how can one develop concrete visions and

motives for developing, marketing and selling more services? 

How can new technology be used for service development?



How should the organization, control and salary systems be developed internally and for retailers?

Furthermore, the incentives for having a PSS business approach were discussed to be manifold, from economic to environmental benefits to the efficient use of physical products. However, it was a consensus among the company participants that it is mainly the business potential of PSS that has been making PSS more interesting to invest in. Based on the questions that the companies wanted to elucidate, a grouping and listing of coming workshop themes was posted. The themes were discussed in front of all meeting members, and a ranking of importance was made. Based on this ranking, the themes of “customers and development” and “new customer relations” were selected for the first workshop meeting. Furthermore, it was decided to have the workshops arranged at the participants’ companies in order to obtain more knowledge about the arranging company’s PSS business and to alternate the travel distances for the participants. The start-up meeting was held at Linköping University in Sweden. 5.2 First network workshop At the first workshop held on the 24th of April 2008, there were 8 company participants and 4 researchers. At this time, several company representatives could not join, but all of them were very clear on the fact they wanted to belong to the network and participate in the future. The hosting company first presented their PSS business approach, followed by the researchers’ presentation of the PSS area of research. In the latter part of the workshop, the topics of value, price and price setting and how to achieve customer value by retrieving correct customer needs were discussed among the participants, the outcome of which is described below. For this discussion, all participants had prepared presentations of how their companies were tackling these issues. By doing so, the participants learned how certain issues can be solved in a good manner. As a result of these discussions, it was found that the customer maturity level of having products brought through PSS fluctuated greatly. As an example, the United States was put forward as a country that was far ahead in leasing and financial services. Another important thing that is needed for a successful PSS is to convince the customer that the costs for spare parts is not as high as the costs for stops in production, and/or that the production will not have a high productivity. Other questions that were raised during the discussion were how to understand the customers’ unconscious needs. An approach that was suggested was to better understand the customers’ customer; this could mean, for example, that the customer perspective would be broader including several new stakeholders. Too few working with service development The companies in the network all have relatively large product development organizations. A common situation for them was that service development department was very small and newly established; this meant that it lacked resources. At one of the companies, the relationship between the ratios of staff working with product development compared to the staff working with service development was 100:3. The companies also put forward

the need for product development and service development to have a close relationship and interaction. There was also an agreement that when performing service development, one has to have a much deeper understanding of the customer - and sometimes the customers’ customer - in order to offer the right services. One of the companies had developed a service fulfillment process showing how the company undergoes the process from how to select and approach customers to how to implement the services and evaluate the performance and costs results and identify improvement areas or possibilities for new types of services. Price setting strategies One major challenge for the companies is how to price offerings. A common problem was to decide at what level to bundle offerings. This can be made at a high level, meaning that customers pay a monthly fee, all-included, or by letting the customer choose from different offering modules. Among the companies participating in the learning network, pricing differed from having a totally open pricing scheme of different modules in an offer to having a monthly fee covering all aspects where the modules’ separate costs are closed. When differentiating the costs, one company used value-based price setting. The company arranged workshops in which they let the customers rank the value of different modules. This has guided the price setting, letting the most valuable parts and modules have a higher price, since the customers value that more and are willing to pay more. This is, however, delicate since the trust of the customers always has to be in focus, and by value-based price setting the trust of the customers needs to be considered. When having value-based costing in maintenance, one aspect to consider is that the cost for standard spare parts is known, and a comparison with other companies is easy to make. Another company, on the other hand, was very keen on hiding the separate costs for the customer in order to be able to bundle offerings and charge a monthly fee. In this case, the price setting was cost-based. The problem then was to actually have control over, and accurately estimate the costs for different services such as maintenance. One of the companies did not really sell services, although they did offer and perform a range of services that increased their customers’ efficiency. Rather, they merely used the services as a way for selling more products, and by providing services they could steer the customers into choosing the products with high margins. Having a higher price on the products is also then one way of getting paid for performing the services. The company is in the process of bundling service offerings for the upper market segment, where customers value services and are prepared to pay extra for them. The plan is to shift the focus from the costs for the products to the total costs. A software is currently being developed to conduct these calculations. Changing the mindset of customers A great challenge for the companies was to change the mindset of the customers. For one company, the challenge was to make the customer understand that the major costs for the customers does not lie in the costs for maintenance and spare parts, but rather in the costs of having to stop the production or that the product has a lower productivity. Then, the customer gets more into thinking about the value of the services or products used, rather than the costs for the physical product or spare parts. It was clear that it was a major difficulty to start offering services that used to be bundled into the product price.

301

Changing the mindset of sales force and retailers Another challenge found for the future was dealing with the promotion of business ideas to local sale companies in different markets, i.e. how should the internal marketing be conducted successfully? Some of the companies mentioned that they had difficulties convincing their own sales force and the retailers to really accept and embrace the services concepts. The sales people are used to selling products, and this requires a new way for them to approach the customers. 5.3 Second network workshop The second workshop was held at the 14th of October 2008 there were 9 company participants and 5 university researchers. In comparison to the first workshop a mix of new and old companies and researchers were participating. Also, some companies that were invited to neither the kick-off nor the first workshop wanted to join this workshop. However, since their relations to some of the already participating companies were delicate (customers and/or competitors) these company could not join at this stage. At this workshop the theme was how to organize sell and deliver product/service systems. Preceding this workshop the company participants had prepared short presentations of how they are dealing with these issues at their company. During the presentations discussions were held among the company members with some influence from the general knowledge of the participating researchers. In short the discussions included: Developing new aftermarket services Selling products as PSS included potential to sell more aftermarket services. For example, one company has started to sell service kits and service knowledge in connection to their normal product sales. The company expects good revenue on these kinds of services which would also help them increase the knowledge of how their physical products are used by their customers. Organization of developing product/service systems One of the participating companies admits having the service part of the PSS coming some time later in the development process. However, the service issues have lately been given a higher priority when developing PSS. Another company reveals not having any relation between the service developers and the product developers. In this case only few are working with service development. However, the company has initiated more focus on the service parts of their PSS. Traditionally technical problems have been solved by technical solutions and not by service solutions. Currently there is a need for methods/tools to collect experiences from the market. Some of the company participants describe that the organization of service development is not as organized and structured as in the product development. This is an issue that needs attention. Selling product/service systems A company that have had much focus on selling products have now started to see the service parts of the PSS as an competitive edge that gives their customers an added value. This is something their competitors cannot yet provide. Also, the PSS have shown to have a better profit margin than the traditional product sales. In this area it was also mentioned that it is important for the sales people to have skills in the knowledge of selling PSS to their customer management themselves or to have someone with them at the point of sales who could

302

speak this language. Otherwise, the chances of a PSS deal would decrease much. Currently many service parts a given away freely as good will. However, these will in the future profitable and come with a price tag. On the other hand one of the participant states: “it is dangerous to charge for something that previously have been given away for free”. On this point the network participants do not agree in full. Furthermore, the second network workshop included some feedback and information from the researchers about on-going research in the area which included a study at the participating companies. These kinds of research involvements during the workshops are not put in focus but add to the knowledge sharing among the company participants and the researchers. 5.4 Future network workshops From the start-up meeting it was decided to have the workshop series running with workshops twice per year. This was determined to keep the network alive, but without having too many meetings per year. The next and third workshop will be held the 25th of March 2009. 6 DISCUSSION The discussion below is based on accomplished workshops in the large company network, and in combination with the authors’ previous research. 6.1 Different purposes for using learning networks Learning networks can be used for different purposes. As described in this paper, one type of learning network was part of a research methodology and used for SMEs with limited knowledge and experience in integrated product and service offerings [20]. In the SME-case, the purpose was to build understanding of different business models and to support the companies in developing their business. In that case, each meeting had a pre-defined theme decided by the researchers, and each network meeting was also a workshop meeting where the participants actually worked on a specific task each time that had specific outcomes. The outcomes were then used as input for the next workshop. The researchers’ roles were very driven, guiding the companies through a development process and a methodology with a specific number of pre-defined meetings. The large companies had a greater pre-understanding of the business logic, and the purpose of the meetings was to continuously improve the companies’ business. The themes were determined based on the company needs (with strong influence by the companies), and the researchers’ role was to act as moderators and participate in discussions, and to provide input and new reflections to the companies. 6.2 Choosing participating companies for continuous improvements networks When choosing participating companies, it is of course firstly important that they do not have a competitive relationship; if they do, then neither trust nor sharing of knowledge can be achieved. However, what is not directly obvious is that the companies working with PSS, even though they have different business, have several common challenges related to PSS. It has been noted that the companies have different types of PSS offerings, and they expressed that in order to get the most use from the network, they would like to be divided into groups with companies with similar situations. This made the large company participants eager to invite other companies in similar situations, something important in order to fertilize their

own need to discuss their challenges and obstacles within PSS. Examples of companies that they wanted to invite were those that had a strong focus on after-sales-related services such as companies with capital-intensive products with long lifetimes. Another category was companies that had more knowledge-intensive services such as advice and supported customers with training and business advice, or companies that used services for selling products. Yet another category of companies were those in a crisis situation, which were forced into changing their business logic for serving the market. The companies have different challenges that they meet. As mentioned in section 5.3 there could be problems of who to include in this kind of learning network since their relation could be of delicate character. These issues must be considered and if the amount of participating companies are to be broader, covering customers and competitors for some companies the openness and topics needs to be changed to fit the new circumstances. 6.3 The PSS challenges for large companies As indicated by the questions that the companies have risen (see section 5), there is definitely the need for more research within this area. The issues are recognizable and vary from organizational questions to product and service development, customer relationship, how to create and get paid for value creation and how to set up the product portfolio. There is clearly room for various kinds of research areas and expertise. Research needs to be both focused but also interdisciplinary in order to understand the total company picture. The following is a list of PSS-related challenges found through this research: How to market PSS? One big challenge found at the large companies was to market their PSS in a good manner, both internally in the company group but also to their customers. The PSS business logic is quite different from the product sellingbased business logic they and their customers are used to. Customers have problems validating PSS offerings, e.g. since many PSS offerings also cover the use phase and the costs that are generated there. Customers are quite often not used to calculating life cycle costs; by tradition, they are more focused on cost prices. These challenges have also been uncovered through company interviews by other researchers e.g. [25]. How to organize the PSS development? This have been described by the participating network companies as an difficult task since it is separated by tradition and the company structures for developing products and services are much different. It is a challenge for the PSS developers to achieve an efficient integrated development of the products and services that are to be included in their offerings. The development teams needs to consist of people of multidisciplinary functions to a larger extent than in e.g. traditional product and/or service development. Also, the life-cycle perspective give opportunities that the PSS providing companies should take care of in a good manner e.g. adapt their product/service design for this purpose. Examples of product design adaptations of products are illustrated by Sundin et al. in [26] One characteristics peculiar to PSS development may be uncertainty of PSS [25]. Some companies recognize the importance of addressing uncertainty during the whole process of PSS development. Another dimension of divisions should be considered in addition to be able to

obtain more optimal way of PSS development, especially in the case of large firms. How should the price be set? A second big challenge found was how the price of the PSS should be set in a good manner. Some companies needed to get paid for services that previously were given away for free. The price needs to balance with how much the customer value the PSS create at the customer. These issues have also been researched previously by e.g. Rosvall & Rosvall [27]. How can new technology be used? Another challenge found was how new technology can support the development and use of PSS. This could include both development support and monitoring and other life-cycle support. A part of these kinds of support systems have been studied by other researchers, e.g. [26]. This is an area that needs more attention, since few systems are in place (see e.g. [21, 26, 28-30]). How to benefit from environmental potentials? The authors’ previous research have identified and quantified how PSS can decrease negative environmental impact. One aspect (not mentioned by the companies) that has to meet is how to get the companies to be interested in the environmental potential of PSS and include those aspects to fully reach the environmental potential. PSS offering companies’ and their customers seems not to be that interested in PSS’s environmental potential. The challenge is how to improve customers’ interest in order to decrease the environmental impact. Several questions are easily stated: Is there a need for policymakers to step up and take an even closer look at services? Can we as researchers promote environmental issues in more business model approaches? This is an issue that needs to be addressed further. 7 CONCLUSIONS AND FUTURE RESEARCH Industrial Product/Service Systems is for many companies a new business approach. The large company network studied for this research faces many challenges in order to achieve a good PSS business. Much of these challenges are related to changing different peoples’ mindset whether it is within the company and/or with external companies and customers. Also, good solutions to these challenges need to be facilitated. Having a learning network approach of dealing with these challenges has so far been perceived as a good manner of tackling the questions raised within the PSS providing companies. The companies have learned from each other and as well as retrieved knowledge from research from the researchers. The researchers on the other hand have received a better understanding of the companies’ faced reality. This would work as a good base for further research that is needed in the PSS industry. 8 ACKNOWLEDGMENTS The authors wish to thank the people from the companies and the researchers participating during the learning network workshop series. Also, the authors express their gratitude to Dr. Margareta Groth at the Swedish Governmental Agency for Innovation Systems (VINNOVA) for moderating the network start-up meeting and for partly financing this research.

303

9 REFERENCES 1. Oliva, R. and R. Kallenberg, Managing the transition from products to services. International Journal of Service Industry Management, 2003. Vol. 14(No. 2): p. 160-172. 2. Sakao, T., et al., How Are Product-Service Combined Offers Provided in Germany and Italy? – Analysis with Company Sizes and Countries. Journal of Systems Science and Systems Engineering, 2008. 17(3): p. 367-381. 3. Ölundh, G., Environmental and Developmental Perspectives of Functional Sales, in Division of Integrated Product Development, Department of Machine Design. 2003, Royal Institute of Technology: Stockholm, Sweden. p. 108. 4. Sundin, E., M. Larsson, and A. Nielsen, Design for Functional Sales - A Case Study of Forklift Trucks at BT Industries, in Proceedings of EcoDesign 05. 2005: Tokyo. 5. Söderström, J., Från produkt till tjänst - Utveckling av affärs- och miljöstrategier i produktorienterade företag, in Handelshögskolan. 2004, Stockholm University: Stockholm. 6. Brännström, O. and B.-O. Elfström. Integrated Product & Service Offerings - Their Rationale and Creation. in International Product Development Conference. 2002. Sophia-Antipolis (near Nice), France: European institute for Advanced Studies in Management. 7. Shimomura, Y., et al. A Proposal for Service Modelling. in EcoDesign 2003. 2003. Tokyo, Japan. 8. Sakao, T., et al., Modeling Design Objects in CAD system for Service/Product Engineering. ComputerAided Design, Accepted to appear. 9. Lindahl, M. and G. Ölundh. The Meaning of Functional Sales. in Life Cycle Engineering: Challenges and Opportunities: 8th International Seminar on Life Cycle Engineering. 2001. Varna, Bulgaria: CIRP. 10. Alonso-Rasgado, T., G. Thompson, and B.-O. Elfström, The design of functional (total care) products. Journal of Engineering Design, 2004. 15(6): p. 515-540. 11. Alonso-Rasgado, T. and G. Thompson, A rapid design process for Total Care Product creation. Journal of Engineering Design, 2006. Vol. 17(No. 6): p. 509-531. 12. Tukker, A. and U. Tischner, New Business for Old Europe. 2006, Sheffield: Greenleaf Publishing. 13. Davies, A., Moving base into high-value integrated solutions: a value stream approach, . Industrial and Corporate Change, 2004. 13(5): p. 727-756. 14. Sakao, T. and Y. Shimomura, Service Engineering: a novel engineering discipline for producers to increase value combing service and product. Journal of Cleaner Production, 2007(No. 15): p. 590-604. 15. Meier, H. and O. Völker. Industrial Product-ServiceSystems - Typology of Service Supply Chain for IPS2 Providing. in Manufacturing Systems and Technologies for the New Frontier - Proceedings for The 41st CIRP Conference on Manufacturing Systems. 2008. Tokyo: Springer. 16. Kvale, S., The qualitative research interview - a phenomenological and a hermeneutical mode of understanding. Journal of Phenomenological Psychology, 1983. 14: p. 171-196.

304

17. Mont, O.K., Clarifying the concept of product-service system. Journal of Cleaner Production, 2002. 10(3): p. 237-245. 18. Tukker, A. and U. Tischner, Product-services as a research field: past, present and future. Reflections from a decade of research. Journal of Cleaner Production, 2006. 14: p. 1552-1556. 19. Lindahl, M., T. Sakao, and A. Öhrwall Rönnbäck, Business Implications of Integrated Product and Service Offerings, in CIRP IPS² Conference. 2009: Cranfield, UK. 20. Lindahl, M., et al. Learning networks: a method for Integrated Product and Service Engineering – experience from the IPSE project. in 41st CIRP Conference on Manufacturing Systems – Manufacturing Systems and Technologies for the New Frontier. 2008. The University of Tokyo, Tokyo, Japan. 21. Sundin, E., et al., Remanufacturing of Products used in Product Service System Offerings, in 41st CIRP Conference on Manufacturing Systems – Manufacturing Systems and Technologies for the New Frontier. 2008: The University of Tokyo, Tokyo, Japan. 22. Ölundh, G. and S. Ritzen, Funktionsförsäljning och produkters miljöaspekter – en studie i tre svenska tillverkningsföretag. 2002, Naturvårdsverket, ISBN 91-620-5234-9: Stockholm. p. 40. 23. Ritzén, S., et al., Learning to change – a network approach to engineering management development, in Proceedings of the International Conference on Engineering Design, ICED 05. 2005: Melbourne, Australia. 24. Kolb, D.A., Experimental Learning – Experience as the Source of learning and Development. 1984, Englewood Cliffs, NJ, USA.: Prentice Hall PTR. 25. Sakao, T. and E. Sundin, Analysis of Integrated Product and Service Offerings from current perspectives of providers and customers, in CIRP IPS2 Conference 2009. 2009: Cranfield. 26. Sundin, E., M. Lindahl, and W. Ijomah, Product Design for Product Service Systems - design experiences from three Manufacturing companies from Sweden. Journal of Engineering Design, To appear. 27. Rosvall, L. and K. Rosvall, Prissättning efter kundvärde - Ett paradigmskifte i modern affärsutveckling (in Swedish). 2001, Lund: Industrilitteratur. 28. Meier, H., Lifecycle-based Service Design for innovative business models. Annals of the CIRP, 2004. 53(1): p. 393-396. 29. Sundin, E., O. Tang, and E. Mårtén. The Swedish Remanufacturing Industry - An Overview of Present Status and Future Potential. in Proceedings of CIRP Life Cycle Engineering Seminar -12th edition - LCE05. 2005. Grenoble, France: Laboratoire 3S. 30. Aurich, C., C. Fuchs, and F. DeVries, An Approach to Life Cycle Oriented Technical Service Design. Annals of the CIRP, 2004. 53: p. 151-154.

Dynamic IPS²-Networks and -Operations Based on Software Agents H. Meier1, E. Uhlmann², C.M. Krug1, O. Völker1, C. Geisert², C. Stelzer² Ruhr-University Bochum, Chair of Production Systems, Bochum, Germany ² Technische Universität Berlin, Institute of Machine Tools and Factory Management, Berlin, Germany [email protected]; [email protected] 1

Abstract This article describes how an IPS² network should be build up by considering the dynamic behavior of the IPS² along its life-cycle. How the network partner could participate and how they allocate their capacities will be also discussed as questions regarding the commissioning by considering the business model. The realization of this concept is based on a multi-agent system. Therefore all service delivery involved IPS² objects like product, network partner and service technicians are represented by software agents. All this will be pointed out by an availability oriented maintenance scenario in the field of micro production. Keywords: Industrial Product-Service Systems; IPS²; Organization; Multi-Agent System; IPS² Network

1 MOTIVATION Industrial Product-Service Systems (IPS²) can be characterized by the fact that they consist of a combination of tangible product and intangible service shares, whose service shares provide a value to the customer via the complete life-cycle [1] [2]. In the different function-, availability- or result-oriented use models [3] the order for the service performance is either given by the customer (function-oriented) or by the provider (availability-/result-oriented). Overall Industrial ProductService Systems deliver value in industrial application for the customer [4]. But this objective is just attractive for the IPS² provider if he gains an acceptable profit by offering this solution. To reach this win-win situation the operation of the IPS² respectively the service delivery has to be organized efficiently. This includes the internal structure of the IPS² provider as well as the buildup of the IPS² network and its control. The cooperation of companies is one of the main aspects of organizations in the service sector. The reasons are high efficiency of value creation and flexible combination of complementary resources by keeping economical independency [5]. The approach to be shown will point out the typology of the service supply chain under the context of the different targets of delivery flows. Due to that each IPS² has got its own network as well as the dynamic of the IPS² and the connected dynamic of each IPS² network, the establishing and control of the IPS² delivery becomes very complex for the IPS² provider. This challenge can be managed just by an information and communication technology (ICT) support. This system has to merge automatically the IPS² network depending on the characteristics of the IPS², its tasks and on the prospective network partner. Further on it has to control the planned delivery processes. Therefore the basis of such a system has to be the automated communication within the network as soon as a concept to handle this in an efficient way by autonomous decisions of system parts considering the current circumstance of each decision. A multi-agent software system will provide the needed advantages. After the software agents are set they will bargain autonomous about the different objectives in the

CIRP IPS2 Conference 2009

305

determined borders. The coordination of the distributed software agents has to be managed by a central agent located at the IPS² provider to reach efficient collaboration within the whole network and to assume the responsibility for the IPS² to the customer. Therefore potentials but also problem fields within an automated customized IPS² delivery will be characterized. 2

BASICS OF THE IPS² NETWORK

2.1 Network Partner Roles The delivery of IPS² needs a new understanding of partnership roles and accountability for actions during the IPS² life-cycle [6]. The roles of the IPS² network partners can be characterized by different delivery flows, communication partners and communication directions and the relationship to the IPS² or to the customer. IPS² module

customer service IPS² IPS² module product/service

product

IPS² provider product

IPS² IPS² IPS² module module module supplier supplier supplier

component component component supplier supplier supplier

Legend: delivery flow network boundary

service service service supplier supplier supplier

communication

Figure 1: Typology IPS² Network Organization Figure 1 shows the relation and the main elements of the IPS² supply chain. It consists of the customer of the IPS², the IPS² provider, the IPS² module suppliers, the

component suppliers and the service suppliers. In the following the different elements will be detailed. Customer: The customer receives the IPS² from the IPS² provider. Depending on the customer needs the IPS² provider has to keep the operability during the use phase. Dependent on the optimal solution for the IPS² the customer pays for the delivered function, the availability or the result. The resources of the customer can be embedded in the delivery processes. The responsibility of the customer for processes of the IPS² changes between the sold business models. IPS² Provider: The IPS² provider has the business relationship with the customer. He represents the one face to the customer and manages all communication in the network. The IPS² provider takes risks depending on the chosen business model. Therefore he coordinates the delivery flow for the IPS² network. Parts of the IPS² can be subcontracted to IPS² module, component or service suppliers by keeping the leadership and responsibility. IPS² Module Supplier: An IPS² module is a product-service subsystem of the IPS². It consists of product and service shares with an integrated development. He delivers the IPS² module by himself or by offering parts of it to sub-suppliers over the life-cycle. The IPS² module supplier can change all necessary processes by guaranteeing the same result to the IPS² provider. Component Supplier: A component is a tangible part of the IPS² without integrated service. It is storable and can be delivered either to the IPS² provider or the customer. Service Supplier: The service is the intangible part of the IPS². It is not storable and therefore it has to be delivered directly to the customer and will become manifest in the IPS². All needed input to carry out a delivery process like material, energy, information [7] and, additional in service delivery, human resources will be described as resources. The needed resources, including those to manage the byproducts [8] have to be planned in advance and have to be allocated in the right time, place and quality. In service delivery this parameters change between each IPS² and even between the delivery processes. While place and time obviously change in connection to the customer the quality, e.g. of a spare part, may vary depending on the chosen business model. Further on, the resource planning and scheduling has to be able to react on the dynamics of the IPS² and of course on breakdowns. Therefore the IPS² provides a set of optimization parameters like the time variance, the process variance or the partial substitution of product and service shares [9]. This optimization has to fulfill the demands of the customer by using all resources over the described IPS² network - including the customer resources. This is just possible with a new operational and organizational structure and the described automated communication within the network. 2.2 Operational and Organizational Structure Processes or business processes are very common objects to illustrate the operational structure and process structuring can be found in a growing number of publications. But an exact definition especially for the use of processes in businesses is not given so far [10]. Most of the publications do not divide between process and

business process and use them synonymously. For operational businesses it is defined that the class of business processes is a subclass of processes. ZELLNER tries to analyze the complete variant of different definitions to extract the commonness for a better process understanding [11]. Every process has an input and an output. The process between the input and output is a combination of activities or tasks that are structured by time or event [11]. The process organization of businesses and also of cooperation and of customer communication is getting more important. Therefore a transformation from a functional business organization to a process business organization can be recognized. The functional business organization has been led to a business unit thinking and thus to coordination problems between different units [12]. Where the process business organization tries to concentrate on the value stream and therefore support the concentration on the customer. Another benefit of process business organizations are transparent processes for the production or service delivery. Process structures describe the separate process steps und the connection between them. The measurement of different activities in cost, time and quality is subject of the service delivery processes. The operation of Industrial Product-Service Systems is to manage the delivery of tangible and intangible parts that are designed integrated as product modules, service modules or product-service modules. Due to the new paradigm shift of an integrated and dynamic interdependent development of products and services, an IPS² process and object description is generated with solutions and solution clusters. The organizational tasks of the IPS² provider are the production planning, the operational delivery planning and to assume the responsibility of the IPS². Thus an operational controlling of the tasks and generated results is required. To work efficiently and to be competitive all work that is not core competence of the own enterprise will attempt to be assigned to other companies or be bought from the market. To fulfill all qualitative and quantitative customer requirements the IPS² provider has to find a reasonable economic size for subdividing the IPS² jobs between possible network partners. Therefore the resource allocation is being done in two different phases during the IPS² operation. Table 1 show the tasks of the resource allocation for the IPS² provider and the consequences for the IPS² network during the initial phase by getting the IPS² product model and the life-cycle phase by getting knowledge about all aspects around and with the IPS². The starting point and hence the initial factor of the job assignment to generate the organizational structure is the developed IPS². At this stage of the IPS² development process the IPS² product model contains detailed descriptions of every part and process with qualitative and quantitative requirements. These requirements are used to build a target state for the organizational IPS² structure. Beginning with the initial phase different possibilities to distribute the responsibility for the necessary processes exist. First the own competencies of the IPS² product model are identified and transferred to the resource planning by the IPS² provider. Beside the identification by competencies also strategic targets have to be considered to allocate the delivery responsibility of IPS² parts. The resource planning depends on the time to the date of delivery and its duration. Possible conflicts with other jobs can be solved by strategic resource upgrading. The IPS² parts that are delivered by external partners are leading to new organizational requirements. The willingness from the IPS² network partners to externalize their IPS² delivery relevant data to the IPS²

306

provider has to be one more requirement. Also the customer has to link his processes with the IPS² to allow the offering of availability-oriented and result-oriented IPS². If the IPS² provider wants to react in a desired time it is necessary to know the production or manufacturing plans or downtimes from the customer. To share data in the IPS² network IPS² execution system interfaces have to be established by the network partners. Resource planning

Network creation:



Strategic capacity planning



Resource attributes changing

Clear responsibilities in IPS² product model



Interface coordination



Commitment of goals and aims

IPS² concept model

Network Optimization: • Substitution of suppliers • Adaptation of IPS² network partner pool

IPS² product model supplier market analysis

Table 1: Tasks of IPS² Provider during IPS² life-cycle The second phase, the actual use of the IPS², describes the use of knowledge out of the life-cycle phase to optimize the job and resource planning [13]. Options to change the job and resource planning are to substitute product and/or service shares, time variances and the upgrading of resource attributes from the IPS² provider or the IPS² network. Changes of the IPS² job demands will affect the partner selection and the job assignment dynamically in existing IPS² networks. 2.3 Job Association The definition of IPS² as the delivery of value in use for the customer in industrial application implies a wide variety of different functions in the IPS² product model. This variety of different functions can only be handled by a cooperation of companies to reach the optimal solution [14]. In this chapter the association of external partners by the IPS² provider will be described. First additional requirements for the IPS² job assignment are emerging from the concept “one face to the customer” and importance of IPS² parts. The “one face to the customer” concept implies the central communication between the customer and the IPS² provider. Even questions from the customer to a part of the IPS² that is being handled by a network partner have to be communicated over the IPS² provider. This does not necessarily contain an explicit action of the IPS² provider, e.g. to send the customer question forward to the network partner. But the IPS² provider should be able to build a history of IPS² delivery information. The job assignment will be done by the IPS² provider and the afterward job control will be done by the individual selected partner. The job association depends on the art of the process and whether the process can be planned easily. For example a critical IPS² process that could lead to a malfunction of the IPS² will be assigned over a longer period of time than non-critical processes. Contracts have to guarantee the delivery even in the case of unpredictable initiation of critical processes. Better planable processes that are not critical for the IPS² operation are assigned for shorter time periods. This would lead to more competition between possible network partners and to an advance of qualitative job results.

307

potential for automation of the processes

Time variance



if partner is not able to deliver

Substitution



Supplier job assignment

subcontracting by using the supplier pool





suitable suppliers selection job advertisement

offer comparison supplier negotiation contracting

job ability indication delivery confirmation / acceptance

contract controlling

warranty

new job advertisements for planable and non-critical jobs

Result for IPS² network



continuous job goal deviation leads to new advertisement

IPS² provider

control loops

Life-cycle phase: generated knowledge from delivery and use of IPS²

Initial Phase: IPS² product model

Phase

The purchase of the IPS² parts from external partners is mostly influenced by service characteristics. The manufacturing and use will be at the same time for the “product” (uno-actu-principle). Thus the three dimensions of a service – potential dimension, process dimension, result dimension – have to be exactly defined for every job description [15]. The tangible parts of the IPS² are described by object lists, technical and production drawings and can be produced definite. A controlling of the process parameters can be done with the manufactured part instead of a service that is more difficult to control before the delivery. Therefore the IPS² parts need standardized description to enable a successful purchase from suppliers. It is helpful for a better communication and a common understanding of the goals and restrictions of the jobs. Next step could be the trading of intangible parts.

contract fullfillment

Legend:

process step

IPS² concretion level

Figure 2: IPS² Job Association Figure 2 shows the business process for the initial job association of IPS² product model parts. The process steps are based on the DIN PAS 1018 model for “the essential structure for the description of services in the procurement stage” [15]. Current service procurements

are starting with the needs recognition and the needs description. The described services will then be calculated to determine costs and prizes. After that step possible suppliers will be detected, then the job will be offered and last the best fitting supplier will be selected. Next the model describes the delivery of services and parallel controlling. The business process ends with the contract fulfillment. For IPS² job procurement the DIN business process model has to be adapted. The processes of needs recognition, needs description and cost calculation are already described in the IPS² concept model and the IPS² product model. Job procurement in the context of IPS² that is starting with the delivery and use phase will begin with the market analysis for suppliers. This process is determined of very heterogenic data and the information collection has to be done by the IPS² provider. The data will be collected in a list or matrix to compare the possible suppliers. After that the processes “suitable suppliers selection”, “job advertisement”, “offer comparison”, “supplier negotiation” and “contracting” are following. These processes have a highly potential for automation. Standardized communication and descriptions are needed and can be used to get an automated information flow between the corresponding partners. The overall time for job assignment is the target and an advantage of the process automation. Another restriction for job procurement is the modularization of the IPS² product model. The different modules have to have a certain kind of independency to minor the coordination efforts to avoid taking more effort in coordination than in the actual job delivery. After the contracting process a monitoring of the delivery ability of the contracted partner is necessary. Until the beginning of the delivery processes the ability will be checked and if other jobs or sudden accidents lead to problems in delivery, a new advertisement of the job will be done. The IPS² partners in the network have to hold all necessary data up to date. A big problem would be the case if no partner could be found to run the advertised job. Solution could be a specialized team of experts out of every network partner that could be entered to help to deliver. But the consequences and the requirements for the IPS² provider, the supplying companies, the network and the customer have to be analyzed in the future.

market [19]. Compared to the IPS² approach available systems are sold as additional physical products that have to be integrated into the spindle after the spindle was designed. To ensure an efficient use of the IPS² a high degree of automation is needed. For an IPS² provider it may be useful to distribute different tasks within the IPS² to several network partners with regard to their core competences. One of the most important duties in networking is the organization and control of service activities. The use of software agents is the focused solution to manage these duties by automated communication and interaction between IPS² objects. 3.1 Software Agents in IPS² There is still no uniform definition of an intelligent software agent. In this research project the characterization by Jennings is preferred [20]: “An agent is an encapsulated computer system that is situated in some environment, and that is capable of flexible, autonomous action in that environment in order to meet its design objectives." According to this definition a software agent can be described in an easy way by its ability to observe its environment, to interact with its environment to reach a defined goal, and in the case of a multi-agent system (MAS) to communicate with other agents. A software agent acts as a virtual representative of a physical object. It is a piece of software that is provided with the profile of the objects needs and is able to act autonomously. Typical application areas for software agents can be found in personal information management [21], the support of business processes [22] and of electronic commerce [23]. An example for an industrial application of software agents can be found in [24]. There, the concept of an agent based production monitoring system in the field of automotive industry is proposed. In the IPS² context software agents can be used to control the IPS² operation during its life-cycle. Every object that is involved in IPS² operation has defined tasks. A MAS is suitable to support and control these tasks by its high degree of automated communication and autonomous action. 4

3 TECHNICAL ENABLER The following chapter shows a scenario for the initialization of an IPS² network and the dynamic adaptation with respect to dynamic changes of the IPS² demands. Before the IPS² network can be initialized, a pool of potential partners is needed. An efficient way to interact with these is the use of software agents. Different types of software agents can be used in the different phases of the IPS². In the phase of initial network building the IPS² provider uses a software agent platform to announce his need for network partners. Software agents of potential partners analyze the task description and in case of capability submit an offer. After building the initial network another software agent platform is used for communication between the network partners. On the hardware side is a need of specialized technical support. The investigation of wear mechanisms and the development of condition monitoring algorithms for the detection of wear and tear of mechanical machine components is part of current research projects [16] [17] [18]. First industrial applications are available on the

DEMONSTRATOR SCENARIO

4.1 Initial Conditions The following scenario describes the software agents’ interaction in the event of a detected imminent breakdown due to wear and tear. In this illustrative example an industrial customer wants to act as supplier for a manufacturer of a product that includes micro components. It is agreed by contract that the customer has to deliver a defined number of parts per time interval. The contract also includes a warranty for the quality of the fabricated parts. Therefore, the analysis of the customer requirements in the early phase of IPS² design lead to an availability oriented use model. In this scenario the resulting IPS² represents a three axes CNC micro milling machine for the manufacturing of micro parts, including all services that are needed to ensure the availability of function relevant components. The further description of the scenario focuses on the milling spindle as the most stressed core component of the machine tool and how to warrant its availability. Due to the analyzed customer requirements the resulting product model of the IPS² has following attributes:

308



Milling spindle with integrated condition monitoring system (CMS) and communication interface as enabler for condition-based maintenance.



Condition-based maintenance of the spindle and periodical maintenance of non-critical machine components. This includes inspection, replacement of wear parts, and repair.



Spare part service.

the IPS² provider has to expand the given IPS² network temporarily. For this reason he may use again the software agent platform for IPS² job association (cf. Figure 2). Depending on whether he finds an alternative service supplier that fulfills all required specifications regarding the three economical aspects of time, cost and quality the scenario differs: Environment Perception Spindle / HMI

4.2 Initial IPS² Network An optimized operation of this IPS² can only be reached, if the IPS² provider takes the responsibility for all needed IPS² processes. Nevertheless, different IPS² processes have to be distributed to different stakeholders within an IPS² network, due to restricted competences and manpower of the IPS² provider. The resulting network consists of partners with special core competences to fulfill the attributes of the given product model. Let us suppose that the following network partners are chosen: •



IPS² provider: He is the only contract and contact partner of the customer. He is responsible for all activities, needed to use the IPS² as agreed in the contract. He is the leader of the consortium that builds the IPS² network. IPS² module supplier: He is the manufacturer of the spindle with an integrated condition monitoring system. The monitoring of wear parts’ condition (data acquisition, signal analysis, data storage, and notification) is the integral service part of the module.



IPS² service supplier: This network partner offers the periodic maintenance service for the milling machine and the condition based for the spindle. This includes the organization of spare parts, which he may obtain from a third party as second tier supplier. The contracting of second tier suppliers lies in the sole responsibility of each network partner. 4.3 Software Agents in the IPS² Network A condition monitoring software agent (CMSA) collects and analyzes condition relevant data during the spindle’s life-cycle. Based on an implemented set of inference rules the agent detects critical trends of the condition of function relevant components, e.g. bearings or parts of the gripping mechanism. When a critical trend is detected, the CMSA calculates the remaining time until an alarm limit will be exceeded. This information will be send by the agent to the IPS² network management software agent (NMSA) which is located on a server at the site of the IPS² provider. If available, more detailed information can be attached to the message, e. g. kind of expected failure. Figure 3 shows the principal structure of interaction for the CMSA. The NMSA logs the information and forwards the message to the IPS² module supplier respectively to his software agent. Based on the received information the IPS² module supplier allocates a plan of service procedures needed to restore the spindle’s origin state. This includes a list of required spare parts, tools and a detailed process description. The maintenance planning software agent (MPSA) sends this service procedures to the NMSA. This information is used by the IPS² provider to organize the logistics for the maintenance activities. In a first step the NMSA sends a request to the maintenance service software agent (MSSA) of the IPS² service supplier. This message contains the detailed job description and the schedule. After checking the internal enterprise resource planning system (ERP) the MSSA declines the job offer, due to a capacity constraint on the part of the IPS² service supplier. To avoid a contract penalty in case of the imminent breakdown of the spindle 309

• Sensors • Counters • PLC messages • Alarms

CMSA Behavior Environmental model

NMSA Environmental model

Behavior

Aims Aims Aims Aims Aims Aims

Action

Figure 3: Principal Structure of CMSA Interaction Case 1 (alternative IPS² service supplier): After contracting an alternative service provider a service technician software agent (STSA) is initialized to control all service related activities from work preparation (e.g. organization of spare parts and tools), journey up to maintenance. All these processes are defined within the IPS² product model and have characteristics to enable monitoring and control. The task of the STSA is to monitor the activities and to adapt correcting-variables of the processes if the actual process workflow differs from the planned one. To ensure process reliability the STSA has to perceive data from the environment, e.g. current position from the global positioning system (GPS), duration of a maintenance process step. The software agent’s behavior can effect two different actions: •

Adaptation of a correcting element in terms of feedback control engineering or



sending a request to NMSA. :CMSA

:NMSA

:MPSA

:MSSA1

:MSSA2

:STSA

notify forward and request response request decline request accept order initialize notify Legend: • :CMSA • :NMSA • :MPSA • :MSSA • :STSA

– Condition Monitoring Software Agent – Network Managing Software Agent – Maintenance Planning Software Agent – Maintenance Service Software Agent – Service Technician Software Agent

Figure 4: Principal Sequence Diagram of the Scenario This control is active until the service process is completed. Additional to the control functionality all process steps are documented by logging their characteristic variables. The principal interaction process of the above described scenario is shown in Figure 4. Case 2 (no IPS² service supplier available): If the IPS² provider is not able to find an alternative service supplier for this job within the remaining life-time of the spindle, the NMSA can upgrade the CMSA by

sending an expanded behavior. In this case, the new behavior is the command to go into a fail-safe mode to avoid consequential damage of other machine components. This results in a breach of the agreed availability warranty and identifies the limits of automation using a MAS. 5 CONCLUSION The paper shows the need of a new network organization that supports the operation and delivery of an IPS² with its special characteristics, e.g. the high dynamic of changes of the IPS² during the life-cycle. To manage this challenge the illustrated network is characterized by its ability to change flexible the network partners. Therefore an automated communication between the potential network partners is essential to fulfill the real-time demands. A multi agent system has the ability to cope with the network demands. Thereby one multi-agent system is needed for initial network configuration and another one for the control of the delivery processes during the IPS² life-cycle. To demonstrate this approach an example scenario for the agent communication concept was described. It shows the different communication levels in case of an imminent machine breakdown. 6 ACKNOWLEDGMENTS We express our sincere thanks to the Deutsche Forschungsgemeinschaft (DFG) for financing this research within the Collaborative Research Project SFB/TR29 on Industrial Product-Service Systems – dynamic interdependency of products and services in the production area. 7 REFERENCES [1] Tan. A. R., McAloone, T. C., 2006, Characteristics of strategies in Product/Service-System Development, International Design Conference – Design 2006, Dubrovnik – Croatia, May 15-18 [2] Matzen, D., Tan A. R., Andreasen, M., 2005, Product/Service-Systems: Proposal for Models and Terminology, 16. Symposium “Design For X”, Neukirchen, October 13-15 [3] Tucker, A., Tischner, U. , 2005, New Business for Old Europe – Product-Service Development, Competiveness and Sustainability, Greanleaf Publishing, Sheffield, UK [4] Meier, H., Kortmann, D., 2007, Leadership- From Technology to Use; Operation Fields and Solution Approaches for the Automation of Service Processes of Industrial Product- Service- Systems, In: Takata, S., Umeda, Y. (Hrsg.), Advances in Life Cycle Engineering for Sustainable Manufacturing BusinessProceeding of the 14th CIRP Conference on Life Cycle Engineering, p. 159- 163, Springer-Verlag London 2007 Limited, ISBN-Nr.: 978-1-84628-934-7 [5] Haarländer, N., Krallmann, H., 2006, Automatisierung der Komposition unternehmensübergreifender Geschäftsprozesse. In: Blecker, T., Gemünden, H. G. (Hrsg.), Wertschöpfungsnetzwerke – Festschrift für Bernd Kaluza, Berlin, S. 1-16. [6] Meier, H., Völker, O., 2008, Industrial ProductService Systems - Typology of Service Supply Chain for IPS² Providing, Proceedings of the CIRP-MS Conference, Tokyo, Japan, p. 485-488, SpringerVerlag, ISBN: 978-1-84800-266-1 [7] Pahl, G., Flemming M., Grundlagen der Konstruktionstechnik, In: Beitz, W., K.-H. Grote,

[8]

[9]

[10]

[11]

[12] [13]

[14]

[15]

[16]

[17] [18]

[19]

[20]

[21] [22]

[23]

[24]

310

Dubbel - Taschenbuch des Maschinenbaus, 19. Auflage, Springer Verlag, Berlin,ISBN 3-540-62467-8 Meier, H., Krug, C. M., 2008, An Approach for Sustainable Resource Planning by Modeling the Processes of IPS², Conference Proceedings - LCE 2008 - 15th CIRP International Conference On Life Cycle Engineering 2008, p. 535-540, self-published, ISBN: 1 877040 67 3 Meier, H., Krug, C. M., 2008, System for Planning of Resources in IPS²-Delivery, Proceedings of the CIRP-MS Conference, p. 469-472, Springer-Verlag, ISBN: 978-1-84800-266-1 Hess, T., 1996, Entwurf betrieblicher Prozesse, Dissertation der Universität St. Gallen, Gabler, Wiesbaden 1996. Zellner, G., 2004, Leistungsprozesse im Kundenbeziehungsmanagement: Identifizierung und Modellierung für ausgewählte Kundentypen, p. 3-31, Logos Verlag, Berlin. Kieser, A., Walgenbach, P., 2003, Organisation, 4. Aufl., Schäffer-Poeschel, Stuttgart. Seliger, G., Gegusch, R., Müller, P., Blessing, L., 2008, Knowledge Generation as a Means to Improve Development Processes of Industrial Product-Service Systems, Proceedings of the CIRP-MS Conference, p. 519-524, Springer, ISBN: 978-1-848000-266-1. Becker, J., Beverungen, D., Knackstedt, R., 2008, Wertschöpfungsnetzwerke von Produzenten und Dienstleistern als Option zur Organisation der Erstellung hybrider Leistungsbündel, In: Wertschöpfungsnetzwerke, Physika-Verlag, Heidelberg. DIN (ed.), 2002, Essential structure for the description of services in the procurement stage, PAS 1018, Beuth, Berlin. Uhlmann, E., Geisert, C., Hohwieler, E., 2008, Monitoring of Slowly Processing Deterioration of CNC-Machine Axes, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, Professional Engineering Publishing, vol. 222, no. 7 (2008). Weck, M., Platen, S., 2003, Intelligent main spindle functionality, wt-online 7/8-2003, p. 517. Ruuska, M., Andersson, P. H., 2003, Spindle Bearing Monitoring Using Acoustic Emission, Proc. of the XVII IMEKO World Congress, June 22-27, 2003, Dubrovnik, Croatia, p. 2164-2167. N. N., 2007, Spindle Data Logging Unit – Spindelintegrierter Datenlogger, installation guide and user manual, Walter Dittel GmbH. Jennings, N. R., 2000, On Agent-based software engineering, Artificial Intelligence 117 (2000), Elsevier, p. 277-296. Jennings, N, Wooldridge, M., 1996, Software Agents, IEE Review, January 1996, p. 17-20. Taveter, K., 2004, Agile Engineering of B2B Automation Systems, ERCIM News No. 58, Special: Automated Software Engineering, July 2004, p. 6061. Moukas, A., Guttman, R., Maes, P., 1998, Agentmediated Electronic Commerce: A Survey, The Knowledge Engineering Review, vol. 13, issue 2 (July 1998), Cambridge University Press, p. 147-159. Sauer, O., 2004, Agent technology used for monitoring of automotive production, Proceedings of the IMS (Intelligent Manufacturing Systems), International Forum 2004, May 16-18, Cernobbio, p. 308-316

Standardization of Service Delivery in Industrial Product-Service Systems H. Meier, C.M. Krug Ruhr-University Bochum, Chair of Production Systems, Bochum, Germany [email protected]

Abstract Industrial Product-Service Systems (IPS²) provide the best value in use for the customer. The high demand for this product model in the future, will lead to a high number of service deliveries. These service deliveries need to be executed in a service network in industry, but are planned and organized by a central intelligence hosted by the OEM of the IPS². This article will describe the possibility to minimize the derivation of execution time by the standardization of the boundary conditions in service delivery. An example scenario for resource planning with minimized derivation of execution time will be shown. Keywords: Industrial Product-Service-Systems; IPS²; Resource Planning; Standardization of Workplaces and Tools

1 MOTIVATION The intangible service shares of IPS² provide a value to the customer via the complete life cycle [1][2].There are some severe differences between the planning of industrial production and the planning of service delivery of IPS². These differences result from the characteristics of services like the immateriality and thus the following non-storability of service, the concurrent producing and consumption (uno-actu-principle) and integration of the customer in the delivery process [3]. One of the major differences is to be found in the variance of execution time. In industrial production, a much more precise forecast of execution time is possible, contrary to the service delivery of IPS² (Figure 1). The qualitative difference between the derivations of execution time is shown in the figure, the ordinate representing the number of performances. The higher uncertainty of the execution time leads to an imprecise planning of the delivery processes and the connected resources, due to the uncertain occupation time of the technical equipment, e.g. of a specialized measuring device. Furthermore the IPS² is occupied for an uncertain service time and is therefore not available for use for the customer. Within the planning of the delivery processes, an additional time, depending on the knowhow of the service technician is taken into account. It is a disadvantage concerning the warranty or availability of manufactured products, especially in availability- or resultorientated use models [4]. If it is possible to forecast the execution time more precisely, resource planning will be more accurate, the reduced resource demand will lead to a cost reduction and to an increase of customer satisfaction. 2 APPROACH Focusing this problem the first question is: Why does the variation of execution time differ that strong between the industrial production and the delivery of services? In industrial production every workplace is designed for its specific production step, that is possible due to the split up

CIRP IPS2 Conference 2009

311

of the whole production task into small parts [5]. Even today, through job enlargement and enrichment, the workscope is increased. By specifying the optimal position of the worker, parts and tools for every production step it allows to produce in clock cycle. Overall, there is the same boundary condition for every task at each production step. In service delivery of IPS² the task will be described in different degrees of detail, also depending on the kind of service in advance. Due to the integration of the external factor every delivery process has to be carried out with different boundary conditions. According to this, service delivery is a kind of extemporaneous process whose management depends on the operating experience of the service technician. Service delivery

n

t 1 t 2 t 3 t 4 t 5 t 6 t 7 t 8 t 9 t10 t11 Industrial production

t

n

t1 t 2 t 3 t 4 t 5 t 6

t Figure 1: Qualitative derivation of service processes and industrial production Following this conception the main objective is to generate standardized boundary conditions as far as possible in service delivery.

In order to reach this objective there are four fields of action to be considered: •







3

Workplace Like in industrial production, the workplaces of the technician have to be declared in advance. The designer has to consider ergonomic issues, i.e. the reach ability of parts and tools, and economic issues, e.g. time for the processing of parts. That means in practice, that the position of the technician and the position of his toolset are determined by the designer in advance. Simulation of the workplace with software tools like eM-Human [6] will support these decisions and will probably lead to changes in machine design. Tools Especially the extemporaneous tools are a common problem in service delivery today [8]. Only cost intensive tools are considered in the planning process; the other tools and their configuration are left to the service technician. The latter implicates a know-how depending on the execution time and in addition, the danger of losing this know-how in case of abrogation of the service technician. This means in practice that for a service process, which is described in detail, a set with required tools will be provided in which every tool has its certain place., thus supporting the technician by saving time for searching tools et cetera. Qualification and Support It is necessary that the service technician who is involved in the process has got the right qualification to manage the process. In case of a sufficiently qualified technician the IPS² provider has to take care that the process is carried out with support so that the planned process parts can be managed in time. This field of action implicates a good communication between the IPS² provider and the service technician located at the customer. Information and Communication It is the backbone of the service delivery in IPS². The use of information and communication technologies is a very important field of action, starting for example with the automated communication between the IPS² provider and the IPS² as well as his suppliers up to the outsourcing of service parts, e.g. failure diagnosis.

THE SEVEN SERVICES CLASSES AND THEIR ABILITY TO USE THIS APPROACH

3.1 The seven types of service Services which are relevant in the capital goods sector can be summarized in the following seven types of services: •

Planning services, e.g. material flow planning, factory planning



Counseling services, e.g. calculation support, personnel counseling



Training services, e.g. operator determination of the need for training



Logistic services, e.g. replacement part service, machine implementation

training,



Function creating services, e.g. start-up. Rampup management



Function maintaining services, e.g. maintenance, repair

• Optimizing services, e.g. process optimization There are differences in the services of a cluster in reference to operation, resource requirements and planning complexity. The individual service types can be given by criteria, which describe the process type, intensity of the interaction with the share of tangible goods, the dimension of the customer integration and automation degree [7]. Process types Four process types can be differentiated in reference to the description level of a service: •

Creative processes,



Expert processes,



Technical time-value processes and

• Logistic processes. In this context, the description of the process is detailed differently, but it is not considered for the standardization of the boundary condition. Logistic processes show the highest level of a standardization potential, followed by technical time-value processes, expert processes and creative processes. The standardization potential of a service process defines the degree of detail of a process representation in the context of the design of an Industrial Product-Service System. It is in particular the sub-processes of the provisional phase, which need to be described, depending on the process types, as, e.g. the results of a creative process are negatively influenced concerning its operation by the concrete specifications from the planning phase. A technical time-value process can be specified to a high degree by the process description in order to secure the result. Intensity of tangible and intangible product shares interaction The intensity of the interaction describes the mutual influence of tangible and intangible goods of an Industrial Product-Service System in the development phase. If there is a high intensity of interaction, there needs to be a special emphasis in the design of the process steps, on the adaptation with the development of the tangible goods. Dimension of customer interaction It is typical for the customer integration to have a large impact on all kinds of service processes, the influence of the customer on logistic processes, however, can be regarded as neutral. In case of industrial services the service is determined by the interaction with the customer’s object much more than the customer himself. The customer and respectively his processes determine the date of service as well, as he has to prepare the service process in his organization. Usually he has just a small influence on the service process itself, except for supporting activities such as switching on/off the air supply etc. Degree of automation The degree of automation of a service shows how far the processes can be executed without a member of staff. One example could be a teleservice solution to monitor the condition of an IPS².

312

Quadrant I: Due to the high creativeness or know-how and the parallel low dimension of customer interaction the standardization of boundary conditions is quite good. One example could be the standardization of the boundary conditions by setting up an expert workplace with defined software tools as well as the transfer of defined IPS² data. There will be just a raw description of the workflow to avoid blocking the creative processes.

Start of production

Quadrant III Due to the high degree of customer integration, the standardization is as hard as described before but qualified by the low creativeness and know-how, the detailed description of the process is possible and thereby the standardization of the boundary condition has a higher possibility than within the first two quadrants.

313

I

II

IV

III

low low

high

Dimension of customer interaction Figure 2: Use ability of the suggested approach 3.3 Example scenario For example, an upcoming machine breakdown was detected [9]. In order to manage the risk of a break down, normally a service technician is sent to fix the problem. Due to the description of the customer, the planner takes care that the service technician and all relevant resources are occupied for a time interval depending on the operating experience of the planner.

a)

Overall process

t

Repair

Assemble

t

Assemble

Repair

Split up processes Diagnosis

c)

Diagnosis

Dissasamble

b)

Dissasamble

Quadrant II: The high creativeness and also the high degree of customer interaction lead to problems. On the one hand the customer defines the boundary condition of the service delivery himself and on the other hand the creativeness and know-how does not allow a detailed description of the process. Therefore quadrant II is the hardest field in standardization of boundary conditions.

high

Quadrant IV The low creativeness and the low degree of customer interaction lead to the best condition for standardization of the boundary conditions. One example could be the change of a wear and tear element. Therefore, the definition of a workplace at an IPS² including a tool set with a fastener integrated in the IPS² was executed so that the service technician has to do well defined movements with calculable duration.

Process type -Creativeness and know-how-

3.2 Usability of the suggested approach The standardization of boundary conditions is just valid if the frequency of performances exceeds a certain level or if the service process and/or its consequences (e.g. downtime of the IPS²) respectively are connected to high costs. Altogether, the basic decision to standardization depends on costs for each service process over the life cycle of all IPS² in the market. The different service types do not allow deciding about the ability to use the suggested approach. This decision is to be made on a more detailed level of process description. The first hints are the special criteria of every service process. So the process type directly indicates the potential of detailed description of a process, depending on the creative demand of the process. The less creativeness is necessary, the more detailed process description is possible. This increases the degree of reproducibility and ensures a constant process quality. The dimension of customer interaction is important, regarding the boundary conditions of the service delivery. Therefore, the active interaction of the costumer should be as small as possible. Given the standardization of processes, automation is a maximum standardized process, whose properties are perfect for planning, due to its independency of the boundary conditions, except the scheduling of the customer. For a more detailed look on the usability the processes has to be split up into parts which have to be characterized by the explained criteria. Summarizing the properties of delivery processes discussed before, a portfolio of usability can be shown (Figure 2). As mentioned, this portfolio is just valid if the earned cost savings exceed the cost for standardization of the boundary conditions. This depends on the savings per process delivery and the number of performances or the cost intensiveness of the service delivery respectively. The two axis of the portfolio are the “Process type” especially the “creativeness and the know-how” and the “dimension of customer interaction”. Both axes are ranging from low to high. Therefore, there are four quadrants to divide the service process in their capabilities to use the suggested approach.

Forecasted end of production

t

Figure 3: Optimizing of execution time His operating experience follows the feedback of service technicians, who managed service processes similar to the current process (Figure 3, a). However, this feedback specifies the whole process and is to fuzzy for an efficient planning. He occupies the resources for a certain percentage of the maximum execution time. This is the

reason why the process square in Figure 3 (a, b) ends before the maximum forecasted end of production. The service consists of the sub processes disassembling, diagnosis, repairing and assembling. Each of these sub processes has its own variance in execution time (Figure 3, b). These specific variances depend on the time to find the right tool in an unsorted toolbox, the unspecified position of the service technician and the uncertain condition at the IPS². The aggregation of these specific variances leads to the overall variance in execution time noticed by the planner. So the aim is to disintegrate the overall process and to analyze which part could be optimized by the suggested approach. The disassembling, the repairing and the assembling of the machine are technical time-value processes with a very detailed description (Figure 2, quadrant IV). The diagnosis is an expert process with a high demand on creativeness and operation experience (Figure 2, quadrant I). So the process has to be split up in sub processes (Figure 4) that the design of the three technical sub processes can be done in detail considering the boundary condition of delivery in advance. Therefore, it may be possible to design a fastener for the tool set in an optimal position depending on the defined position of the service technician.

Expert process

Disassemble Diagnosis

Repair

Assemble

Technical time-value processes

Figure 4 Split up of a process The diagnosis of the failure is as announced an expert process. In order to manage this in an efficient way, information and communication technology can be used, e.g. by an automated transfer of machine data to an external expert in advance and sending him pictures after dissembling the machine by the service technician. The involvement of the expert by information and communication technologies guarantees the failure diagnosis in a certain time and allows concentrating the expert man power in one place by using the expert knowledge in several processes. This procedure allows sharpening the schedule. The specific variances in execution time become more accurate and the overall variance of execution time decreases strongly (Figure 3, c). 4 CONCLUSION The paper has shown that there is a difference between the planning of industrial production and the planning of service delivery. The latter is connected to the different

boundary conditions for every service delivery in opposition to the industrial production where the boundary conditions are always the same. So the suggested approach is to create a homogeneous boundary condition in service delivery. The basis for this process are standardized workplaces with standardized tool sets. The four groups of services, divided by creativeness/know-how and the degree of customer integration, feature different usability of the suggested approach to design the workplace at the IPS² for service delivery and to allocate a specialized tool set for the specific service delivery. It should be supported by the use of information and communication technologies and the qualification and the supporting of the service technician respectively. Finally an example scenario was given to demonstrate the advantages of suggested approach. 5 ACKNOWLEDGMENTS We express our sincere thanks to the Deutsche Forschungsgemeinschaft (DFG) for financing this research within the Collaborative Research Project SFB/TR29 on Industrial Product-Service Systems – dynamic interdependency of products and services in the production area. 6 REFERENCES [1] Matzen, D., Tan A.R., Andreasen, M., 2005, Product/Service-Systems: Proposal for Models and Terminology, 16. Symposium “Design For X”, Neukirchen, October 13-15 [2] Tan. A.R., McAloone, T.C., 2006, Characteristics of strategies in Product/Service-System Development, International Design Conference – Design 2006, Dubrovnik – Croatia, May 15-18 [3] Meier, H., Krug, C. M., 2008, System for Planning of Resources in IPS²-Delivery, Proceedings of the CIRP-MS Conference, p. 469-472, Springer-Verlag, ISBN: 978-1-84800-266-1 [4] Tucker, A., Tischner, U. , 2005, New Business for Old Europe – Product-Service Development, Competiveness and Sustainability, Greanleaf Publishing, Sheffield, UK [5] Taylor, F.W., 1995, Die Grundsätze wissenschaftlicher Betriebsführung: The Principles Scientific Management, BeltzPVU, Psychologie Verlagswissen, Weinheim, ISBN 3-621-27267-4 [6] Siemens Product Lifecycle Management Software, Product eM-Human: http://www.ugsplm.de/produkte/tecnomatix/human_pe rformance/em_human.shtml, Viewed on 15.08.2008 [7] Kortmann, D., Dienstleistungsgestaltung innerhalb hybrider Leistungsbündel, Dissertation, 2007, Shaker Verlag, ISBN 978-3-8322-6622-6 [8] Lampe, M., Strassner, M., Fleisch, E., 2004, A Ubiquitous Computing Environment for Aircraft Maintenance, ACM Symposium on Applied Computing 2004. Nicosia, Cyprus, March 14-17, 2004 [9] Ruuska, M., Andersson, P. H., 2003, Spindle Bearing Monitoring Using Acoustic Emission, Proc. of the XVII IMEKO World Congress, June 22-27, 2003, Dubrovnik, Croatia, p. 2164-2167.

314

Engineering Network Configuration: Transition from Products to Services Y. Zhang, J. Srai, M. Gregory, A. Iakovaki The Institute for Manufacturing, Cambridge University, Mill Lane, Cambridge CB2 1RX, the UK [email protected]

Abstract Existing approaches to the design and operation of engineering networks are largely product-oriented and pay little attention to the intangible, customer-involving and relationship-based nature of services. With the trend of servitization in manufacturing companies and the emergence of service science, manufacturers, particularly those who are engaged with complex and long-lifecycle products and systems, need to update their engineering networks to support integrated product-service offering. This paper develops a conceptual framework to demonstrate the configuration features of product- and service-oriented engineering networks. It will provide theoretical insight and practical guidance on the design and operation of integrated product and service systems. Keywords: Engineering Network Configuration, Service Engineering Networks, the Product Lifecycle, Integrated Productservice Systems

1

BACKGROUND AND INTRODUCTION

The research on international engineering operations is experiencing a stage of cross-discipline integration driven by the increasing complexity of engineering activities and the network of organisations involved in their delivery. Theoretical insights have been proposed to interpret new practices in coordinating internationally dispersed engineering activities. Examples include off-shoring [10], outsourcing [11], and global product development [1]. At the same time, practical guidance has been offered on the collaboration between different functions and organisations in dynamic business environments. Examples include concurrent engineering [12] and collaborative engineering [13]. In addition, increasingly capable information and communication technologies make it possible to bring together traditional computer aided engineering tools and business process management systems effectively. The concept of product lifecycle management has emerged to provide an integrated solution based on efficient communication and collaboration [14]. Most of the changes or transitions are heavily product-oriented with relatively little concern of the intangible and complex nature of services. Traditional engineering management concepts which were built on a simple assumption of stable business environments are challenged by the increasing complexity and uncertainty of engineering operations. New organisational forms have emerged to better coordinate dispersed engineering activities in dynamic business environments, e.g. matrix structures [15], centre(s) of excellence [16] and the virtual enterprise [17]. Emerging concepts and practices are converging on global engineering networks (GEN) for their efficient, flexible and innovative natures. Zhang, Gregory and Shi (2007) developed an integrating framework for GEN to guide the design and operation of internationally dispersed engineering systems [4]. However, the framework was based on case studies of new product development oriented engineering operations, and paid relatively little attention to service and support issues or through-life integration.

CIRP IPS2 Conference 2009

315

The decline of manufacturing and the rise of services is an important global development [18]. There is a trend for manufacturers to integrate services in their core product offerings [19]. The reason for having an integrated product-service system is multi-fold. For example, significant revenue can be generated from services [20]; customers are demanding more services [21]; or services can be a source of sustainable competitive advantage because they are less transparent and hence more sustainable [21]. However, operational issues are relatively neglected in service research largely due to the difficulty in defining services and service processes [2227]. There is little theoretical insight or practical guidance on how industries can effectively organise engineering resources or coordinate engineering activities to support integrated product and service offering. Existing research on engineering networks is largely product oriented and pays relatively little attention to the intangible, customer-involving and relationship-based nature of services. With the trend of servitization in manufacturing companies [6] [19] and the emergence of service science [7], manufacturers, particularly those who are engaged with complex and long-lifecycle products and systems, require a better understanding of the nature of services and its impact on the design and operation of engineering networks. Booz, Allen and Hamilton (2006) estimated that the worldwide spend on engineering services will exceed $1.0 trillion by 2020, and about one quarter of the activities will be off-shored or out-sourced to emerging economies [28]. This will lead to a radical change to engineering network design and operation, not least in order to address the issues of geography dispersion, international inter-firm collaboration, customer relationship management and through-life integration. The reported research aims to understand how to design and operate engineering networks to effectively support integrated product and service offering. This paper develops a conceptual framework to demonstrate the different sets of organisational requirements for new product development oriented engineering operations and service and support oriented engineering operations.

Research on Global Engineering Networks Research on International Supply Networks Research on International Manufacturing Networks

The Preliminary Framework

The Refined Framework

The Tested Framework

Literature on the Product Lifecycle Management and Services

Exploratory Case Studies to Refine and Validate the Framework

Two Pilot Applications to Test the Applicability

Figure 1. An overview of the research design. Integrated product-service systems (IPS2) should meet the requirements from both product-orientation and service-orientation. The framework integrates the existing theoretical and empirical knowledge on engineering management, services and network organisations. Key elements of engineering network configuration are defined based on the research of engineering networks [4], supply networks [8] and manufacturing networks [9]. The elements are refined and adjusted to reflect the intangible, customer-involving and relationship-based nature of service and support oriented engineering operations. The framework was improved and validated through exploratory case studies and tested by pilot applications. Main directions for future research and fundamental propositions are discussed based on this framework. 2. RESEARCH APPROACH The reported research in this area has developed the framework for engineering network configuration largely through case studies. A preliminary framework was proposed through integrating the outputs of the research programmes for global engineering networks (GEN), international supply networks (ISN) and international manufacturing networks (IMN). The preliminary framework was further developed to reflect the influence of inter-firm relationships, product lifecycle management and service orientation; and then refined and validated with exploratory case studies in different industry sectors. The applicability of the framework was tested with two pilot applications. Figure 1 presents an overview of this research. The configuration framework was based on a big of number of case studies to understand network configuration from different perspectives, including • Global engineering networks: 7 in-depth cases in the industry sectors of aerospace, automobile, electrics and electronics, and fast moving and consumer goods (FMCG) [4]. • International supply networks: 10 exploratory cases and 10 in-depth cases in the industry sectors of aerospace, electronics, FMCG, garments and pharmaceuticals [8]. • International manufacturing networks: 15 cases in the industry sectors of aerospace, electronics, heavy engineering and pharmaceuticals [9]. A generalised framework to describe network configurations was developed through integrating the above cases. The configuration framework was later refined and validated with exploratory case studies to capture configuration archetypes for product oriented engineering networks and service oriented engineering networks. Case companies with product oriented engineering networks usually give higher priorities to new

product development related engineering activities (see Figure 2). Companies with service oriented engineering networks believe that service and support are critical to their business (a simple indicator is that a significant amount of revenues are from services). Table 1 presents a list of the exploratory cases. The framework was later tested by two pilot applications: case Y and case Z. The two pilot cases were selected with the criteria being to minimise the influence of contextual factors, and to demonstrate the difference between product orientation and service orientation. To meet the first criteria, both cases were selected from the same industry sector, and common geographic region, and were firms of similar and significant scale. (Both companies are within the top 5 service firms in the selected sector and region). Addressing the second criteria, the two engineering networks focus on different parts of the product lifecycle. One of them leads in design and manufacturing of complex equipment in the defence sector. The other case company is a leading after-sales and maintenance service provider in the same sector. The results of these pilot cases are discussed in Section 4. Cases

Industry Sector

Revenues (in 2007)

A B C D E F G H I J

Automobile Electrics FMCG Aerospace Automobile Electronics Aerospace Aerospace IT services Petrol Chemical

$174 billion $29 billion €40 billion $820 million £310 million £160 million £7.4 billion €25 billion US$22billion US$284 billion

Product / Service Orientation product product product product product product service service service service

Table 1. An overview of the cases. 3 LITERATURE ON SERVICE ENGINEERING NETWORK CONFIGURATION 3.1 Engineering activities through the product lifecycle Engineering operations focus on different activities along the product lifecycle (see Figure 2). Product oriented engineering operations (e.g. new product development) usually set a high priority to the activities from initial concept to manufacture. Service oriented engineering operations (e.g. service and support) usually set a high priority to the activities from manufacture to disposal. In reality, the activities are not isolated but interrelated, sometimes requiring iterative development. The design and operations of integrated product-service systems should be based on an overall understanding of engineering activities along the product lifecycle.

316

Engineering Activities along the Product Lifecycle Idea & Concept

Design & Development

Service & Support

Manufacture

Disposal & Recycle

Product Oriented Engineering Operations •

explore new technologies



explore new market opportunities



generate new ideas of products or services



generate concepts for product/ service improvement



define product/service concepts and features



define product transfer requirement



assess local capability and fitness for transfer

assess concessions



process engineering changes





confirm build standard meets design standard



capture customer feedback



product enhancement



identify/forecast future sales



manage contractor and partner



review product effectiveness

selection of equipment and services providers



process equipment obsolescence management



issue technical instructions



review readiness for production



continuous improvement across the mfg. network





complete maintenance/support policies



prove compliance with operational capability specification



service and support concept adjustment



validate and produce full support and training packages



prepare maintenance test procedure



produce initial support facilities and equipment



produce product support plan

• •

complete engineering definition



conceptual/detailed design



prototype development



mfg. process design



organisation & system design



down select option solutions against customer • requirements



market & technical assessment



business & financial analysis



business case development



complete overall product specification



negotiate amendments to customer requirements





produce and agree specification



verify supportability performance



product introduction & ramp-up



product de-bugging



production trials

• •



potential supplier assessment



resource viability confirmation

complete level of repair analysis







finalise frequency of maintenance tasks





service and support concept development



review decommissioning element of engineering plan

knowledge transfer



support board of inquires and answer queries

decommission all product unit



identify hazardous material



identify high value salvage material for recovery

issue repair schemes



propose resale

document and implement product specification changes



product components obsolescence and upgrade

archive documentary evidence as required by legislation





review service support capability

archive design information



support planning and coordination with customer



recover security or IP sensitive material

define in-service maintenance procedure



support on customer locations



support via in house locations

manage maintenance suppliers



external support

Service Oriented Engineering Operations

engineering strategy development, engineering standards setting, project management, contracting and approvals, learning and development, product safety, resources and recruitment, research and technologies, terms and conditions, tool sets and support people, best practice identification and transfer, documentation processes, lifecycle management processes, project management processes

Figure 2. Engineering activities along the product lifecycle [4]. Product lifecycle management (PLM) in an engineering context aims to optimise engineering operations across the product lifecycle. It has been proposed as a strategic approach to creating and managing product related information from the initial concept, through design, development and manufacture, to service and disposal [14]. It provides a common platform for the synergies of technologies, processes, resources and business systems throughout the product lifecycle [29]. PLM systems are usually enabled by IT solutions for product and portfolio management, product design, manufacturing process management, and product data management. These solutions have been adopted and shown benefits in a wide range of industry sectors, especially the aerospace and automotive industries [30]. A critical issue to the implementation of PLM is to develop business strategies and operational processes across the different stages of a product lifecycle.

happens simultaneously. In addition, a customer may experience a service differently each time. This kind of heterogeneity makes it difficult to analyse the process of services or measure the outputs. Furthermore, services are perishable. It is generally not possible to stock a service for future use if it is not consumed when available. Finally, human aspect is a core element for service operations because of the significant people involvement in the process of service production and service provision. To be successful, service providers need to be customer centric- adapting to their often real-time dynamic needs whilst collaborating on both solution design and codelivery. Distinguishing Features of Services

Intangibility (heterogeneity)

3.2 Services and engineering operations Services have been considered as the application of specialised competencies through deeds, processes and performances for the benefit of another entity or the entity itself [31]. They are fundamentally different from physical products on the basis of intangibility, simultaneity/inseparability, perishability, heterogeneity, human involvement and customer contact [27]. Services can be partly intangible with the process of services being the application of specialised skills and knowledge. Physical products/goods are usually the distribution mechanism for service provision. At the same time, customers may contribute to the production process of services; and the production and consumption of services

317

Co-creation of Value (simultaneity, inseparability)

Relationship (human/customer centric)

Product Oriented Engineering Operation (e.g. new product development) • well defined specifications • measurable and prespecified outcomes • standardised processes and outcomes • value is determined by the producer/engineering service provider • customers could be separate from the value creation process • it is possible to store outcomes • • •

transition oriented relatively low impact of human aspect relatively low customer centric

Service Oriented Engineering Operation (e.g. service and support) • output based ‘service level agreements’ • subjective and userdependent outcomes • variable processes and outcomes • value is perceived and partly co-determined by the customer • customers are involved in the value creation process • unable to store outcomes but it is possible to store service capability • relationship oriented • high impact of human aspect • highly customer centric

Table 2. A comparison of product-oriented and serviceoriented engineering operations.

In brief, a service-centred view is participatory and dynamic. The value of service provision will be maximised through an iterative learning process on both the service provider and the customer. The logic of service processes is focused on intangible resources and the co-creation of value through mutually benefiting relationships [31]. Physical products oriented methodologies and theories are challenged by the increasing importance of services in the field of operations management. The unique nature of services requires practitioners and researchers to think about their business strategies and operational processes from a new perspective [32]. This radically changes the principles for engineering network design and operation. Table 2 presents the distinguishing features of services and their impact on engineering operations. 3.3 Engineering network configuration Literature on strategic management and organisational studies implies that organisations function effectively because they put different characteristics together in complementary ways [33-37]. Miller (1986) observed that organisational features are usually interrelated and mutually reinforcing [35]. Organisations might be driven towards common configurations to achieve internal harmony among the elements of structure, environment and strategy. Organisational parameters should be logically configured into internally consistent groupings composed of tight constellations of complementary elements. This concept of cohesive configuration could be predicatively useful for the study of organisational structures and organisational capabilities because the number of possible ways in which constructional elements are combined is reduced. For example, Mintzberg (1979) identified five viable organisation configurations (i.e., simple bureaucracy, machine bureaucracy, professional bureaucracy, divisionalised form, and adhocracy) based on their features of structures (e.g. operating core, strategic apex, middle-line, techno-structure, and support staff) and coordination mechanisms (e.g. direct supervision, mutual adjustment, standardisation of work, outputs, and skills) [34]. Ghoshal & Nohria (1993) articulated four configurations of multinational corporations with the dimensions of differentiation and integration, i.e., ad hoc variation, structural uniformity, differentiated fit, and integrated variety. These configurations would be apt for different environmental conditions, each with different degree of requirement for global integration and local responsiveness [36].

strategic capabilities was investigated to explain the current transformation towards more globally integrated or coordinated configurations. Srai and Gregory (2008) considered the configuration of supply networks as the particular arrangement or permutation, of the supply network’s key elements including, the network structure of the various operations within the supply network and their integrating mechanisms, the flow of materials and information between and within key unit operations, the role, interrelationships, and governance between key network partners, and the value structure of the product or service delivered [8]. Each supply network configuration would exhibit different intrinsic capabilities. Exemplar supply network capability and configuration profiles were identified through establishing specific capabilityconfiguration relationship patterns, e.g. distinct approaches to end-to-end network integration, highly responsive risk-pooling supply models, global scale contract manufacture, mass customisation on-demand, product-service integration and alternative types of multidomestic product supply. This could help determine a supply network’s potential for re-configuration, i.e. the ability to rearrange key elements of the supply network to enable improvements in the supply or development of products or services. Zhang, Gregory and Shi (2007) proposed an overall framework for the design and operation of global engineering networks (GEN) from the perspective of context, capability and configuration [4]. Key organisational elements of GEN include network structure, coordination, governance and support. The configuration of an engineering network could be described with the features of the above elements, e.g. the degree of dispersion and interdependence of network structure, the degree of standardisation of network coordination, the degree of centralisation of network governance, and the degree of unification of network support. An integrated GEN configuration is characterised by concentrated and interdependent engineering centres, formal and structured coordination, detailed and operational governance, and uniform support across the network. Configuration Elements

Global Engineering Networks [4]

International Supply Networks [8]

Plant’s characteristics; geographic dispersion

Structure, including geographic dispersion, resources and roles of engineering centres, and rationales for network structure design

The network structure of the various operations within the supply network and their integrating mechanisms

Operations Flow and Processes

Horizontal/vertical coordination; operational mechanisms; dynamic response mechanisms; product lifecycle and knowledge transfer

Coordination, including operational processes and coordination mechanisms

The flow of materials and information between and within key unit operations

Governance and Coordination

Dynamic capability building and network evolution

Governance, including authority structure and performance measures

The role of and governance mechanisms between key network partners

The configuration approach has been increasingly adopted in research on operations management, e.g. engineering operations [4], supply chain management [8] and international manufacturing [9]. By doing so, the researchers are able to simplify and classify network systems, and capture their characteristics and capabilities. Shi and Gregory (1998) identified eight international manufacturing network configurations according to their degree of geographic dispersion and coordination, i.e. home based manufacturing, home based global manufacturing, regionally uncoordinated manufacturing network, regional exporting manufacturing network, multidomestic manufacturing network, global integrated manufacturing network, global-local manufacturing network, and global coordinated manufacturing network [9]. A set of structural and infrastructural elements of manufacturing systems were used to describe the configurations, including factory as a node, geography dispersion, horizontal coordination, vertical integration, response, knowledge sharing, operation and evolution. The relationship between network configurations and

International Manufacturing Networks [9]

Structure

Support Infrastructure

Relationships

Support, including tools, IT systems, and people The role and interrelationships, between key network partners

Table 3. Key elements of network configuration.

318

It usually demonstrates strong capability for integration and synergising. An autonomous GEN configuration is characterised by dispersed and independent engineering centres, informal and unstructured coordination, generic and strategic governance, and customised support for customers, technologies or countries. It usually demonstrates strong capability for adaptation and restructuring. There are also engineering networks with strong capabilities for innovation and learning. They are configured between the two extremes between integrated GEN and autonomous GEN. Table 3 presents the key elements employed by the researchers studying the ‘configuration of network organisations’. For an engineering network involving multiple players, taking a multi-organizational perspective, these individual research strand inputs can be usefully integrated as bellows. •









Structure refers to the physical footprint of engineering resources, including the size, number and types of Engineering centres, the rationale for location decision, and interrelationship and resource sharing between engineering centres. Network structures are characterised by the degree of dispersion (dispersed vs. concentrated), and the interdependence between engineering centres (independent vs. interdependent). Operational Flow (& Processes) refers to the flow of material and information between members of the network to create valuable output to customers, e.g. new product development processes, lifecycle management processes, supply chain management processes, service and support processes. Operational flows are characterised by their degree of standardisation (standard vs. tailored /bespoke). Governance (& Coordination) refers to the mechanisms to direct and control the network, especially authority structures and performance measurement systems. Governance mechanisms are characterised by their degree of centralisation (centralised vs. decentralised) for commercial control, engineering authority and metrics. Support Infrastructure refers to enablers for network members to collaborate with each other, especially engineering tools, information systems, engineering resource, people, culture and behaviours. Network support are characterised by their degree of unification (uniform vs. customised) and globalisation (global vs. local). Relationships refer to the interaction with internal/external partners, especially suppliers, customers and users. Network relationships are characterised by their strategic importance (strategic vs. tactical), degree of trust (trust vs. transactional), and scope (global vs. local).

synergies, resource sharing, and reusing existing knowledge and solutions. At the same time, an engineering network may seek for greater effectiveness through quick response to environmental changes, market/technology driven innovation, mobile engineering resources, and flexible operation approaches. Zhang, Gregory and Shi (2007) differentiated two types of effective engineering networks. One focuses on innovative product development and the other focuses on strategic flexibility [4]. Thus, engineering networks could be configured with strategic intent for efficiency, innovation and flexibility (see Table 4). An efficient engineering network aims to achieve efficient operations on a global scale through minimising waste and maximising value and capability utilisation, e.g. economies of scale/scope, international operations synergies, leveraging expertise or precious resources on a global scale, sharing and reuse existing solutions. It is appropriate for complex products/services in relatively stable business environments. An innovative engineering network aims to satisfy business and customer needs effectively through new product/service/process development, e.g. customer intimacy, technology leadership, and market/technology driven innovation, learning across disciplines or organisations, leaving room for creativity or diversity. It is appropriate for simple products/services in relatively dynamic business environments. A flexible engineering network aims to improve the ability of the network to adapt to uncertain circumstances though flexible working approaches, mobile engineering resources and reconfigurable network structures, e.g. local responsiveness, and quick response. It is appropriate for complex products/services in dynamic business environments.

Efficiency • economies of scale/scope • international operations synergies • leveraging expertise or precious resources • sharing and reusing knowledge or existing solutions

Innovation

Flexibility

• technology leadership/techn ology-driven innovation

• reconfigurable network structure

• customer intimacy/marketdriven innovation • learning across disciplines, businesses, and organisations • leaving room for diversity and creativity

• mobile engineering resources • flexible working approaches • quick response to environmental changes

Table 4. The performance preference of global engineering networks [4-5].

3.4 Deliberate intent for engineering network configuration

4. ENGINEERING NETWORK CONFIGURATION FRAMEWORK AND PILOT TESTING

Traditional engineering systems were organised for the effectiveness (e.g. the project approach) and efficiency (e.g. the functional approach) of engineering operations. Effectiveness indicates how closely the output of an engineering system meets its goals or customer needs; and efficiency indicates how economically the resources are utilised to produce the output [38]. Zhang, Gregory and Shi (2008) revealed the strategic intent of different forms of engineering networks from an evolutionary perspective [5]. The study demonstrates that an engineering network may seek for greater efficiency through economies of scale/scope, international operation

Figure 3 presents an overall framework for engineering network configuration along the product lifecycle. The exploratory cases of product-oriented engineering networks and service-oriented engineering networks demonstrate different configuration characteristics to support their strategic intents for efficiency, innovation and flexibility. For product oriented engineering operations, an integrated engineering network configuration usually demonstrates strong capability for efficiency; and an autonomous engineering network configuration usually demonstrates strong capability for flexibility.

319

Engineering Network Configuration

Engineering Operations along the Product Lifecycle Product Orientation

Service Orientation

Efficiency

Innovation

Flexibility

concentrated and specialised resources close to manufacturing bases

dispersed resources close to technology bases or customers/users

dispersed and independent resources close to customers or users

Operations Flow

common processes

common processes for reference

local processes for customer needs

Governance

centralised control

centralised control on major operations

local authority

centralised control

Structure

Efficiency dispersed and specialised resources close to customers and manufacturing bases common processes tailored for customer needs

Support Infrastructure

uniform support

uniform support on major operations

customised support

uniform support customised for customer needs

Relationships

strategic partnership with suppliers

strategic partnership with suppliers on key programmes

transitional relationship with suppliers

strategic partnership with suppliers and customers

Exploratory Cases Pilot Cases

Case A, E

Case B, C, F

Case D

Innovation

Flexibility

dispersed resources close to customers and technology bases

dispersed and independent resources close to customers

common processes tailored for customer needs centralised control on major operations uniform support customised for customer needs strategic partnership with suppliers and customers on key programmes

Case H

Case Y

local processes for customer needs local authority uniform support customised for customer needs strategic relationship with customers and transitional relationship with suppliers

Case G

Case I, J Case Z

Figure 3. Engineering network configuration framework. Service and support focused engineering operations have a different set of features due to the nature of services in intangibility, customer-involvement and relationship-based. The network structure tends to be dispersed with customers, the process, governance, and support system are usually tailored for customer needs, and the relationships with customers and users are critical to successful engineering operations. This framework would enable industries to optimise their current engineering networks or design new engineering networks for integrated product and service offering. The design and operation of integrated product-service systems should consider the requirements from both product orientation and service orientation. Companies can assess the current situation of their engineering networks against the configuration elements and optimise their engineering networks through aligning these elements to their major strategic objectives. They can also use the framework as a template to design new engineering networks according to environmental changes, e.g. the increasing importance of inter-firm collaboration or the demand for through-life engineering capabilities. This is particular helpful for companies in their transition process from traditional manufacturers to services providers, as observed in the pilot application cases. The definitions and categorisations of the configuration elements and the capture of strategic intents, in this paper have been piloted in two complex equipment service providers: case Y and case Z. Case Y is a complex equipment designer and manufacturer. It is the regional operation of an international defence company. Case Y employs approximately 2,600 people at over 50 sites in the operational areas of global network, marine, air and land. Case Z is a regional scale equipment and maintenance service provider. It employs about 2,500 people in the operational areas of aerospace, land, marine and electronic systems. Case Z’s engineering operations are dispersed in a large number of centres across the region. The two companies are currently in the process of post merger integration. The merged business aims to be the leading through-life capability partner of the local government. For each of the pilot cases, the engineering network configurations were mapped through interviews with groups of front-line managers and in workshops involving senior engineering managers. The configuration

framework was used to capture the ‘current state’ and providing a basis for exploring reconfiguration options for the merged business (see Figure 4). Both cases give relatively high priority for innovation in concept development. Achieving greater efficiency is the key strategic intent in manufacturing and support, while short/medium term flexibility and innovation is required to respond to well-established platforms and capability inadequacies (especially for case Z). Case Z’ engineering operations focus more on service and customer support. Therefore, its engineering resources are relatively dispersed and independent; its operating processes tend to be adaptable and informal; its governance system is relatively decentralised; its support systems are more customised to local needs; and its relationships with suppliers are relatively weak. All the configuration features of the above two cases reasonably comply with the archetype expectations of product-centric and servicecentric organisations, and are reflected in the conceptual framework (Figure 3) except for the relationship with customers. In theory, a service oriented engineering network like case Z should have stronger relationship with customers or users. But the mapping result shows that case Y and case Z have a similar level of partnership with customers; and case Y’s relationships with customers are perhaps even stronger. The active transition of case Y into progressively more service based operations, and the consequent mindset changes this brings, may influence the mapping outcome. The pilot cases also demonstrate the need for integrating product- and service- oriented configuration features. The proposed configuration profile for the merged business (see Figure 4) encompasses configuration features from both product orientation and service orientation. The feedback from the pilot case studies shows that the framework provides a common language which will help dispersed engineering centres and different functions communicate with each other to achieve good consensus or identify common problems. At the same time, the working tools developed from this framework (e.g. tools to identify key success factors, to assess strategic intents, or to generate and evaluate configuration options) help demonstrate a high-level vision of a company’s engineering network while breaking the whole issue into manageable elements. This will facilitate companies to form their global engineering strategy and to identify the critical issues in the process of transition.

320

Engineering Network Configuration

Structure Dispersion Interdependence

Operations Flow Standardisation

Governance & Coordination

Global/Corporate Centres

Global/Corporate Centres for Key Activities

Regional/Group Centres

BU Centres

Local Centres

Interdependent across the World/Corporation

Interdependent on Key activities across the World/Corporation

Interdependent within Regions/Groups

Interdependent within BU

Independent Centres

Common Processed across the World/Corporation

Common Processes for Common Processes key Activities across the within Regions/Groups World/Corporation

Common Processes Tailored for Local Needs

Ad-hoc Practices

Global/Corporate Control

Regional/Group Control Regional/Group with Central Influence Control

Distributed Control with Central Influence

Distributed Local Control

Global/Corporate Control

Regional/Group Control Regional/Group with Central Influence Control

Distributed Control with Central Influence

Distributed Local Control

Common System across the World/Corporation

Common Basic System Common Systems within across the Regions/Groups World/Corporation

Common Systems within BU

Local Systems

Global/Corporate Resource Management

Global/Corporate Management of Key Resources

Regional/Group Resource Management

Resource Management within BU

Local Resource Management

Global/Corporate Culture

Global/Corporate Orientation on Key Activities

Regional/Group Cultures

BU Cultures

Local Cultures

Supplier Partnership

Global/Corporate Partnership

Strategic Partnership on Regional/Group Partnership Key Activities

BU Partnership

Transactional Relationship

Customer Partnership

Global/Corporate Partnership

Strategic Partnership on Regional/Group Partnership Key Activities

BU Partnership

Contractual Relationship

Commercial Control Engineering Control

Support Infrastructure Engineering Systems Engineering Resources Culture

Relationships

Proposed configuration for Case Z the merged business Figure 4. Engineering network configurations of case Y and case Z.

Case Y

Future research will further develop the framework through in-depth case studies across a variety of industry sectors. The follow-up study will pay particular attention to the comprehensiveness of the configuration elements and their inter-relationships, as well as the potential trade-offs between performance preferences or strategic intents. At the same time, organisational characteristics of engineering networks to achieve specific strategic intents will be captured. Typical combinations of the configuration elements (or configuration archetypes) will be identified in a wider range of business saturations. Configuration features of service oriented engineering networks and product oriented engineering networks will be further investigated.

Theoretically, this conceptual framework extends the theory of engineering network configuration from product oriented engineering operations to service oriented engineering operations. It improves the understanding of integrated product and service systems from an organisation perspective. It also contributes new insights to the knowledge domains such as organisational configuration, network organisations, product lifecycle management, and service science. Practically, this research can support industries to effectively design and operate their engineering networks for integrated product and service offering. The transition challenges in moving to a service oriented business are highlighted as the critical engineering dimensions in a service environment have been identified.

5. CONCLUSION

However, the framework stems from three strands of research, i.e. global engineering networks, international supply networks and international manufacturing networks. This obviously provides benefits in cross-discipline learning; but at the same time brings challenges in consistency and compatibility. Further integration and more empirical studies would improve the validity and reliability. Future research will refine and test the framework, thus far validated in a number of exploratory cases and piloted in two complex equipment service providers. These follow-up in-depth case studies will aim to identify the configuration features of engineering networks in different service contexts and explore the transition themes in further detail, not least the impact on human resources, people and culture aspects, required capabilities, and performance measures.

This paper demonstrates different organisational requirements for engineering activities along the product lifecycle with a focus on product oriented engineering operations and service oriented engineering operations. It reveals the intangible, customer involving and relationship based nature of services, and assesses its impact on engineering operations. At the same time, this paper identifies the key elements of ‘engineering network configuration’, including network structure, processes, governance, support and relationships. These elements could be configured into consistent constellations with strategic intents for efficiency, innovation and flexibility. This paper concludes by integrating the above insights and proposes a conceptual framework for engineering network configuration along the product lifecycle.

321

6. REFERENCES [1] Eppinger, S. D. and Chitkara, A. R. 2006. “The New Practice of Global Product Development”, Sloan Management Review, 47(4), 22-30. [2] Karandikar, H. and Nidamarthi, S. 2006. “A Model for Managing the Transition to a Global Engineering Network Spanning Industrialized and Emerging Economies”, Journal of Manufacturing Technology Management, 17 (8), 1042-1057. [3] Zhang, Y., Gregory, M. & Shi, Y. 2006. “Foundations of Global Engineering Networks: Essential Features of Effective Engineering Networks”. Paper presented on the IEEE International Conference on Management of Innovation and Technology, Singapore. [4] Zhang, Y., Gregory, M. & Shi, Y. 2007. “Global Engineering Networks (GEN): The Integrating Framework and Key Patterns”, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, 221, 1269-1283. [5] Zhang, Y., Gregory, M. & Shi, Y. 2008. “Global Engineering Networks (GEN): Drivers, Evolution, Configuration, Performance, and Key Patterns”, Journal of Manufacturing Technology Management, 19(3), 299-314. [6] Ren, G., and Gregory, M., 2007. “Servitization in Manufacturing Companies: A Conceptualisation, Critical Review and Research Agenda. The 16th Annual Frontiers in Service Conference, San Francisco, California, USA. [7] IfM and IBM, 2008. Succeeding through service innovation: A service perspective for education, research, business and government. Cambridge University, Cambridge, UK. [8] Srai, J., and Gregory M., 2008. “A Supply Network Configuration Perspective on International Supply Chain Development”, International Journal of Operations Management, 28(5), 386-411. [9] Shi, Y. and Gregory, M. 1998. “International manufacturing networks: to develop global competitive capabilities”, JOPM, 16, 195-214 [10] PTC. 2005. Gaining Competitive Advantage through Global Product Development. White Paper, Parametric Technology Corporation (PTC). [11] BusinessWeek Research Services. 2006. Global Engineering Development (GPD) - moving from strategy to execution. BusinessWeek Research Services Reports, NY. [12] Backhouse, C. J. and Brookes, N. J. 1996. Concurrent Engineering: what’s working where. Design Council, Gower. [13] Willaelt, S. S. A, De Graaf, R. and Minderhoud, S. 1998. “Collaborative engineering: A case study of concurrent engineering in a wider context”, Journal of Engineering Technology Management, 15, 87-109. [14] Subrahmanian, E., Rachuri, S., Fenves, S., Foufou, S. and Sriram, R. 2005. “Product lifecycle management support: a challenge in supporting product design and manufacturing in a networked economy”, International Journal of Product Lifecycle Management, 1(1), 4-25. [15] Payne, A.C., Chelsom, J.V., Reavill, L.R.P. 1996, Management for Engineering, Wiley, NY. [16] Reger, G. 2004. “Coordinating globally dispersed research centres of excellence- the case of Philips Electronics”, Journal of International Management, 10, 51– 76. [17] Powell, A., Piccoli, G. and Ives, B. 2004. “Virtual teams: A review of current literature and directions for future research”, ACM SIGMIS Database, 35(1), 6-36.

[18] World Bank. 2006. World development indicators, Washington, DC: World Bank. [19] Oliva R. and Kallenberg, R. 2003. “Managing the transition from product to services’’, International Journal of Service Industry Management, Vol. 14, No.2, 160-172. [20] Poole K. 2003. “Seizing the Potential of the Service Supply Chain’’, Supply Chain Management Review’’, July – August, 54-61. [21] Desmet S., Van Dierdonck R., Van Looy B. “1998. “Services in a world economy” in Services Management: An Integrated Approach, FT Prentice Hall, 45-57. [22] Ellram L., Tate W., Billington C., (2004), ‘’Understanding and managing the services supply chain’’, Journal of Supply Chain Management, 40 (4), 17-32. [23] Metters, R. nad Marucheck, A. 2007. “Service Management- Adademic Issues and Scholarly Reflections from Operations Management Researchers”, Decision Sciences, 38(2), 195-214. [24] Bretthauer, K., 2004. “Service management”, Decision Sciences, 35(3), 325-332. [25] Chresbrugh, H., and Spohrer, J. 2006. “A research manifesto for service science”, Communications of the ACM, 49(7), 35-40. [26] Sengupta H., Heiser R., Cook S. 2006. ‘’Manufacturing and service supply chain performance: a comparative analysis’’, Journal of Supply Chain Management, Vol. 42, Issue 4. [27] Baltacioglu, T., Ada, E., Kaplan, M. D., Yurt, O. And Kaplan Y. C. 2007. “A new framework for service supply chains”, The service industries Journal, 27(2), 2007. [28] Booz, Allen & Hamilton. 2006. Globalisation of Engineering Services. Industrial Report, August, New Delhi. [29] Ameri, F. and Dutta, D. 2005. “Product Lifecycle Management: Closing the Knowledge Loops”, Computer Aided Design & Applications, 2(5), 577-590. [30] Grieves, M. 2006. Product Lifecycle Management: Driving the next generation of lean thinking. New York: McGraw Hill. [31] Vargo, S.L. and Lusch, R.F., 2004. “Evolving to a New Dominant Logic for Marketing”, Journal of Marketing, 68, 1-17. [32] Wise, R. and Baumgartner, P. 1999. “Go downstream: the new imperative in manufacturing”, Harvard Business Review, 77(5), 133-141. [33] Child, J. 1972. “Organisational Structure, Environment and Performance: The Role of Strategic Choice”, Sociology, 6(1), 1-22. [34] Mintzberg, H. 1979. The Structuring of Organisation. Prentice Hall, Englewood Cliffs, NJ. [35] Miller, D. 1986. “Configurations of Strategy and Structure: Towards a Synthesis”, Strategic Management Journal, 7, 233-249. [36] Ghoshal, S. and Nohira, N. 1993. “Horses for coursesorganisational forms for multinational corporations”, Sloan Management Review, 34, 23–35. [37] Mintzberg, H., Bruce, A. and Joseph, L. 1998. The Strategy Safari: A guided tour through the wilds of strategic management, London: Prentice Hall. [38] Neely, A., Gregory, M., Platts, K. 1995. “Performance Measurement System Design”, International Journal of Operations & Production Management, 15(4), 80-116.

322

Roadmap to Self-Serving Assets in Civil Aerospace 1

A. M. Brintrup1, D. C. Ranasinghe1, S. Kwan1, A. Parlikad1, K. Owens2 University of Cambridge, Institute for Manufacturing, Mill Lane 16, Cambridge CB2 1RX, UK 2 The Boeing Company [email protected]

Abstract The “intelligent object” paradigm first occurred in holonic manufacturing, where objects managed their production. The “self-serving asset” is a further evolution of those early concepts from manufacturing to usage phase. The usage phase bestows a different set of requirements including maximisation of the asset’s life-in-service and benefits to the asset’s stakeholders. Addressing these requirements with a selfserving asset may lead to more streamlined decision-making in service operations, reduce erroneous or suboptimal decisions, and enable condition-based maintenance. We present a future direction for service systems by considering self-serving assets in the aerospace industry, and outline a technology roadmap for the transformation. Keywords: Multi-agent systems, Self-serving asset, Intelligent object, Aerospace service supply chain

1 INTRODUCTION The intelligent object has a unique identity, is capable of communicating with its environment through sensors and languages, and can participate in decisions concerning its destiny [1]. Unique identity can be possessed using barcodes or automated identification systems such as Radio Frequency Identification (RFID). Communication ability can be obtained using sensors, wireless networks, common interfaces and languages. Participation in decisions can mean providing raw data to processing mechanisms such as Decision Support Systems (DSS) to help users understand the current state and decide on the future state of the object, or it can mean the object making decisions autonomously, by a built in decision making mechanism. In a recent review of the intelligent object across the product lifecycle [2], we found that most research focused on the beginning of life – in the form of holonic manufacturing, and in the middle of life – in the form of objects monitoring their health. The former granted autonomy to objects in driving their production, usually with the use of multi-agent systems [3-5]. The latter coupled products with sensors, and enabled data provision to decision support systems for external decision makers to undertake courses of action [6-12]. Therefore the latter application of the intelligent objects did not yet give true autonomy at the individual product level to decide and act upon its service needs. We introduce the self-serving asset paradigm to grant this autonomy to the intelligent object at the middle of life. The functionality we seek to add to the asset is a survival goal, and the capability to manage its own service lifecycle by balancing the interests of its various stakeholders. For instance, an aircraft component may monitor its health and expiry date, store and transmit its service life history, and schedule or trigger maintenance operations when required. The asset decides on the supplier based on a number of criteria such as its location, cost, previous performance, contractual obligations, and the fit of offered components. It can engage in negotiations with other assets requesting the same services to determine who will get serviced first. The result is an open, largely autonomous service chain, where decisions are traceable and transparent. Such an “active decision-making” capability is in sharp contrast with today's notion of object

CIRP IPS2 Conference 2009

323

intelligence, which enables the object to provide information required to support decisions made by another system (i.e. passive decision support). Thus our self-serving asset must: 1.

be self-aware, in terms of identity, location, health, expiry dates, and operation schedule 2. have the goal to maximise its life in service by autonomously deciding on its service needs, managing the procurement of their replacement parts, taking into account its resources, and perceptions from its environment, 3. act on the interests of its different stakeholders, such as selecting a supplier based on its previous performance, or deciding on service schedules to minimise disruption to operation cycle, 4. have the ability to engage in communication with other assets or intelligent systems when searching and competing for resources. In the remainder of this paper we briefly review service and maintenance in aerospace and extract needs resulting in the self-serving asset vision. We then present a roadmap toward the realisation of the self-serving asset, highlighting market drivers, necessary systems, standards and adoption, and evolution of technology. Finally we summarise risks awaiting the vision, and conclude our findings. 2 SERVICE and MAINTENANCE IN AEROSPACE This section briefly examines the current aerospace service supply chain structure based on data obtained from maintenance activities of three aerospace companies, and identifies key stakeholders that has led to our notion of a self serving asset. The following are the different types of health management activity in aerospace:  Scheduled: This type of maintenance is conducted for

preventative reasons before the failure of a component occurs. The actual condition of the part is unknown before the service. It is done with static time intervals for inspection, repair or overhaul. Schedules may be determined using modelling and simulation.  Reactive: This type of maintenance is reactive to failures and operates in an unscheduled manner. If we have sophisticated integrated diagnostic hardware and

software available, the events that led to the failure can be reconstructed to ease the troubleshooting process, leading to diagnostic service such as Integrated Vehicle Health Management [13]. Real time usage data is necessary to operate in the diagnostic service style, which is obtained by the use of embedded sensors that monitor various usage parameters.  Prognostic: This maintenance style takes the diagnosis process of the fault based maintenance a step further, and schedules maintenance based on forecasts on remaining component life. Actions taken have the goal of maximising remaining service life rather than solely solving the current problem. This is the best style in terms of fitting in with the aircraft’s schedule, and maximising the remaining life of the component. Currently the commercial aviation industry focuses predominantly on preventative, scheduled maintenance but hopes to move towards diagnostic and prognostic health management, categorised under the term: condition-based management (CBM) [14-15]. This will enable parts to be serviced before a failure occurs, maximising their life. It also enables streamlining of service operations, as schedules for prognostic failure can then be drawn more efficiently, as opposed to unscheduled, reactive maintenance. This goal is in line with the self-serving asset vision as assets are self aware and are capable of providing condition based health management, given appropriate data processing capabilities. In CBM, determining the equipment operating status is accomplished in three ways:  Embedded sensors in equipment and monitoring on-

the-fly  Portable devices that connect to an interface to gather

data from embedded sensors; or  Applying an external sensor on the part, using standalone wear instruments or gauges such as tire-wear Using the above technologies, a scheduled inspection is performed by analysing the data collected. Once a course of action is decided (which may involve using decision support systems (DSS)), a service/replacement order is given if required. The inspection process involves manually checking paperwork and entering data on service information systems, which takes an average of 25 minutes per part (averaged across three companies that were studied as part of this work). Procurement officers at the maintenance, repair and overhaul (MRO) facilities select suitable external vendors if the part cannot be repaired or replaced in house. This decision is made mainly based on time limitations, costs and maintenance capabilities. A higher level decision needs to be made when maintenance needs to be outsourced, and would be based on factors such as contracts and relationships with external vendors (e.g. external repair agents, OEMs). If there are no available parts, e.g., if they are out of

production, the issue is referred to a manufacturer. The capabilities of the manufacturer may have been recorded at the MRO or can be obtained by personnel at the MRO. Once the service provider is decided upon and agrees to provide the service, a maintenance schedule is drawn, taking into account the flight-schedule and current/future location of the aircraft, and logistics providers are alerted. Analysis using MEDA, a tool designed by Boeing for investigating factors that contribute to maintenance errors, found that documentation and operator errors are the most frequent contributing factor to maintenance errors, highlighting the need for traceability, autonomy and visibility to meet the future challenge of managing a global service supply network. Another issue is the increasing complexity of the aerospace service supply chain. Suppliers are globally sourced, and competition to provide more streamlined and less costly service operations is increasing. Leasing of the aircraft brings in multiple stakeholders on the parts. The stakeholders include: the airline, owning or leasing the plane, the MRO shops that repair and maintain parts for an airline, and the third party suppliers of parts. Their common goals include the maximisation of aircraft availability and minimisation of time on ground. The result is a requirement for simpler, consistent, cheaper, and cooperative service systems. When considered in the light of the above challenges in current service supply chain operations, the key benefits that the self-serving asset brings include:  CBM is incorporated as asset is aware of its health

and expiry. “Awareness” and “goal-directed behaviour” can be facilitated using intelligent software agents tied to a part through a unique identification such as RFID. Agents perceive their environment through sensors, and process the data. Wireless networks and common communication languages can facilitate the data exchange. The agents execute actions based on this perception, to reach to their encoded goals. Asset selfawareness in real-time leads to the reduction of unplanned maintenance, benefiting stakeholders, i.e. the airline, MROs and the OEMs, of increased life in service.  Through automated decision-making, decisions are

taken faster, leading to a more responsive supply chain and simpler flow of operations. Decision making is traceable and consistent.  Through automation, errors and time of operations are reduced, and a leaner approach to lifecycle management is formed, which in turn increases aircraft availability.

Figure 1: Service operation with the self-serving asset

324

3 THE SELF-SERVING ASSET: A VISION Figure 1 shows a snapshot of the new service operation based on the self-serving asset. Each part is represented by its dedicated software agent, accessible by a unique identifier recorded on an asset identification tag. Complex serviceable and critical assets such as landing gears, engines, fuel systems are monitored by onboard or embedded sensors which collect asset state information such as hydraulic pressure, temperature and flow rates of fluids. Less complex, replaceable assets (such as oxygen generators or life vests) would not need sensors, but have expiry dates embedded in their agent software’s belief sets. Data is collected on the central maintenance computer and transmitted to ground station via wireless communication channels. The servers on ground have the assets’ representative agents that process the data. On the event that a fault is detected or predicted, asset agents evaluate the current circumstances and execute a plan of action. For instance, an asset agent may start searching for suitable suppliers by communicating with supplier systems. Asset agents are able to choose suppliers, award contracts and trigger logistic operations. Asset agents also communicate with one another in the face of competition and resolve any conflicts that arise. Once they issue an order (service request), the monitoring phase starts until an operator completes maintenance and interacts with the software to confirm that maintenance has been carried out successfully. Asset agents then update the configuration and condition of the aircraft, as well as the supplier’s performance in their belief set. In the next section we discuss the roadmap towards achieving this vision. 4 ROADMAP TOWARDS SELF-SERVING ASSETS In this section, we describe the roadmap, starting with the market drivers, and highlighting each item that will help bring us closer to the self-serving asset. Evolution of technology, market drivers, standards and legislation as well the standing point of stakeholders all have an impact on the vision. In order to outline the requirements as well as a technology evolution pathway, we have developed a roadmap for the deployment of the self-serving asset concept on the aerospace service supply chain (Figure 2). The roadmapping methodology have been taken from [16]. Technology roadmapping is a methodology to help companies plan in achieving their technology related visions. First the roadmap is structured with the various factors impacting the self-serving asset vision, including systems, market drivers, standards, adoption, and technology enablers. Next unstructured interviews were carried out with Boeing personnel and the Cambridge research team to develop an understanding of the selfserving asset vision. Then a literature survey is conducted to extract the state of the art and develop how the various aspects considered should evolve over a given time horizon such that the self-serving asset vision is brought to life. The roadmap is then further developed and validated with a total of five industrialists from aerospace service sector, each from different companies. The interviews also helped extract risk drivers awaiting the self-serving asset vision. The roadmap considers the following aspects: 

Market and business drivers: Unlike the military aerospace industry, where performance is the major driver of change, civil aerospace requires drivers of change that emanate from the market place, from government legislation, from compelling economic

325

 





rationale or some combination of these to justify a new product/technology development. We identify external market drivers reflecting the business motivations. They are grouped into short-medium and long term. System: This section shows the set of systems that needs to be developed to bring the self-serving asset to life. Standards: The commercial aviation industry relies heavily on information exchange between trading partners. In line with this reliance, we outline data exchange standards and legislations required before new technology can be introduced. Adoption: Since the roadmap encompasses the entire service supply chain, collaboration is essential for successful deployment of the self-serving asset. This layer captures actions to be undertaken by different stakeholders. Technology enablers: There are various technology requirements to bring the above scenario to life. This section shows the various technology requirements to enable the self-serving asset.

4.1 Market Drivers High prevailing fuel costs: Non-fuel costs have decreased in recent years with reductions in labour, and increases in operation efficiency [17]. Although this is likely to continue, fuel cost as a percentage of operating expenditure has a general upward trend, meaning that more efficiency increase in the aerospace operations are needed to decrease costs and balance the increase of fuel costs. Reduction of no-fault found: Aviation data suggest that there are in excess of 400,000 of no-fault-found cases per year, where a false alarm is given and no fault is found after the investigation. The number represents 23% of all (1.76 million) component removals. With an average cost of $800 per removal, including labour, tracking, testing, no-fault-found removals cost the industry approximately $300 million per year [18]. “The only quotable estimate, provided by Airbus, was published 10 years ago by the Air Transport Association. At that time, Air Transport Association estimated annual no-fault-found costs for an airline operating 200 aircraft at $20 million, or $100,000 per aircraft per year [19]”.This points a strong market driver in automated condition-based maintenance. Restructuring of MRO market: As the civil aviation market expands, competition among airline operators, independent service providers and original equipment manufacturers (OEMs) will intensify for taking a share of the service and maintenance market. [20] indicates that airline maintenance companies currently account for nearly 60% of market share, followed by independent service providers and OEMs and suggest that with the globalisation of third-party MRO providers consolidation will lead to increased buying power for these players and an accompanying reduction in inventory pools. The process of consolidation will continue until a leaner, more efficient MRO operation structure emerges. Leaner service operation environments will be necessary to simplify the service operations. Global sourcing of suppliers: The commercial aviation industry’s globalisation process is in its infancy. Significant investments particularly in India, China and Russia are taking place. As a result, engineering partnerships with Western OEMs have begun to accelerate the development of the capabilities of

emerging markets. In developing new aircraft, the involvement of suppliers from different parts of the world creates complex management, integration and coordination challenges giving rise to new collaborative models. Increasing demand for air travel: The Department for Transport in U.K. forecasts short haul traffic is expected to grow at an average of 4.8% per year over 2005-2020, slightly less than the long haul growth rate of 6.1% over the same period [21]. Demand for aircraft in emerging markets is increasing, with China, India, Russia are expected to purchase more than 3,500 planes (roughly 15% of global demand) over the next two decades [22]. Global penetration of low-cost carriers: A consequence of the passenger delivery growth is strong airline fleet expansion among low-cost carriers [23]. This is a significant development for the maintenance sector since low-cost carriers typically contract their MRO requirements to third-party providers, leading to higher pressures for collaborative, consistent service provision. Switching to fly-by-the-hour: OEMs are actively pursuing a new fly-by-the-hour business model. Rolls Royce predicts that the 40% of civil engines under TotalCare in 2004 will grow by 80% by 2010 [24].The new model has numerous advantages over the traditional model – including lower cost of asset ownership by allowing airline operators to treat an asset as a service. Payment is based on asset usage, with outsourced partners taking on MRO responsibility. This switches focus from capital spending to variable costs, allowing airlines to focus on their core business, expecting fast, reliable service delivery. Example set by the military sector: The military sector is primarily driven by performance, and is the pioneer of condition-based aircraft maintenance. An example is the Joint Strike Fighter Prognostic Health Management Programme (1997-2037) [25-26]. The nature of commercial aerospace market is fundamentally different because the airline customers rarely request dramatic technological innovation; instead, their approach to improving aircraft performance tends to be incremental, with a heavy emphasis on cost [18]. The civil aerospace companies will adopt a condition-based health management system only when the economic benefits are realised in the military sector, and when the technology is proven reliable, mature and safe. Mandates for lower emissions: The European Commission proposed to include carbon dioxide emissions in the European Union Emissions Trading Scheme for commercial flights arriving or departing European Union airports, including US registered aircraft. The proposal aims at capping carbon dioxide emissions at the average emission levels between 2004-2006 [27]. The Advisory Council for Aeronautics Research in Europe have created the “Vision 2020” for European aeronautics which sets out a number of ambitious targets such as 50% reduction of carbon dioxide and 80% reduction of NOx emissions [28]. In order to achieve these targets, aircraft manufacturers and suppliers need to modernise and expand their fleet with more environmentally friendly aircrafts; also in the wider context of life cycle management of aircrafts. The above market drivers show the changing face of the civil aviation industry and point to requirements for a leaner, more collaborative and transparent service supply chain. The self-serving asset emerges as a key bundle of technology in help companies move towards this goal.

4.2 Systems, Standards and Adoption In terms of systems, standards and adoption, incremental improvements to the current systems are needed to be undertaken. A primary factor is achieving a uniform method of uniquely identifying assets with an agent representation on the network. This will ensure data integrity and quality throughout life, and support agent activities when facing multiple suppliers. Solutions can be found in maintaining and extending the existing 1D or 2D barcode identification to all aircraft parts that need to be serviced, a development which all suppliers, OEMs and MROs shall adopt. Another enabling technology that is being adopted by the aerospace industry is Radio Frequency Identification (RFID), for uniquely identifying objects using microelectronic transponders that communicate wirelessly with dedicated systems. Unique and automated identification is a key requirement for the self-serving asset, making developments in RFID technology critical for its realisation. In 2005, Federal Aviation Administration (FAA) approved the use of passive Radio Frequency IDentification (RFID) tags on ground in its Spec2000 specification1. This calls for the industry to establish a standard for compatible interfaces between proprietary systems. RFID tags can be integrated with existing barcodes and are likely to coexist with RFID for several years. As tag costs become lower their use is expected to increase. Different assets require different tag data storage capacity, to support service operations as well adhering to regulatory mandates. For example, a life vest would only require a low memory passive tag with a unique ID whereas a hydraulic pump or actuator could require high-memory tags for storing sensor data as well as maintenance data and service history [29]. This observation points a requirement for a system that facilitates the use of different types of tags together. There are different regulations regarding the use of RFID across various regions of the world. This becomes an issue as the MRO markets globalises. For instance, RFID uses licence free parts of the electromagnetic spectrum and these allocations are different depending on the regulatory bodies in that region. For example, the widely deployed UHF RFID Systems must use the 840.25 to 844.75 MHz and 920.25 to 924.75 MHz ranges approved by China's State Radio Regulation Committee (SRRC) in that country but RFID UHF bandwidth across the European Union ranges from 865 MHz to 868 MHz while RFID UHF bandwidth in North America ranges from 902 to 928 MHz. These differences add to the complexity and the cost of the ultimate system. Systems that manage these regulatory differences will be necessary. A major concern for the use of RFID is the security and integrity of data. Commonly used cryptographic mechanisms rely on large complex computations and computing resources that are generally available in embedded systems. However these mechanisms are not suitable for low power, low cost resource constraint devices such as passive RFID tags [30]. There are many ongoing research projects that are addressing both security and privacy concerns posed by RFID technology, a review of which can be found in [31]. Standards regarding data exchange, encoding of service related data on tags, are to be developed to facilitate data exchange in the aerospace service supply chain,

1

326

http://www.spec2000.com/10.html

Decision making models and conflict resolution models used by the self-serving asset need to be accepted across the supply chain, as well as the concept of automated decision making. Regarding data transfer, current info bus systems provide the ability to transmit flight information, including sensor data, flight controls and safety systems; as well as data from non-essential systems to the central maintenance computers where the information is processed. An extension to improve the central info bus technology can be achieved by segmenting the info bus into layers and giving safety-critical system data the priority for transmission. Based on the criticality of data, different communication networks can be used. For example, realtime monitoring data are processed directly by onboard health management systems. If a fault is detected and maintenance is required; a signal is sent to notify the ground stations. Other non-critical data are recorded in a no volatile memory and later be downloaded to a central repository for analysis off board. A possible improvement to this system is direct wireless download to a central repository. Due to complexity of analysis, currently much of the prognostic reasoning is performed off- board. In the future, more onboard reasoning could be achieved if more server space and computational power is available. By shifting the reasoning capability onboard increases efficiency of response to maintenance and service needs by mitigating delays associated with ground based DSS that require data to be sent through manual or wireless channels. In addition, data transmission services to ground base are only available when an aeroplane is at a certain distance from airport facilities therefore it is critical to use the available time to organise and schedule maintenance and service activity to minimise disruptions to airlines and flight schedules. A key aspect of the self-serving asset is autonomous decision making. Decisions to be made include the type of maintenance needed as well as timing, tools, parts and selection of suitable suppliers. Today decision support systems (DSS) are common place, which inform decision makers about trade-offs and available options based on inputs of current circumstances such as resources availability, flight schedules and cost of replacement parts from various supply chain options. However DSS usually ask for an external decision maker to make the decision. A step forward from DSS is distributed and autonomous decision making, comprising a smart system based on emergent behaviour such as agent based systems. These will require real-time update of information on supplier availability and pricing, a conflict resolution model and a scalable, high speed computational architecture. Lowlevel decision making is increasingly automated enabling companies to manage their global supply network more efficiently. Finally, all the deployed systems across multiple parties must be integrated to achieve a truly distributed system. This requires standards for data exchange and corporate trust for open information exchange between collaborators. Partners can then securely share timely information and give each contributor appropriate access to the information over the internet using a common data exchange standard. For instance, two existing standards, the EPCglobal EPCIS interface [32] and STEP standards for product data exchange [33] can be considered to provide appropriate metadata models for data exchange.

327

4.3 Technology Enablers The self serving asset concept is primarily based on leveraging developments in various technology fields such developments in computer processors, energy harvesting, Micro-Electro-Mechanical Systems (MEMS) technology and microelectronics, High-speed computation architectures are required for supporting increased processing requirements of systems hosting agent based software. Current developments in multi-core processors as well the developments in 64 bit processor architectures capable of supporting increased capacities of Random Access Memory for more in memory operations and processing of data are critical for the deployment of agent based systems in large scale. Continued miniaturisation of sensors, through development in MEMS leading to nano sensors as well as developments in wireless sensor networks supported by energy sources developed through energy harvesting will allow sensors to be embedded more extensively to increase the state space monitored by current systems while satisfying a key requirement in reducing the extra weight added onto the aircraft to bring the system to life. Advances in RFID technology to develop more reliable, cost effective, high memory and secure tags will be able to support the requirements of automating manual records of service and maintenance activity as well as support the link to asset agents executing on networked systems. Eventually large scale integration and miniaturisation of microelectronics will pave the way for ultimately running lightweight agents onboard assets capable of forming an intelligent entity capable of managing its operational life in its totality. Previous research had found that multi-agent systems (MAS) are particularly suitable to model a supply chain where each party has its partial own view of the environment, its own goals and behaviour results in an emergent system [34]. Autonomous decision making capability is also natural to MAS. However the use of MAS to represent each serviceable part and provide autonomous decision making capability will depend on improvements in the scalability of agent software as this has been an issue in past industrial applications [35]. Developments in communication technology such as ZigBee [36], which is a low-cost, low-power, wireless mesh networking standard, are required to provide short range communication between intelligent assets in a reliable and secure manner. Developments in new communication technology will also play a role in advancing self-organizing ad-hoc digital radio networks, which will allow intelligent assets to organise into intelligent entities and support inter agent communication at a time when the agents are onboard assets as opposed to being on a ground based facility.

Figure 2: Roadmap to self-serving assets 5 BENEFITS and RISKS SERVING ASSET VISION

AWAITING

the

SELF-

A preliminary risk assessment for migrating from today’s practice to the future self-servicing asset model is performed using data from three aerospace companies. A total of five respondents, all of whom have held positions dealing with the aerospace service operations were interviewed using semi-structured interviews. Although a larger sample size is needed, we present a preliminary risk assessment from this basic exercise. When presented with this roadmap and the self-serving asset vision, an even spread of risks in technology, industry, social, legislative and economical areas emerge (Figure 3).

In terms of technology, unexpected breakdowns in the IT system were of most concern, given the reliance on autonomous software. The maturity level of sensor and diagnosis technology and complications in global networking were other concerns. Acceptance of automated decision making, security breaches, and transparency in the supply chain were found to be the risk factors in social and cultural terms. The industry’s potential reluctance to move to condition-based maintenance in the fear of losing steady revenue, lack of collaboration are among the risk factors in industrial terms. Costs of implementation, the difficulty of the business case, and management scepticism on the technology might be the economical factors impacting

Figure 3: Self-serving asset risk drivers

328

deployment. Finally, approval of the condition-based maintenance processes, and technology as well as approval of the use of RFID tags in aircraft (which provide the important unique identifier to the assets) were identified as legislative risk factors. The risk factors highlighted above are critical in deploying the self-serving asset vision. Further investigations to quantify the risks and linking them to different stakeholders on the service supply chain is needed, followed by a comprehensive risk mitigation strategy. 6 CONCLUSIONS Current service and maintenance in civil aerospace is facing major market changes. Increasing fuel costs, demand for air travel, and global outsourcing, are some factors that require lowering the cost of service provision and increasing the quality of service, to enable suppliers and manufacturers to remain competitive. The sector is under increasing pressure to deliver better service with fly-by-hour contracts, and mandates to better manage parts lifecycle. The self-serving asset comes about as the result of such market drivers: a technology that aims give autonomy to aircraft parts in monitoring their health, deciding when to order service, and where to order it from. The self-serving asset has the goal to maximise its life, and serve to its multiple stakeholders. The vision comes with a set of requirements for technologies, systems, standards and adoption to facilitate its realisation. Standard regulations regarding data exchange, tag encoding, secure, high speed wireless communications, integrated systems for different tag types and frequencies, integrated sensor networks, development of scalable agent architectures, form some of these requirements. A preliminary risk assessment shows management scepticism, high costs, lack of regulatory approval, and system breakdowns being among the highest risk factors. With this short assessment we aim to provide researchers and practitioners a common view on this new vision emerging from needs of the current service sector, and draw attention to the necessary developments that are highlighted on the roadmap. 7 REFERENCES [1] Wong C.Y., McFarlane D., Zaharudin A., Agarwal V. (2002), The intelligent product driven supply chain, Systems, Man and Cybernetics, IEEE International Conference, 4: 6-9, October 2002 [2] Brintrup A., Ranasinghe D., McFarlane D., Parlikad A., (2008) A review of the intelligent product across the product lifecycle, International Conference on Product Lifecycle Management, Seoul, Korea, July 2008. [3] Chirn J. L. and McFarlane D. C. (2000a), A component-based approach to the holonic control of a robot assembly cell. Proc. IEEE Int. Conf. on Robotics and Automation, 4:3701-3706. [4] Chirn J. L. and McFarlane D. C. (2000b) A holonic component-based approach to reconfigurable manufacturing control architecture. Proceedings of the 11th Int. Workshop on Database and Expert Systems Applications, pp. 219-223. [5] Bussmann, S. and Sieverding, J. (2001) Holonic control of an engine assembly plant-an industrial evaluation. Proc. IEEE Systems, Man, and Cybernetics Conference, Tucson, Arizona, USA, pp. 169 - 174.

329

[6] (PROMISE 2004) "Product Lifecycle Management and Information Tracking using Smart Embedded Systems" home page: http://www.promise.no [7] (DYNAMITE 2006) "Dynamic Decisions in Maintenance (DYNAMITE)" home page: http://osiris.sunderland.ac.uk/%7Ecs0aad/DYNAMIT E/Index.htm, accessed on 02/2008. [8] Zhang W, Halang W, Diedrich C. (2003) An agentbased platform for service integration in Emaintenance. In: Proceedings of ICIT 2003, IEEE international conference on industrial technology, vol. 1, Maribor, Slovenia, pp. 426-33. [9] Yu R, Iung B, Panetto H. (2003), A multi-agents based E-maintenance system with case-based reasoning decision support. Eng Appl Art Intell 16:321-33. [10] Djurdjanovic D, Lee J, Ni J. (2003), Watchdog agent, an infotronics-based prognostics approach for product performance degradation assessment and prediction. Adv Eng Inf 17 (3): 109-25. [11] Hung M, Chen K, Ho R, Cheng F. (2003) Development of an e-diagnostics/ maintenance framework for semiconductor factories with security considerations. Adv Eng Inf 17 (3-4):165-78. [12] Bangemann T, Reboul D, Scymanski J, Thomesse JP, Zerhouni N. (2006), PROTEUS-An integration platform for distributed maintenance systems. Comput Ind , 57(6):539-51. [13] Keller K., Baldwin A., Ofsthun S., Swearingen K., Vian J., Wilmering T., Williams Z. (2007), Health management engineering environment and open integration platform, 2007 IEEE Aerospace Conference. [14] Deaton J. E. and F. A. Glenn. The Development of Specifications for an Automated Monitoring System Interface associated with Aircraft Condition, Int. Journal of Aviation Psychology 9(2): 175-187. [15] Schaefer, C.G. and D. J. Haas (2002), A Simulation Model to Investigate the Impact of Health and Usage Monitoring Systems (HUMS) on Helicopter Operation and Maintenance. American Helicopter Society 58th Annual Forum, Montreal, Canada, 58(2):1653-1665, June 11-13. [16] Phaal R., Farrukh C.J.P., Probert D.R. (2004), Technology roadmapping: A planning framework for evolution and revolution, Technological Forecasting and Social Change, 71(1-2):5-26, Roadmapping: From Sustainable to Disruptive Technologies. [17] IATA Industry Financial Forecast (2008), available from: http://www.iata.org/economics, March 2008. [18] Lorell M. A., Lowell J. F., Kennedy M., Levaux H. P. (2000), Cheaper, Faster, Better? Commercial Approaches to Weapons Acquisition: Lessons from the Commercial Aerospace Market, 89-136. [19] Burchell B. (2007), Untangling No Fault Found, Overhaul & Maintenance, Aviation Week , February 2007. [20] Society of British Aerospace Companies (SBAC) and A.T. Kearney (2008), Airline MRO Providers Under Pressure, available from: www.defenseaerospace.com/produit/32690_us.html, January 2008. [21] DoT (2000), Department of Transport, Air traffic forecasts for the United Kingdom 2000, available from: http://www.dft.gov.uk/pgr/aviation/atf [22] Bedier C., Vancauwenberghe M., Sinter W. (2008), The Growing Role of Emerging Markets in Aerospace, McKinsey Quarterly, April 2008. [23] SR Technics (2008), Partnerships for Success: A Company Portrait, available from:

[24]

[25]

[26] [27] [28]

[29] [30]

[31]

[32]

[33]

[34]

[35] [36]

http://www.srtechnics.com/datas/docs/sr_technics_co rp_brochure.pdf Terret M. (2005), Rolls Royce TotalCare: the dependable way of life, available from: http://www.rolls-royce.com/media/presentations/ services3 Smith G; Schroeder J. B., Navarro S., Haldeman D. (1997), Development of a prognostics and health management capability for the Joint Strike Fighter, AUTOTESTCON '97; Proceedings of the System Readiness Technology Conference, Anaheim, US, 676-682. Hess A., Calvello G., Dabney T., (2004), PHM a key enabler for the JSF autonomic logistics support concept, IEEE Aerospace Conference 2004. Lynch K. (2008), EPA to invite comment on aircraft emissions, Aviation Week, April 2008. Arguelles P., Bischoff M., Busquin P., Droste B., Evans R., Kroll W., Lagardere J., Lina A., Lumsden J., Ranque D., Rasmussen S., Reutlinger P., Robins R., Terho H., Wittlow A. (2001), European Aeronautics: a Vision for 2020, Meeting Society's needs and winning global leadership, available from: http://www.acare4europe.org/docs/Vision%202020.p df Harrison M. (2007), Guidelines for Lifecycle ID and Data Management, AeroID whitepaper, available from: http://www.aero-id.org Ranasinghe, D. C, Goshal, R., Grasso, A., and Cole, P.H. (2008), "Lightweight Cryptography for Low Cost RFID: A new direction in cryptography", Handbook: Applications, Technology, Security and Privacy, CRC Press, 2008. Cole, P.H., and Ranasinghe, D.C, ed., Networked RFID Systems and Lightweight Cryptography: Raising Barriers to Product Counterfeiting, SpringerVerlag, Nov. 2007. EPCIS (2007), EPC Information Services (EPCIS) Version 1.0.1 Specification, available from: http://www.epcglobalinc.org/standards/epcis/epcis_1_ 0_1-standard-20070921.pdf Pratt M. (2005), ISO 10303, the STEP standard for product data exchange, and its PLM capabilities International Journal of Product Lifecycle Management 2005, 1(1):86 - 94 Frey D., Woelk P. O., Stockheim T., Zimmermann R., Integrated multi-agent-based supply chain management, Enabling Technologies: Infrastructure for Collaborative Enterprises, 12th IEEE International Workshops on WET ICE 2003, 24-29, 9-11 June 2003 Deters R. (2001) Scalability and information agents. SIGAPP Appl. Comput. Rev. 9(3): 13-20, Sept. 2001. ZigBee Specification (2008), available from: http://www.zigbee.org/en/ spec_download/ download_request.asp.

330

The Chinese Service Industry as a Challenge for European SME: A Systematic Approach for Market Entry R. Schmitt, S. Schumacher, C. Scharrenberg Department of Production Quality and Metrology, Fraunhofer Institute for Production Technology IPT, Steinbachstr. 17, 52074 Aachen, Germany {robert.schmitt, sven.schumacher, carsten.scharrenberg}@ipt.fraunhofer.de

Abstract Compared to industrial firms, the market share of foreign companies in the Chinese service industry is rather low. Especially western small and medium-sized service providing enterprises face different problems when they try to establish a business in China. Therefore an approach has been developed within a research project, which aims to support these companies in planning and successfully realising a service market entry in China. Methods are presented, which allow a systematic benchmark and selection of the target markets and a detailed definition and agreement of the cooperation between the Chinese and the western company. Keywords: Internationalisation, Service, China

1 INTRODUCTION In the recent past, the service sector has grown rapidly. In many industrial countries, the service sector even accounts for a higher percentage in economic development than the traditional manufacturing sector. Like manufacturing companies before them, service providing firms are also experiencing the process of internationalisation by following their clients into a foreign market or seeking a new market in a foreign country. With several reforms and the opening-up policy, China has utilised a large amount of foreign investments. However, the multinational corporations in China were mainly manufacturing companies. Until recently, China had failed to recognise the important role service can play in economic development. As a result, the service industry has remained relatively underdeveloped compared to other fields of industry. The manufacturing sector will remain the main driver of economic growth in China for some time to come. However, the services sector is also now being touted as the country’s next main engine of growth, as services will provide a solution to China’s serious employment challenges, sustain economic growth and raise living standards. China’s accession to the WTO in 2001, with its commitment to liberalise services, is expected to herald a new era in the country’s services trade [1]. The huge potential of the Chinese service market attracts many multinational companies including German service providing firms. Although German manufacturing companies have obtained much experience in entering the Chinese market, this remains a relatively new topic for industrial service providers. Therefore, a procedure model that allows, especially small and medium-sized service providing enterprises, a successful entry into the Chinese service market is developed within a joint research project of research institutes and industrial partners. This paper focuses on two methods, which are essential elements of this procedure model.

CIRP IPS2 Conference 2009

331

2

THE CHINESE SERVICE MARKET: POTENTIALS AND RISKS Internationalisation is traditionally viewed as a process through which a company moves from operation solely in its domestic market place to international markets [2]. Though internationalisation was typical for manufacturing companies in the past, the service sector has gained a great share of international trade and foreign investments. One reason for this ongoing development is the tendency towards combined offers of products and services in industrial product-service systems (IPS²) to fulfil customer needs. Following motivations for internationalisation of integrated solution providers have been suggested [3] [4] [5]: 1. Client-following: With the increasing globalisation of economic activity, service providing companies come under growing pressure to follow their multinational clients. Thus, client-following is a major demanddriven motivation for internationalisation. 2. Market-seeking: Based on supply-driven motivations, increasingly more service providing companies proactively seek new international markets. As a market of sufficient size may only be claimed through internationalisation, industrial service providers seek for a wider client base. Furthermore, multilateral agreements such as the General Agreement on Trade in Services (GATS) and the EU Services Directive press towards internationalisation of services. As the survey “Service within the mechanical engineering industry”, conducted by the Fraunhofer Institute for Production Technology, indicates, the Asian market offers a high growth potential for the service industry. Within the Asian market, 70% of the respondents estimate China to be the market with the highest potential for growth of the service sector (Figure 1) [6]. This economic valuation is sustained by regarding the turnover distribution of the respondents. Asia is the second largest sales market of the questioned companies, but they generate only 10% of

70%

India Japan

16% 6% Proportion of mentions of the questioned companies

Others

8%

Figure 1: Potential for growth of service in Asia. Additionally, the contribution of services to China’s overall output is substantially lower than in many economies of comparative levels of income. In the context of world development experience, the services sector in China should account for a much greater proportion of the country’s total output than it currently does [1]. Over the period from 1991 to 2002, the portion of the service sector increased for 1% to 34% of China’s gross domestic product (GDP). In 2007 the proportion was around 39% [8]. Compared to Germany, with a portion of the service sector of nearly 70%, the portion of the service sector to China’s GDP is still low [9]. China has acknowledged the importance of services. In its 11th Five-Year Plan (2006–2010), the Government announced further opening up and promotion of the development of the service sector in order to substantially expand its presence in the national economy. In conclusion, China’s service sector is seen as the next target for many multinational corporations as the opening of the sector offers enormous market opportunities [1] [7]. However, the study shows that internationalisation is a big challenge for German industrial companies, which want to offer integrated product-service systems in foreign markets. Up to 40% of the respondents do not know the requirements of their foreign clients and, therefore, up to 66% of the asked companies confirm that they do not offer tailored service products for their foreign target markets. As a result, up to 50% fail to fulfil the requirements of their foreign customers [6].

Professionalisation

48% 64%

72%

78%

15% 12% 37%

1000

Years of presence in China

31% 46%

48%

6%

73% 14%

16%

63% 40%

10%

36%

17%

57%

Expand capacity

10 Positive

Figure 3: Profitability of service companies in China. Regardless of the motivation for internationalisation, small and medium-sized enterprises seem to need a procedure model, which allows them to enter the Chinese service market successfully and to act profitable from the start. In scientific literature, only research concerning companyinternal success factors for internationalisation [11] or the effects of globalisation on small and medium-sized enterprises [12] can be found. Therefore an approach has been developed within a joint research project of research institutes and industrial partners, which aims to support western integrated solution suppliers as well as pure industrial service providers in planning and successfully realising a service market entry in China. The main focus of the approach is on the initial project phase, trying to avoid early failures and to ensure an efficient decision making at the project start. A second important aspect is

332

the arrangement of a cooperation with a local Chinese partner. Methods are presented which allow a systematic benchmark and selection of the target markets and a detailed definition and agreement of the cooperation between the Chinese and the western company. 3 METHODS The procedure model, which is developed within the research project is structured in three phases. initial phase

planning phase

implementation phase

Figure 3: Three-phase procedure model. Elements of the first phase are initial activities for preparing a possible China engagement and provide a basis for subsequent activities. For example, the analysis of target markets, an evaluation of market chances and risks, as well as the contacting of potential clients and business partners belong to this phase. The objective of the phase is to establish a solid information basis for decision making with regard to a China engagement. As an application-oriented tool for conducting a comprehensive market analysis, the scoring model, which is described in the next chapter, can be applied. The detailed planning of the internationalisation process is accomplished within the second phase of the procedure model. For example, concepts for marketing, the offered services, the quality management and the organisational structure are developed. For establishing and ensuring a basis for a successful cooperation, the method ServiceBlueprint was adapted within the research project and is described in a later chapter. With regard to the intercultural cooperation and the establishing of a organisational culture, the objective of the third phase is to ensure a successful start of the new service business in China. The new insights gained in this phase are used for continuous improvements of the concepts in the second phase. 3.1 Scoring model According to Applebaum, scoring models were the first approaches for a systematic evaluation of the worth of a company’s location in comparison to other potential locations within a market area [13]. As scoring models have been used as a decision guidance for the choice of a location when opening a new subsidiary within a geographical region, this method has been adapted within the developed procedure model for the evaluation of potential target markets or rather target countries. PRODUCT 1 weight criteria

CHINA

USA

rating product rating product

20

customer structure

3

60

4

80

10

market growth

5

50

6

60

15

market volume

4

60

4

60

30

entry barriers

2

60

1

30

























10

Infrastructure

3

30

2

20

∑100

∑ 460

∑ 375

Figure 4: Example for scoring model evaluation. Thus, the scoring model, as a classical tool for strategy planning, provides a basis for a methodical evaluation of options for actions within the process of

333

internationalisation. The objective of this method is to reach a better understanding of the interactions and players within different target markets. By identifying their own as well as the market potentials a company’s strategic course can be determined. Therefore, economic, technical, psychological and social influencing factors are considered. These criteria are based upon quantitative and qualitative characteristics. For a successful utilisation of the method, following steps have to be conducted accurately in a workshop: 1. Choosing market alternatives: At the beginning, the different markets, which come into consideration, have to be chosen. 2. Choosing products: It has to be defined, which service products should be sold in the target markets. Evaluations are conducted and documented in separate tables for each product. These tables embody separate evaluation columns for the specific target markets. Thus, a product-based evaluation of the target markets is conducted. 3. Identifying the influencing factors: Vitally important for ensuring the success of the evaluation is the choice of the influencing factors, which should be chosen depending on the product. A fixed number of criteria is not prescribed. Within the scope of the research project a catalogue of influencing factors has been developed, which should provide assistance with finding the decisive influencing factors. 4. Weighting of influencing factors: Weighting factors, which indicate the relevance of each criterion with regard to the market attractiveness have to be defined. The sum of all weighting factors must be 100. The rating of the values is not trivial and requires a careful consideration. 5. Evaluation: Every single influencing factor for each product is now evaluated for the different target markets. The scale used for these evaluation factors can, i.e. reach from 1 (very bad) to 6 (very good), but also other increments are possible. The crucial point for a purposeful evaluation is a detailed discussion between the workshop participants about each evaluation rating. It is important to keep in mind that the highest rating for a criterion should not be given for the strongest characteristic value of a considered market, but for the optimal characteristic value. Therefore, it makes sense to define the optimal characteristic value of each influencing factor in the run-up to the evaluation. 6. Documentation of reasons: The reasons for specific evaluation ratings should be documented separately. Thus, the motivations become comprehensible for later reflections. Furthermore, it encourages discussions about the evaluation ratings and, with retrospective effect, about the influencing factors. Pros and cons should be documented for each criterion and target market. Additionally, the reasons for evaluating a market more positively compared to another should be recorded. 7. Analysis and comparison: After finishing the evaluation, the weighting factor of each influencing factor is multiplied with the corresponding evaluation value of each target market. After summing up the obtained values for each market, a conclusion about the market attractiveness can be reached by a comparison of the aggregated values. 8. Deriving further decisions: The utilisation of the scoring model facilitates a structured discussion about the market attractiveness and leads to a concerted understanding between the workshop participants.

When mapping out a market entry strategy, the identified strengths and weaknesses of each market should be taken into consideration. When using the scoring model, the results of the market attractiveness for the different target markets can differ with respect to each product. In the end, the company’s management has to decide about the target markets for an internationalisation process by balancing the reasons and calculating the involved risks. Besides being an assistance for the decision process when choosing the target market for an internationalisation of service products, the scoring model also reveals, which competencies are to be strengthened or built-up for a successful start of the service business in the according target market. 3.2 Service-Blueprint The Service-Blueprint facilitates a transparent visualisation of services with all detailed sub-processes and their classification into the two dimensions of customer proximity and time. This allows a systematic analysis of every single process step. In standard Service-Blueprints the classification of process steps does not depend on the organisational unit that carries out a process step within the service company. Furthermore, division of work between strategic partners is not supported [14]. Prior to establishing a new subsidiary in a foreign target market, most companies cooperate with strategic partners in different forms. Within any form of cooperation it is important for the success of a service business to define which partner conducts which process steps to ensure an efficient and smooth workflow. Therefore, the standard Service-Blueprint was adapted within the research project to meet the demands of cooperation (Figure 5). It allows a clear and structured splitting of the process of service provisioning according to the cooperation partners. Thus, internal and external business interfaces can be welldefined and misunderstandings and delays due to ambiguous allocation of roles and responsibilities can be avoided by using the adapted Service-Blueprint. customer activities line of interaction

onstage activities

recognised by customers

line of internal interaction

support activities line of external interaction

external activities

secondary activities

backstage activities hidden activities

supplier activities

line of visibility

activities strategic partner

Figure 5: Adapted Service-Blueprint. Just like in standard Service-Blueprints, the top layer of the adapted Service-Blueprint, with regard to the customer proximity, comprises the customer activities. These are the process steps within a service provisioning that are conducted by the customers themselves [15]. The layers below are used for the process steps, which are conducted by the service providing company or a strategic partner. The less the customer proximity of a process step, the lower the layer the process step is assigned to.

Below the layer of customer activities is the layer of onstage activities. All activities of the service providing company, which can be recognised by the customer, belong to this layer. To provide the customer with a single contact person, it is preferred that only the process steps, which are conducted by the service providing company should belong to this layer and not those of the strategic partner. The layer of backstage activities is situated underneath the layer of onstage activities. As it can be gathered from the name, this layer comprises all process steps with a direct reference to the customer, but which can not be recognised by the customer himself. The dividing rule between the layers of onstage activities and backstage activities is the “line of visibility”, because the activities below this line can not be experienced by the customer. By relocating the “line of visibility” to lower layers, the customer gets a deeper and more transparent insight into the service process. This can lead to an increased willingness to pay for the provided service and can therefore be vitally important for the success of the service business in China, because up to now the willingness to pay for services is rather low in Asia. However, it is important to deliberate about the activities which benefit the willingness to pay and should therefore be revealed to the customers and which do not do so and should therefore be remain hidden for the customers. The layer of backstage activities is followed by the layer of support activities. The process steps which put customer information into customised service solutions belong to this layer. The separating line between the layers of backstage activities and support activities is called “line of internal interaction”, because normally two different internal departments of a service provider interact at this line. Up to the layer of support activities, the structure of the adapted Service-Blueprint corresponds to that of the standard version. The layers for preparation and facility activities, which lie below the layer of support activities in the standard Service-Blueprint, comprise the process steps which have no direct reference to the customer’s order. Since the focus of the adapted Service-Blueprint is on the customer’s order processing in cooperation with a strategic partner, both layers are omitted and substituted by the layer for the activities of the strategic partner. The dividing line between the layer of support activities and the layer of activities of the strategic partner is called “line of external interaction”, because it presents the interface between the service providing company and a strategic partner. Which activities are conducted by the strategic partner depends on the offered service and the form and intensity of the cooperation. Thus, these activities can be of the same value as the activities of the service providing company or provide only support for the process steps conducted by the service provider. Due to these different definitions of cooperation, it is also possible that the strategic partner has direct contact to the customer within specific activities, although its activities are assigned to the layer with the highest customer distance. The decision, which process steps should or can be conducted by the strategic partner, can be supported by evaluating them on the basis of benefit as well as the general and the strategic importance. The benefit or the general importance of a process step is high, if the process step leads to an adding value or influences the service quality significantly. The strategic importance of a process step depends on the level of unique selling propositions. In addition, before outsourcing a process step it should be considered, that cooperation also hold disadvantages. It is necessary to take the risk of an

334

unintentional outflow of know-how or an opportunistic behaviour of the strategic partner. Therefore, it is recommend to combine the process steps, which should be conducted by the strategic partner, to cooperation packages. In doing so, the following criteria for a cooperation package must be considered: 1. It has to be interesting for the strategic partner and must provide enough substance, as winning a partner for a cooperation package is only possible if conducting the package leads to generating profit. 2. The package should feature a clearly defined Input and Output to prevent additional costs due to increased coordination efforts. 3. The interfaces should be defined precisely and clearly, to facilitate the depiction of the contents and the demanded results of the package to the partner. As in the standard Service-Blueprint, the adapted version provides no further classification of process steps within the dimension of time. All activities are documented in their logical and chronological order from the left to the right. In conclusion, the adapted Service-Blueprint is not only a tool to picture the service process. It enables a systematic analysis and design of all activities within a service process. It is a tool, which describes the contact points between the customer and the service provider, the so called “Moments of Truth”, on the one hand and the contact points between the service providing company and a strategic partner on the other hand. 4 SUMMARY This paper describes the present situation of German service providing companies and their internationalisation efforts with special focus on the target market China. The attractiveness of the Chinese service market, due to its immense growth potential, is counter-balanced by several problems and risks especially small and medium-sized service providers have to face when trying to establish a business in China. Subsequently, two selected methods of a procedure model, which were developed within a joint research project of research institutes and industrial partners, are presented. This procedure model aims to support both western integrated product-service suppliers as well as pure industrial service providers in planning and realizing a successful service market entry in China. The scoring model, primarily used as a decision guidance for the choice of a location when opening a new subsidiary within a geographical region, has been adapted within the developed procedure model for the evaluation of potential target markets or rather target countries. It also reveals which competencies are to be strengthened or built-up for a successful start of the service business in the according target market. Usually, the first step within the internationalisation process is to build up a cooperation with a strategic partner in the target market. Decisive for the success of a service business is to define which partner conducts which process steps to ensure an efficient and smooth workflow. Since the division of work between strategic partners is not supported in a standard Service-Blueprint, this method was adapted within the research project to meet the demands of cooperation. The adapted methods have been successfully verified by the industrial partners of the research project, but continuous improvement and detailing is still subject to further research.

335

5 ACKNOWLEDGMENTS The support of the German Federal Ministry of Education and Research (BMBF) as well as the Project Management Agency – part of the German Aerospace Center (PT-DLR) is greatly acknowledged. 6 REFERENCES [1] Australian Government, Department of Foreign Affairs and Trade, 2005, Unlocking China’s Services Sector, Adcorp Canberra [2] Javalgi, R. G., Griffith, D. A., White, D. S., 2003, An Empirical Examination of Factors Influencing the Internationalization of Service Firms, Journal of Services Marketing, Vol. 17, No. 2, pp.185-201 [3] Groenroos, C., 1999, Internationalization strategies for services, Journal of Services Marketing, Vol. 13, No. 4/5, pp.290-297 [4] Netland, T. H., Alfnes, E., 2007, Internationalisation of professional services – A 1999-2005 literature review, College of Service Operations, 2007 Conference [5] Roberts, J., 1999, The Internationalization of Business Service Firms: A Stages Approach, The Service Industries Journal, Vol. 19, No. 4, pp.68-88 [6] Fraunhofer Institute for Production Technology, 2006, Survey: Service within the mechanical engineering industry, Aachen [7] Chadee, D. D., 2002, Foreign ownership structure of service equity joint ventures in China, International Journal of Service Industry Management, Vol.13, No. 2, pp.181-201 [8] German Embassy in China: Economic facts compact, Beijing, 2008, p.1 [9] German Government; Federal Statistical Office: Statistical Yearbook 2007 for the Federal Republic of Germany, Wiesbaden, 2007, p. 637 [10] Wuttke, J., 2007, European Union Chamber of Commerce in China, Roland Berger Strategy Consultants: European Chamber Business Confidence Survey 2007 [11] Brouthers, L. E., Nakos, G., 2005, The Role of Systematic International Market Selection on Small Firms’ Export Performance, Journal of Small Business Management 43(4), pp.363-381. [12] Etemad, H., Wright, R. W., 1999, Internationalization of SMEs: Management Responses to a Changing Environment, Journal of International Marketing Vol.7, No.4, 4-10 [13] Applebaum, W., 1966, Methods for Determining Store Trade Areas, Market Penetration and Potential Sales, Journal of Marketing Research 3/1966, pp.127-141 [14] Lorenzi, P., 2004, Service Scout – Dienstleistungsbedarfe antizipativ erkennen und in Netzwerken systematisch erfüllen, Cuvillier, Göttingen, pp.192204 [15] Kleinaltenkamp, M., 1999, Service Blueprinting – Nicht ohne einen Kunden, Technischer Vertrieb, pp.33-39

Industrial Services Reference Model M. Gerosa, M. Taisch Politecnico di Milano, Department of Management, Economics and Industrial Engineering P.zza Leonardo da Vinci 32, 20133 Milano, Italy, {marco.gerosa, marco.taisch}@polimi.it Abstract The need to integrate service providers into an existing customer supply chain requires the collective know-how of the coordination mode, including the ability to synchronize interdependent processes, to integrate information systems and to cope with distributed learning. About this topic the EU-funded InCoCo-S project is developing a new standard business reference model with key focus on operation & integration of business related services in supply chains. Based on the requirement analysis concrete business processes have been developed to integrate services in the existing customer supply chain both on a strategic and operational level. Keywords Service supply chain, process modeling, reference model

1 INTRODUCTION Nowadays, the service businesses industries have developed into an important economic force and have become an integral part of modern society [1,2]. More than eighty-five percent of all North American and European companies have outsourced at least one function [3]; sixty percent of Fortune 500 companies surveyed have at least one logistic outsourcing contract [4]. Especially over the last ten years, organizations have increasingly improved their own service operations. Many service sectors have sought and made use of various enhancement programs to improve their operations in an attempt to be highly competitive. [5]. Industrial service is becoming increasingly important to manufacturing firms for a number of reasons. To improve profitability it is not enough to sell just a product; the real impact on profitability comes from exploiting downstream opportunities, by providing the customers with products such as financing, maintenance, spare parts and consumables [6,7]. Moreover about seventy percent of the overall European GDP is generated by the service industries in total. Over fifty-four percent of the overall European GDP is generated by Business Related Services and thirty percent of all Business Related Services are consumed by the production sector [8]. This highlights the increasing relevant importance that supply chains driven by the exchange of services among service providers and customers (service supply chains) have in Europe. In such context, the need for improving integration of service providers into an existing supply chain requires a distributed know-how of the coordination modes, including the ability of synchronizing interdependent processes, to integrate information systems and to cope with distributed learning. In this paper we deal with the design and benefits of a new Reference Model for Industrial Services resulting from the framework of the EU-funded InCoCo-S project (Innovation, Coordination and Collaboration in Service Driven Manufacturing Supply Chains, see website in the references). Moreover, the so called Industrial Services Reference Model answers to the need, for new and innovative business processes requested by the efficient integration of service provider networks. On the basis of requirement analysis, the InCoCo-S consortium developed a reference model that integrates services in the existing customer supply chains on an operational level.

CIRP IPS2 Conference 2009

336

2

DESIGN OF A NEW REFERENCE MODEL

2.1 Reference Model definition A reference model is a model representing a class of domains [9]. It is a conceptual framework that could be used as the blueprint for system development. Reference models, usually labeled as “common practices” are generic conceptual models that formalize recommended practices for a certain domain. They claim to capture reusable efficient state-of-the-art practices and have the main objectives of streamlining the design of enterprise individual models by providing a generic solution, reducing the costs of designing models and facilitating the management and control of the organization. [10] Fields of application of reference modeling address all levels and business fields of enterprises; therefore depicted domains can be very different. They can range from selected functional areas such as accounting or Customer Relationship Management to the scope of an entire industry sector, e.g. higher education. The main objective of reference models is to streamline the design of enterprise individual (particular) models by providing a generic solution. The application of reference models is motivated by the ‘Design by Reuse’ paradigm. Reference models accelerate the modeling process by providing a repository of potentially relevant business processes and structures. Moreover, with the increased popularity of business modeling, a wide and quite heterogeneous range of purposes can motivate the use of a reference model. Generally speaking, capturing reusable efficient state-of-theart practices, a reference model serves the following key objectives: (i) Streamlining the design of enterprise models by providing a generic solution, (ii) Reducing the costs of designing models, (iii) Facilitating the management and control of the organization, (iv) Facilitating description and optimization of organizational issues, (v) Helping to develop enterprise specific models including the reutilization of business knowledge. 2.2

Industrial Services Reference Model development methodology In the InCoCo-S project, the Reference Model was built following the Road Map to develop a reference model, suggested by Fettke et al.[11], made by three different Phases, corresponding to the definition of Domain, Features and Language (Phase 1), Modeling Approach (Phase 2), Construction Methodology (Phase 3).

The identification of Domain, Features and Language was relatively easy, having quite a clear picture of a service supply chain reference model that could help in improving integration of service providers into an existing supply chain considering also new and innovative business processes. Therefore to fulfill the second phase a deep literature research for relevant models that may serve as a baseline for the development of the InCoCo-S reference model has been carried out. Thereafter a “screening” process was performed to identify only relevant models, to be taken into consideration. After these activities the following reference models were considered to be relevant: SCOR reference model [12], Y-CIM reference Model [13], Aachener PPS reference model [14], GSCF [15] and CPFR [16]. Among the existing reference model, no one seemed to fit on full basis to the representation of interactions between manufactures and their service providers. Nevertheless, even though SCOR shows to be inadequate from the point of view of the service sector, it represents the best choice as a baseline for the development of the Industrial Services Reference Model (IRM) since: (i) service related activities, even though not described in detail are inferred. This fact facilitates the generation of an interface model for the service provider domain that could be in the future interfaced with SCOR; (ii) includes KPI that facilitate a better coordination and control during the supplier-manufacturer relationship; (iii) includes best practices that help companies activities benchmark and improvement. Since a suitable reference model basis was found, the following activity to be performed was the enhancement, development and evolution of the SCOR reference model to fit with the InCoCo-S special requirements. 2.3 Grounded Theory Methodology (GTM) Different procedures can be used to enhance, develop and evolve the starting point reference Model to fit with specific requirements. Nevertheless recommended procedures are very general or too specific and give only a summary description of the process followed. And the use of these recommended procedures result inadequate as basis of a scientific study. However, when selecting a research methodology two class of requirements should be considered: research requirements for a scientific method and the requirements relevant to the topic under study. On one hand, according to Strauss regarding research requirements, the researcher should take into consideration the following criteria: appeal, goals of the researcher, cost rigor, interpretations, usefulness and so forth. [17]. On the other hand the reference model to be developed should also have some features to reduce and control the complexity of the modeling process. These requirements correspond to

the six principles of GoM theory developed by [18]. These principles are: accuracy, relevance, cost effectiveness, clarity, comparability, and systematic construction. [19] Among the different scientific research strategies with a qualitative orientation, the Grounded Theory Methodology (GTM) calls special attention. The use of GTM attracted the attention of researchers not only of social and educational fields. Since its creation in 1967 by Glaser and Strauss, it has been used not only by academic researchers but also by managers, businessmen and professionals as they couldn’t find a better way to explain practical phenomenon.[20] Comparing the different scientific qualitative methods available, taking into consideration both the requirements of the research and topic under study, the grounded theory method fits exactly with the research needs. Furthermore, it is particularly suitable to those investigations for which little theory has been developed, where theory should emerge from data [21], which is exactly the case of the methodology to develop reference models that are large in number but the methodology used to develop them is hardly found. The synthesis of the main reasons why it was decided to use the grounded theory as the research approach to develop reference models are hereafter explained: (i) theories are derived from observation, the Reference Model had to emerge from the practice and the practice will be represented on it. (ii) GTM allows the researcher use his own experience and practical knowledge to generate new theories, (iii) the process of constant comparison assures a practical oriented and long lasting end product, (iv) it is a scientific method and has been used successfully in business and management research. The founders (Glaser & Strauss) of this qualitative research methodology define it as an inductive method “derived from the study of the phenomenon it represents, that is, discovered, developed and provisionally verified through systematic data collection and analysis of data pertaining to that phenomenon. Therefore, data collection, analysis and theory should stand in reciprocal relationship with each other. One does not begin with a theory and then prove it. Rather one begins with an area of study and what is relevant to that area is allowed to emerge.” [22] In general, grounded theory is the systematic generation of theory from data. Grounded theory is derived inductively from observation following a specific research procedure that allows the researcher to develop a theoretical account of the general figure of a topic while simultaneously grounding the account in empirical observations or data [23]. Theories are developed based on empirical observation, due to this special characteristic; it has been used mostly for

Figure 1: GTM adaptation to the specific case

337

domains that have no theory available. But whenever it is possible, the best domain theory should be used to give the theoretical framework of the initial study [24]. Since the SCOR reference model alone does not fulfill all the requirements of the new reference model, on the basis of the SCOR model a new one was developed with a procedure based on the GTM. For the specific case of this study, the GTM adaptation used can be represented as shown in Figure 1.

y

Support. Support Phase provides the infrastructural support to run and execute the service operations right from the Adapt phase to Operate phase. Key goal of support phase is to first define consistent & global business rules, governing principles which shall guide the overall service organization. The second key goal of support phase is to ensure unrestricted & uniform access to information across all service activities and all hierarchy levels.

3.2 3 THE INCOCO-S REFERENCE MODEL The Industrial Services Reference Model (IRM) is a rolebased process reference model based on enterprise architectural standards. It integrates AS-IS process mapping and TO-BE process design capabilities with operational performance setting and measurements to enable service providers and their business partners to achieve best-inclass results, built on proven industry practices and supported by solutions and tools. The comprehensive IRM process repository contains standardized descriptions of service processes on different levels of detail, structured in hierarchies to enable users to perform root-cause-analyses of performance variance. IRM processes are linked through input-output relationships and offer partner interface processes to other BPM standard models. The IRM structure covers supply-chain related business services in manufacturer-service provider networks. 3.1 IRM Level 1 – Strategic Process Types Level 1 processes define the entire scope of the service operations which are incorporated in the model. IRM incorporates all processes from the first service contact to continuous service operations using a service lifecycle approach. Based on this approach, five key process types have been identified namely: Plan, Adapt, Build, Operate and Support. y Plan. Since services cannot be stored as inventory, the plan service is most crucial for a successful service business and covers ongoing activities within service business. During the Plan phase a framework of procedural methods is first established for the entire span of the service supply chain and its operations. y Adapt. The Adapt Phase is primarily concerned with adapting the Service Portfolio offered by the service provider to the specific customer requirements in order to develop a service offer, which fulfil the customer needs. Key goal here is to fulfill the customer needs with the existing / enhanced service portfolio in an effective, efficient and reliable manner. y Build. The Build Phase is the phase of service implementation, where the service provider and the service network bring together all the resources – hardware, software and personnel oriented to implement the service solution for the customer. The key aim of Build Phase is to reliably and efficiently implement the service solution so that the service operations can commence to the utmost satisfaction of the customer and service provider. y Operate. Operate phase envelop value-adding core business transactions: here the services agreed between the customer and service provider are operated as per the service terms & conditions. Achieving the service levels and bring the benefits to the customer through enhanced performance are the key goals of the service provider. In addition, the service provider further strives to continuously improve service performance by evaluating the service performance and identifying potential for improvement.

338

IRM Level 2 – Configurations for Service Clusters Level 2 refers to customer fulfillment strategies to Plan, Execute and Support a certain business process in an industry specific environment. Whereas the SCOR Model differentiates between, Make-to-Stock, Make-to-Order and Engineer-to-Order approaches to satisfy customer needs in a supply chain environment, InCoCo-S has defined 5 service clusters to deliver business related services to supply chain customers. Using an iterative approach the IRM has been defined for five different services namely Logistics, Maintenance, Retrofit, Packaging and Quality Control services. The clusters processes have been defined together with the active participation of the industrial partners, especially business cases and SMEs. 3.3 IRM Level 3 – Process elements Process elements are a decomposition of Level 2 configurations and usually the lowest level of detail in a reference model. Level 3 processes define the transition from a generic reference model to a customer specific workflow. IRM Level 3 includes all the processes within the considered service which are of competence of the service provider. It models service supply chains as sets of connected processes: more than two hundreds processes at level 3 have been identified. This structure represents the backbone of the IRM model. To this backbone many other element are attached to form the entire IRM organism. Some other relevant information have been then added to the basic framework, including input/output, Best Practices and Key Performance Indicators. Best Practices and Performance Indicators are strictly related since the results achieved through Best Practices can be measured in an improvement of Key Performance Indicators. As a result, each process of IRM is characterized by a set of input coming from other processes, outputs, going to other processes, best practices that could improve the process performance and Key Performance Indicators that allow process performance measurement. An example of such result can be seen in Figure 2, where a process is represented with all its main attributes. 3.3.1 Input/output relationship All the processes are connected the one to the other thanks to Input / Output relationships. In the IRM each process is characterized by some inputs and outputs connected together to form a flow, which defines sequence and associated interdependency. Inputs can come from outputs of another process or from an entity external to the service provider (e.g. a customer or a supplier). In the same way the output of a process can go to one or more other processes and/or to some external entities. This sequence of inputs and outputs represent the flow of information that ideally accompanies the execution of the service, from the first contact with the customer to the service termination. Each output of a process is the input of one or more other processes and this allows “moving” along the IRM process structure. Process interactions and relationships are highlighted by inputs and outputs and form the basis of the framework.

Figure 2: Example of an InCoCo-S service – supply chain process 3.3.2 Performance Indicators (PIs) Performance indicators (PIs) present important information for the control and management of processes in terms of efficiency and effectiveness and the resulting service outcome. They measure whether set performance goals were met, and support the identification of improvement potentials. In the IRM, every process has PIs assigned on each level, indicating what needs to be measured while performing the process. On the strategic level, operational PIs are aggregated to key performance indicators (KPIs), which are important for benchmarking purposes, internally when comparing the performance of different departments or plants performing similar activities, or externally when benchmarking the service operation performance with the competitors. The PIs are structured in a certain manner, forming altogether the so-called Service Performance Measurement System (SPMS). The SPMS was designed to quantify the efficiency and effectiveness of service operation activities from a holistic perspective. The SPMS is structured into three dimensions: Service Object, Service Activity (forming together the Service Encounter interface) and Customer Service Satisfaction. Service activity (SA) is the measure of the service providers own internal service process performance in terms of how efficient and reliable the processes are. The dimension of Service object (SO) focuses on the performance of the objects (e.g. performance of machines in case of maintenance services) in the customers` manufacturing supply chain which are being serviced by the service provider. The basic idea of the Customer Service Satisfaction is to have a dimension quantifying the gap between actual service operation performance based on the objective measures from the Service Encounter Interface and the perceived service operation performance from the customers` perspective. For each category five key target areas have been identified to measure service performance holistically from both service provider and customer perspective: y Service Reliability: These PIs relate to the ability to achieve an intended or agreed service operation level or availability. Reliability refers to the ability to perform a required operation under stated conditions for a stated period of time. y Service Responsiveness (Time): These PIs reflect the time between the beginning and completion of SA and measure time related to SO (e.g. order lead time).

y

y y

Service Flexibility / Adaptability: These PIs provide the basis for measuring the ability of SA and SO to adapt to changing requirements in terms of time and volume/intensity. Service Assets / Costs: These PIs highlight the financial expense to carry out SA and run the SO. Service Efficiency / Productivity: Includes relative PIs measuring static PIs in relation to time and costs and shows how efficient the resources are being used in transforming inputs to outputs.

3.3.3 Best Practices (BPs) Best practices can be defined as the most efficient (least amount of effort) and effective (best results) way of accomplishing a task, based on repeatable procedures that have proven themselves over time for large numbers of cases. Best Practice consists in a technique, method, process, activity, incentive or reward that is more effective at delivering a desired outcome with fewer problems and unforeseen complications. In the IRM model, once the performance of the process has been measured and performance gaps identified, it becomes important to identify what activities should be performed to close those gaps. For each process a set of best practices is identified according to the relevance in the ambit of the service – supply chain and to the capability of making the process achieve “best-in-class” performance. To assign one or more best practices to process, several best practice were deeply analyzed, to understand the service supply chain process(es) where, if adopted, could result in “best-in-class” performance. During this task, a classification of best practices, reflecting a structure similar to the IRM phases was used. [25] Moreover, beside the best practices identified in literature, new ones were also defined in order to address the peculiarities of the maintenance service supply chain. These novel best practices were usually developed as modifications and adaptations of existing managerial practices and mainly derived by the experience of the project industrial partners. 4

POTENTIAL USAGES OF IRM

4.1 Analyze & Optimize Business Processes Users who want to define organization specific processes and develop & evaluate AS-IS and a TO-BE scenarios before implementation of a business transformation project can apply the DMAIC (Define, Measure, Analyze, Implement and Control) approach, derived from an alignment of BPM

339

and Six Sigma toolboxes. The IRM methodology supports the user in decisions on breakthrough or continuous improvement focus beforehand, defining strategic goals and operational targets in enterprise context, concentrating on quick and easy leverage critical to value and quality, defining fact-based baselines and agreeing upon measurement systems and KPIs to be used. A comprehensive set of practice proven tools can be used to agree on process improvement potential, define the process maturity of involved community, describe desired impact on business results, develop AS-IS, SHOULD-BE, TO-BE scope and scale, define process owners and targets, document level of detail and degree of visual aids, document improvements to be achieved and areas to be addressed. 4.2

Standardize the business processes for internal & external communication Benchmarking / Knowledge Management Companies whose focus is on optimization, standardization and automation of internal process execution are supported in program and process management methodologies to develop and prioritize scope and scale of improvement projects. Hereby the user is guided through a multistep iteration, based on sequences of processes he can choose from depending on the project roadmap. Examples include the description of end-to-end process flows on different levels of details, allocation of resources and responsibilities, the identification of process inputs and outputs and the description of interfaces and measurement points on the respective process level. Once the analysis of a given AS-IS performance in process execution has been done, a description of input-output relationships and their impact on performance levels can be addressed. A cross-functional team now can elaborate a project roadmap and measurement plan, collect and evaluate company specific data and process disconnects, analyze and prioritize required outputs relevant for a desired TO-BE performance. Again the comprehensive toolbox supports users if needed in defining long-, medium-, and short term improvement targets, developing potential scenarios and their impact on the company baseline and applicable ICT solutions. 4.3 Use of IRM for Benchmarking Purposes Having standard processes using the IRM structure, the according performance indicators allow for a comparison of process effectiveness and efficiency. When harmonizing the own organization's process flows, different departments or regional offices can be easily compared and improvement potentials identified. The aim within the IRM user group is to establish a database with values for the different performance indicators, especially the key PIs, which represent real data. These values will be linked to strategic setups, so that other IRM users can match their own performance against competing strategies. 4.4 Facilitation of Software Implementation IRM is used to align a multitude of IT Systems with business processes into a streamlined IT application which supports the business processes in a synchronized manner. The software solutions can be developed on the basis of the internal process as defined by the in-house process management team. In addition, IRM can also be used by a supply chain partner to align the service operations of their service providers and to benchmark the performance of different service providers using a common service scorecard. IRM supports interactions with other proprietary models such as SCOR, the Aachen PPC Model and GSCF in the supply chain

340

domain in order to develop an integrated process framework incorporating both service and supply chain domains. 5 IRM VALIDATION AND STANDARDIZATION According to GTM the results achieved with the IRM are consequence both of the literature analysis and of the model refinement achieved thanks to the Industrial Partners support. In particular, a first validation of the IRM has been conducted on the basis of industrial business cases. Refining the model through industrial partners help has been just the first step to achieve a real validation. In fact on the basis of the InCoCo-S project external validation plan, more than 30 extensive validation tests were carried out to the end of March 2008. In the validation tests the IRM was used in industrial services offering companies to map the AS-IS scenario, in order to warrant the flexibility of the model and its adaptability to different business cases. Moreover the IRM was used to suggest possible reconfiguration and remodeling of the service offered either reorganizing processes or using best practices to improve performance. It must also be considered that the 3rd level of detail of the Industrial Service Reference Model is not to be considered the “final” one because further decomposition of processes is possible on the basis of company specific information. From this point of view, during validation, the possibility of easily represent with such higher details real industrial processes was also tested. Furthermore, thanks also to validation, testifying its capabilities, Industrial Services Reference Model was proposed for standardization, becoming CEN Workshop Agreement (CWA) WS 39, a valid industrial standard for business services modeling and improvement. This is also be eased by the full compatibility of IRM with well known and in use complementary models, like SCOR: the possibility of interconnect SCOR to model manufacturing processes and IRM to model business services and the similar hierarchical, process based structure, is for sure an important advantage to improve IRM diffusion and attitude towards its use. 6 CONCLUSIONS Starting from the assumption that the increasing importance of the business service offer has to be faced with new and innovative tools to improve companies’ effectiveness and competitiveness, this paper showed the main characteristics of the Industrial Services Reference Model, and how, using the framework a company can model its processes and find hints for improvement, both considering the suggested process interaction and using Key Performance Indicators and Best Practices to monitor and improve performances. This in order to warrant to future users of the IRM framework the possibility, not only to model the service offered but also to supply tools to improve the modeled processes. The final objective of the InCoCo-S project, to manage the interdependency between the manufacturing supply chain and service providers was achieved by developing a service oriented reference model that could increase and sustain the overall performance and competitiveness of both manufacturing supply chain and their service providers. 7 ACKNOWLEDGEMENT This work has been partly funded by the European Commission through NMP Project InCoCo-S: Innovation, Coordination and Collaboration in Service Driven Manufacturing Supply Chains (No. NMP2-CT-2005017192). [26] The authors wish to acknowledge the Commission for their support. We also wish to acknowledge our gratitude and appreciation to all the InCoCo-S project

partners for their contribution during the development of various ideas and concepts presented in this paper. 8 [1] [2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

REFERENCES. Illeris S. 1996. The service economy: a geographical approach. J. Wiley & Sons, 1996. Cook, D.P., Goh, C.H., Chung, C.H., 1999. Service typologies: a state of the art survey. Production and Operation Management 8 (3), 318–338. Logan, S.M., Fraught, K., and Ganster D., Outsourcing a satisfied and committed workforce: a trucking industry case study. International Journal of uman Resource Management 15, 147, 2004 Vaidyanathan, G. A framework for evaluating third party logistics, Association for Computing Machinery. Communications of the ACM, 48, 89, 2005 Ma, Q., Tseng, M.M., Yen, B., 2002. A generic model and design representation technique of service products. Technovation 22, 15–39. Knecht, T., Leszinski, R., Webe, F.A., 1993. Making profits after the sale. The McKinsey Quarterly 4, 79– 86. Wise, R., Baumgartner, P., 1999. Go downstream— the new profit imperative in manufacturing. Harvard Business Review 77 (5), 133–141. European Commission (2003). The competitiveness of Business Related Services and their contribution to the performance of European enterprises. Fettke, P. and P. Loos (2003). "Classification of Reference Models. A Methodology and its Application " Information Systems and e- Business Management 1(1): 35 - 53. Rosemann, M.: Application Reference Models and Building Blocks for Management and Control (ERP Systems). In: Handbook of Enterprise Architecture, eds.: P. Bernus, L. Nemes, G. Schmidt. SpringerVerlag, Berlin, Germany, pp. 595-615. Fettke, P., P. Loos, et al. (2005). Business Process reference Models: Survey and Classification. Proceedings of the Workshop on Business Process Reference Models., Nancy - France Supply Chain Council (2006), Supply Chain Operations Reference Model SCOR® Version 8.0, http://www.supply-chain.org Scheer, A.-W. (1994), Business Process Engineering : Reference Models for Industrial Enterprises, 2nd ed., Berlin, Springer.

[14] Luczak, H., Kees, A. (1998). Das Aachener PPSModell. In: Proc. Referenzmodellierung '98: Anwendungsfelder in Theorie und Praxis, 14. Juli 1998, RWTH Aachen (1998) 2-1 - 2-9 [15] Global Supply Chain Forum (2007), Global Supply Chain Forum (GSCF) Supply Chain Management Framework http://www.fisher.osu.edu/centers/scm/ about-the-forum. [16] VICS (2007), CPFR, http://www.vics.org.CE-NET: The Concurrent Enterprising Network of Excellence. WWW page. http://www.ce-net.org/index.html. [17] Strauss, A. and Corbin, J. Basics of Qualitative Research: Grounded Theory Procedures and Techniques. Newbury Park, London, New Delhi: Sage Publications, 1990. [18] Becker, J. 1995. Strukturanalogien in Informationsmodellen. Ihre Definition, ihr Nutzen und ihr Einfluß auf die Bildung von Grundsätzen ordnungsmäßiger Modellierung (GoM). In: Wirtschaftsinformatik ´95. Wettbewerbsfähigkeit, Innovation, Wirtschaftlichkeit. Hrsg.: W. König. Heidelberg 1995, S. 133-150. [19] Becker, J. (2006). "LS-IS: Grundsätze ordnungsmäßiger Modellierung (GoM)." From http://www.wi.uni-muenster.de/is/projekte/gom/index 6.cfm [20] Glaser, B.G. and Strauss, A.L. The Discovery of Grounded Theory: Strategies for Qualitative Research. NY: Aldine de Gruyter, 1967 [21] Glaser, B., A. Lowe, et al. (2000). "The Grounded Theory Review." The Grounded Theory Review: An International Journal, 2. [22] Minnis, J.R. (1985). ‘‘Ethnography, case study, grounded theory and distance education research.” Distance Education, 6,2,189-198. [23] Pandit, N. (1996). "The Creation of Theory: A Recent Application of the Grounded Theory Method." The Qualitative Report 2(4). [24] Endres-Niggemeyer, B. (1999). "Grounded theory methodology for knowledge engineering." http://www.cse.unsw.edu.au/~achim/Empirical99/En dresNiggemeyer.pps. [25] Gerosa M., Montorio, M., Taisch, M., Cassina, J., (2007), Improve collaboration and coordination in service - supply chain operations: the InCoCo-S approach, CARV2007, Toronto, 22-24 July 2007. [26] InCoCo-S website: http://www.incoco.net

341

CIRP – IPS2 2009

The following topics are covered: • PSS design: methodologies and challenges

Most western manufacturing companies are shifting their focus of business strategy towards selling services or functionality instead of products. The product-service system (PSS) strategy has a great impact on customers, product life cycle and company strategy. The design of PSS is a complex problem, and must meet the challenges of the changing financial and resource models that align with PSS strategy.

• PSS requirements engineering and management

New developments are taking place in industrial markets in the form of Industrial Product-Service Systems (IPS2). IPS2 represents a change in the competitive strategy for manufacturing companies, enabling innovative function, availability and result oriented business models.

• PSS evaluation techniques

The proceedings present multidisciplinary research encompassing concepts, methodologies and infrastructure development for successful IPS2.

• Service and supporting network in the PSS environment

Professor Rajkumar Roy

Dr. Essam Shehab

Rajkumar Roy is Professor of Competitive Design and Head of the Decision Engineering Centre at Cranfield University. He is also the President of the Association of Cost Engineers. His research interests include design optimisation and cost engineering for products, services and industrial product-service systems.

Essam Shehab is a Senior Lecturer in Decision Engineering at Cranfield University. His research and industrial interests cover multi-disciplinary areas including design engineering, cost modelling and knowledge management for innovative products and industrial product-service systems.

• Product, service and PSS knowledge representation • Service knowledge capture and reuse in PSS design • PSS information and knowledge management • Digital product life cycle systems for PSS • Service engineering • PSS life cycle management • Review of PSS design approaches • Impact of in-service and disposal issues on design • Impact of informated product in use on design • PSS: business requirements and techniques • Organisational complexity in the PSS model • Life cycle cost modelling for PSS

• PSS: strategy and transition

Industrial Product-Service Systems

CIRP – IPS2 2009

Industrial ProductService Systems (IPS2) Proceedings of the

1st CIRP IPS2 Conference

Rajkumar Roy Front cover car manufacturing image courtesy of Toyota

Essam Shehab

Rajkumar Roy, Essam Shehab Editors