Industrial Product-Service Systems Industrial ...

7 downloads 47 Views 20MB Size Report
automotive industry requires pre-production services. (such as ..... systems, i. e. for the automotive industry. The unique ...... as product maintenance and training that continuously ..... this case, service technicians are sent to the customer in.

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 Est