15.4 Increasing energy efficiency through simulation ...

15.4 Increasing energy efficiency through simulation-driven process evaluation J. Stoldt 1, D. Neumann 1, T. Langer 1, M. Putz 1, A. Schlegel 1 1

Fraunhofer Institute for Machine Tools and Forming Technology, Chemnitz, Germany

Abstract Continuous improvement of the production efficiency is one of many goals a company has for attaining a sustainable market position. When considering traditional objectives, benchmarks are used to compare numerous improvements concerning required inputs and created outputs. Adopting this approach for increasing the energy efficiency of individual manufacturing steps is difficult as their comparability is usually low. However, motivated by the need for competitiveness, regulatory mandates, and a desire for proactive green behaviour, companies seek for sustainable measures to make their manufacturing operations environmentally benign and thus require means for the ecologic assessment of their production processes. This paper presents a novel approach to benchmarking and process evolution, which minds both traditional economic and energy related key performance indicators (KPI). For this purpose, a procedure model making use of an enhanced material flow simulation system has been developed to evaluate and scrutinize the energy efficiency in production processes. Keywords: Benchmarking; Energy efficiency; Process evolution; Simulation

1 INTRODUCTION Companies need to be innovative throughout their operation, if they want to sustain or improve their market position. This includes, for instance, the development of products, the advancement of business processes, or the optimisation of production operations. Hence, it is a management task to promote and monitor innovative developments. One common tool to foster business improvements is benchmarking. Recent surveys rank it as the most popular management instrument [1]. Its success is based on the ability to increase a company’s “performance by identifying and applying best documented practices,” using key performance indicators (KPI) to evaluate the efficacy of organisations [1]. The comparison with other companies and the afore-mentioned focus on “best practices” are the reasons for its great improvement potential [2]. While numerous KPI exist for specific purposes, it is difficult to compare the ecological efficiency of different production processes and technologies. However, in light of expected supply shortages of raw materials, as well as fossil energy carriers (i.e. oil, coal, etc.), and with respect to the effects of an increased usage of natural energy sources (i.e. wind, sun, tides, etc.), companies have to mind their resource usage behaviour. This is intensified by customer demands, regulatory mandates, the pursuit of competitive economic advantages, and the desire for proactive green behaviour, which act as motivators for environmentally benign manufacturing [3]. In conclusion, there is a strong need for management instruments that help to foster the energy efficiency of complex production processes. If they are expected to gain broad acceptance they need to be truly generic, thus allowing for the evaluation and comparison of various production technologies, as well as combinations thereof employed in any desired factory. While benchmarking is already a state of the art tool for improving the resource efficiency of processes [4], this paper introduces a new approach which allows for the comparison

of different production technologies. Unlike other work in this field of research (e.g. [5-7]) it is focussed on the evaluation of manufacturing processes on an arbitrary level of detail. Following the “performance per watts” KPI in information technology (IT) [8], this novel methodology aims to evaluate the performance of production facilities and equipment. In order to support the implementation of energy efficiency increasing measures, it will make extensive use of energyenriched material flow simulation. Hereby, it will be possible to effectively predict the impact of planned changes on the energy usage and the overall system productivity. Using simulation software for benchmarking purposes is not unique to the approach presented in this paper (e.g. [9]), however, the ability to be able to conjointly regard traditional (economic) and energy-related (ecological) KPI in a single simulation model is. The work presented in this paper is to be understood as a proof of concept for the developed methodology. Hence, all of the examples demonstrate the basic capability of this novel approach but do not use data acquired in a real production environment. Hereafter, the general methodology and the newly developed (energy-related) process efficiency coefficient will be elaborated. Section 3 discusses how measures increasing energy efficiency can be devised and implemented utilising this new approach to benchmarking. The following section 4 describes how the methodology can be implemented into a simulation model and how the necessary data can be obtained from existing IT systems in the production environment. A brief summary will then conclude this paper. 2

BENCHMARKING THE ENERGY EFFICIENCY OF PRODUCTION PROCESSES

Including ecological considerations into the benchmarking methodology requires changes to the basic methodology as

G. Seliger (Ed.), Proceedings of the 11th Global Conference on Sustainable Manufacturing - Innovative Solutions ISBN 978-3-7983-2609-5 © Universitätsverlag der TU Berlin 2013

473

J. Stoldt; D. Neumann; T. Langer; M. Putz; A. Schlegel

well as the definition of novel KPI. These are described in the following subsections. 2.1

General methodology

The continuous advancement of all business processes allows for the gradual improvement of the organisational performance. Further combining benchmarking with value chain management allows for even greater efficiency throughout the production system. If these conventional considerations are expanded to include aspects of energy efficiency, the potential to realise sustainable improvements increases significantly. Traditional benchmarking has been the subject of numerous publications; accordingly, there are varying suggestions for its conduct. Andersen describes the procedure utilising a wheel metaphor, emphasising the iterative character of the method [10]. According to this author, the following five steps should be performed: 1. Plan: Critical success factors, select a process for benchmarking, document the process, and develop performance measures. 2. Search: Find benchmarking partners. 3. Observe: Understand and document process, both performance and practice.

the

partners'

4. Analyze: Identify gaps in performance and find the root causes for the performance gaps. 5. Adapt: Choose "best practice", adapt to the company's conditions, and implement changes. In contrast to a traditional benchmarking process, the new approach presented in this paper aims to allow for the immediate comparison of technological process regarding energy efficiency. For this purpose, a number of adjustments have to be made, resulting in the following seven steps: 1. Definition of traditional and energy-related KPI, level of abstraction, and processes/ equipment to be analysed. 2. Measurement of relevant data and description of the processes with their respective key production features. 3. Analysis of relevant data, evaluation of KPI, and ranking of individual processes. 4. Analysis of result and feature varieties as well as identification of related improvement potentials. 5. Identification and adaption of “best practices”. 6. Simulation-based alterations.

assessment

of

planned

process

7. Implementation of effective measures. The measurement of data in step two can be made either in the actual production environment or in a simulation model thereof. This allows for prospectively benchmarking production systems which are still in a design phase in order to determine an optimal solution, from both an ecologic and an economic point of view. The explicit definition of the simulation of planned process alterations as part of the methodology aims to minimise the risk of failure through unintended side effects of their implementation. These are more prone to appear due to the increased complexity of the considerations caused by simultaneously aiming for both an economic and an ecologic optimum.

How this general methodology for benchmarking can benefit from value chain management will be detailed in the next subsection with respect to a process efficiency coefficient. The latter is the core KPI for the afore-described approach. 2.2

Quantifying (energetic) process efficiency

Individual production processes should be assessed just like companies are. The most relevant figures for any operation are profit (P), revenues (R) and costs (C). These have the relationship:

P  R C

(1)

Revenues are generated by selling the actual products which is usually not possible for individual steps of the production process. Accordingly, there is a need to determine a replacement for revenues. Considering that any production step should aim to add value to the final product, added value (AV) is suitable. Consequently, the target figure can no longer be called profit, as no actual profit has been made at any point in the manufacturing process. Hence, the pseudo target figure rectified added value (RAV) replaces profit:

RAV  AV  C

(2)

A process efficiency coefficient (EC) can be deduced from this formula, keeping in mind that greater efficiency equals a greater rectified added value:

EC 

AV C

(3)

It is apparent that by definition, the costs may not be zero, which is, however, unlikely in a real production environment. This definition includes costs as a single factor although there are various factors within the overall production process which induce costs, such as labour, write-offs, raw materials, consumables, maintenance, energy, peripherals (e.g. lighting, or climate control), and so forth. While some may be quantified for individual process steps, many may not be. A mathematical expression of the relation of the costs and these cost factors (CFi with i=1,…,I as the respective factor) – which are usually determined by means of a measurable consumption (Consi) and a cost per unit rate (CPUi) – is: I

I

i

i

C   CFi   Consi  CPUi

(4)

In order to work around the tedious effort necessary to determine the overall costs – if even possible for individual production steps – the approach presented in this paper does not aim to use a single KPI (e.g. overall efficiency coefficient) but rather a flexible array of similar KPI. For this purpose, costs are replaced with the actual quantifiable consumption. Accordingly, the process efficiency coefficient (ECi) is redefined to be unique for each cost factor:

ECi 

AV Consi

(5)

A more specific version of this coefficient is the energy efficiency coefficient (eECi), which replaces the abstract concept of consumption with the actual consumption of energy or energy carriers (eConsi), such as electricity, compressed air, or cooling water. It is defined as follows:

eECi 

AV eConsi

(6)

474

Increasing energy efficiency through simulation-driven process evaluation

of each process is determined by the slope of the line connecting the point of the process and the point of origin.

F

Value added

A similar KPI has also been suggested by Reinhart et al. [11]. In contrast to these authors, it should be noted that the energy efficiency coefficients are expected to be applied to various carriers and alongside other traditional performance KPI (e.g. added value per hour of labour). The individual results should then be interpreted together in order to assess the process’ overall efficiency. Which factors should be considered has to be determined in step one of the benchmarking methodology (see previous subsection). Acquiring data for the consumption of any cost factor is dependent on the characteristics of these. The added value should always be determined by means of value chain management methods, e.g. value stream design.

3. 2.

C

1. 4. B

E A

D Consumption

Figure 2: Visualisation of efficiency improvement approaches. 3

MEASURES FOR INCREASING ENERGY EFFICIENCY

The presented benchmarking procedure allows for the comparison of different processes of a single company or of different companies. These processes can be planned and optimised taking differences between themselves and other more efficient implementations of the same or a different technology into consideration. In order to improve the results, a systemised approach should be followed. Figure 1 depicts a more in-depth overview of the tasks which are included in steps four and five of the benchmarking methodology presented in section 2.1.

Ranking of analysed processes Joint and individual comparison of determined KPI Identification of root causes for performance differences Deduction of related improvement potentials Identification of “best practices” Development of measures to adopt “best practices” Accessment and implementation of measures Figure 1: Process steps for deduction of measures. Accordingly, the first step is to compare the determined KPI of the ranked processes with each other. In order to assess the differences thoroughly and to identify correlations, the various KPI should be compared individually and conjointly. Having determined the major performance differences, their individual root causes should be investigated. These are very much dependent on the compared processes and their technologies. Hence, they provide input on the question how the process efficiency can be increased. The deduction of this particular piece of information, i.e. improvement potentials related to the results of the comparison, is part of the following step. Once the potentials to be tapped have been decided, “best practices” should be identified. While the compared processes may already be labelled the “best practice”, they do not necessarily have to be. Consequently, a search for the best available solution should be made. Based on the results of the previous steps, measures aimed at the adaption of the “best practice” have to be developed. For this purpose, the two dimensions of the presented (energy) efficiency coefficient should be considered. Utilising these, all assessed processes can be visualised on a plane, as depicted in figure 2. It should be noted that this portrayal is unique to any specified KPI (ECi / eECi). The actual efficiency

In order to devise measures which improve the efficiency of a production process, four fundamental approaches exist (see figure 2), which should be combined with the knowledge of the identified “best practices”. These are explained hereafter: 1. Decreasing consumption: Basically two ways to reduce the consumption of energy exist. On the one hand, equipment losses can be minimised by utilising more efficient components and higher grade consumables. On the other hand, avoidable wasteful behaviour (see [11]), such as prolonged equipment idle time, can be reduced. Decreasing the consumption of other cost factors is also closely related to the elimination of waste, as propagated by Lean Management [12] and the Toyota Production System [13]. 2. Increasing added value: The basic idea of this approach is to increase the added value without changing the consumption of cost factors. This can be achieved by intensifying the labour, for instance, so that more work is completed in the same time. 3. Combination of 1. and 2.: While most devisable measures are likely to be a combination of 1. and 2., there are some situations when considerable effort is put into decreasing consumption and increasing added value. The integration of entire process chains (e.g. [14]) is a prime example for this, where functionality of multiple machines is integrated into a single one. Thereby, the added value created by a single machine is increased and redundant components, such as multiple controllers, can be removed. 4. Process substitution: Instead of adjusting, advancing, and improving existing processes to improve their efficiency, they may also be substituted with entirely new processes. This approach is especially feasible, if newly developed technologies promise a significantly higher efficiency despite higher costs for their implementation. During the development of measures aimed at the adoption of “best practices”, the desired and anticipated effects on all regarded KPI should be considered. As this can be quite complicated, especially when using a large array of KPI, the basic methodology suggests simulations prior to the measure implementation. This will be discussed in more detail in the following section. 4

SIMULATION IN THE BENCHMARKING PROCESS

The importance of simulation within the benchmarking methodology has already been discussed in the previous

475


sections. More precisely, it can be used to generate initial input for the generation of KPI and to validate the effectiveness of devised measures prior to their implementations. The fundamental approach is hereafter exemplified by means of a use-case. 4.1

Implementation of the methodology

In order to move towards a holistic benchmarking which allows for comparing various processes, a wide variety of data needs to be assessed. This includes, amongst others, the consumption of energy, energy carriers, and raw material, as well as the required amount of labour, the size of material stocks, and equipment availability data. Regarding these conjointly becomes increasingly difficult, if larger production systems and more stochastic events (e.g. random equipment break downs) are considered. Discrete event simulation software solutions are ideal tools to reduce the necessary effort for such investigation. Hence, the hereafter presented approach makes use of the established Tecnomatix Plant Simulation software to determine input for the evaluation of specific KPI. For this purpose, a model of an exemplary production system has been created, which represents the manual assembly and finish part of a car body production. Versions of produced cars with both three and five doors are regarded during simulation. The production system itself consists of a main line with five interlinked and clocked work stations (WS), as well as eight subsystems, which supply – in teams of two – parts to the earlier four work stations. These four assemble the doors of the five-door version, the bonnet and tailgate, the doors of the three-door version, and the wings to the otherwise complete car bodies. Afterwards, employees in the fifth work station, a light tunnel, check for any imperfections which would diminish the quality of the final product. The individual subsystems, which supply the afore-mentioned parts to the main line, are made up of between one and five robot work groups (WG). A schematic of this system is depicted in the lower half of figure 3. All these work groups and work stations have been modelled utilising the VDA Automotive Toolkit [15]. As Plant Simulation does offer no functionality to regard the energy consumption of the production equipment during run time, an own solution, called eniBRIC, has been developed [16]. These eniBRIC modules enrich existing material flow objects – in this case from the VDA Automotive Toolkit –, providing a way to estimate the consumption of energy, energy carriers, and other process prerequisites. In particular, electricity, compressed air, cooling water, lighting, smoulder suction, laser light, and ventilation, as well as their respective suppliers are regarded. Once the simulation model had been created and enabled to regard energy consumption, data for its parameters had to be selected and entered. As this is a very important part of any simulation study, the following section 4.2 will cover it in more detail. Afterwards, preparations for the benchmarking process have been made. This entailed writing routines to collect all necessary input data for the KPI evaluation, paying close attention to selected measuring points. With respect to the definition of the EC and the eEC (formulas (5) and (6)), a consumption for any regarded cost driver had to be identified and subsequently measured during simulation. Furthermore, a data base of the value added by any simulated process needed to be created and stored for later references. These

sets of data are combined during or after individual simulation runs to evaluate the individually examined KPI, i.e. EC and eEC, as can be seen in the upper half of figure 3. eECE400

energy consumption

EClabour

labour time consumption … …

5 WG

5 WG

doors (five-door version)

3 WG

1 WG

bonnet / tailgate

efficiency coefficients

ECi

… 3 WG

benchmarking input data

added value (data base)

… … 3 WG

doors (threedoor version)

1 WG

subsystems (# of work groups)

1 WG

wings

light tunnel

work stations

Figure 3: Schematic of regarded production system and data relationships within the simulation model. The simulation model has been prepared accordingly, aiming to benchmark the processes (i.e. work groups and work stations) with respect to their consumption of electricity at 400 V (eECE400) and labour time (EClabour). In order to determine appropriate measuring points, preliminary considerations regarding the effects that should be included in the investigation have been made. For this purpose, a process chain of three non-elastically interlinked processes with identical cycle time is regarded (see figure 4). If a failure (red) occurs during operation (green) in the second process (P2), the time until it completes the current part and accepts a new part increases. Consequently, the prior process (P1) will be blocked (blue), as it can not commence work on a new part unless the previous one has left. Similarly, the following process (P3) has to wait (yellow) until P2 has finished part 3. Considering these interactions, it is clear that the measuring points can be at the beginning and the end of the processing, excluding foreign influences, or always at the beginning of the processing. As these interactions should be regarded for the following evaluation, the latter approach has been chosen for this investigation. Hence, the consumption of energy and (labour) time between two entering parts is logged for each process during simulation runs. The KPI are then evaluated using the average consumption. P1

Part 3

P2

Part 2

P3

Part 1

Part 4 Part 3

Part 5 Part 3

Part 2

Part 4 Part 3

Figure 4: Consequences of errors in a process chain. An exemplary application of the simulation model in a minor benchmarking study shall illustrate how the simulation approach benefits the general methodology. Having already defined KPI and the processes to be analysed, the status quo is investigated through simulation, as suggested for step two of the methodology (see section 2.1). Since the data is automatically used to evaluate the KPI, the results can be analysed immediately. These show clearly that the main line acts as the bottleneck of the entire production system, causing frequent blockages of the supplying subsystems. Skipping in-depth elaboration on best practices to avoid these blockages, a measure is devised to reduce them. In particular, the main line cycle time is reduced by 2 seconds, which resembles approach two identified in section 3. Taking into consideration that increasing the work intensity of manual labour – as present on the main line – can be very hard to achieve, a preliminary investigation should be made to

476

Increasing energy efficiency through simulation-driven process evaluation

estimate the potential. Hence, the simulation model is altered accordingly and used again in step six of the methodology, proving the general effectiveness of the devised measure. Table 1 depicts some exemplary data collected as part of the simulation effort in this benchmarking study. For this purpose, the added value of all processes has been set to 100 monetary units. Both experiments (before and after optimisation) had a duration of 45 days and have been repeated 50 times in order to reach statistical certainty despite randomly occurring events (e.g. break downs). Table 1: Exemplary results of the benchmarking study. Process name WS1 WGbonnet WGwings1 4.2

mean

eECE400 after

before

after

1,61

1,66

0,88

0,91

+0,05

∆ mean

0,80

∆

+0,03 0,83

1,46

+0,02

∆ mean

EClabour

before

1,68

1,50 +0,04

1,73 +0,05

1,46

1,50 +0,04

Data acquisition in a corporate environment

The easy and flexible acquisition of realistic and complete data for the parameterisation of the presented simulation model is of great importance. This is benefitted by the substantial and still increasing support of production systems by IT-based (data) management solutions. In order to foster the general quality of the data used for the simulation, it is an asset to extract the model parameters from these existing IT systems. Figure 5 gives an exemplary overview on possible data sources, the information they can provide, and specific data sets. Shift shedules ERP

 Work hours  Staff requirement  Work days

Work center calendar  Planned shifts  Produced products

Production program  Work plans

MES

Availability data

 Availability  MTTR/MTBF

Quality data

 Rejection rate  Rework rate

Allocation plans

 Occupancy rate  Historic data

Simulation parameter data

Consumption profiles  Trends EMS

PLM

Operating states

 Distinctive cons.  Number of states

Avg. consumptions

 per energy carrier  Reference values

System layout

 System dimensions  General layout

Dependencies

 Interlinkages  Supplier-Consumer

Process defaults Product data PDM

Bill of materials Machining details

 Cycle times  Time variability  Dimensions  Identification details  Assembly groups  Raw materials  Set-up times  Capable machines

Figure 5: Sources for simulation parameter data. Hereafter, the focus lies on the following sets of information:

 production program;  product quality;  consumption of energy and energy carriers;  common operating states of equipment;  equipment availability;  production system defaults (e.g. cycle time). Enterprise Resource Planning (ERP) solutions are state of the art in most companies. Through direct access to their underlying data base or using specific software-based application programming interfaces (API) it is possible to gather and to extract information regarding the production program of the system under consideration. This is important to ascertain which products are produced in which quantity, for instance. Furthermore, work plans for individual products can be determined, although these will usually have to be formalised to be usable in a simulation. Process- and quality related data is usually managed by specific modules of Manufacturing Execution Systems (MES) [17]. These pieces of information are frequently complemented by data on the consumption of energy and other process resources, which in turn may be acquired by Energy Management Systems (EMS) as process information relating to either individual products or employed equipment. For the approach to simulation including energy-related considerations introduced in the previous section, yet not exclusively for this one, it is important to determine certain operating states and their respective energy and energy carrier consumption. Such information can be derived automatically or semi-automatically from recorded consumption profiles, depending on the complexity of the profiles and the available software. If the simulated equipment and processes are to behave realistically, detailed knowledge of their availability and break down behaviour is required. Nowadays, corresponding data is predominantly gathered and analysed during production and stored in specialised availability management systems. Based on the concept of multivalent data usage various approaches to reuse already acquired data for other purposes than their originally intended one exist. The method for the product data based productivity assessment is an example in this context [18,19]. It aims to deduct information on the effectiveness of equipment in interlinked production systems from product data available in traceability solutions. Additionally, information regarding production process defaults – which often depend on the manufactured product types – is required to parameterise simulation models. This includes cycle times, the variability thereof etc. and is usually stored in Product Lifecycle Management (PLM) solutions. Product Data Management (PDM) software is also state of the art in most companies and holds most product related data, depending on the company specific implementation. All these systems store the data independently, which makes the data acquisition for simulation studies difficult. Hence, Fraunhofer IWU searches for and develops new approaches to creating links between all the individual data sets [20], in order to drastically reduce the necessary effort. 5

SUMMARY

In order to increase the energy efficiency of production systems a novel approach to benchmarking has been

477


presented. Its basic methodology along with the efficiency and energy efficiency coefficient allow for the assessment of the performance of individual manufacturing processes on an arbitrary level. This information can be used to identify improvement potentials and derive measures to exploit the latter, effectively improving both the ecological and economic performance of the regarded production system. Furthermore, it has been shown how the developed benchmarking methodology can be complemented with simulation studies and where necessary input data for this purpose may be acquired. At this point, the (energy) efficiency coefficient can only be utilised to assess the performance of value adding processes while many indirect processes, such as in-plant logistics, can not be considered. Future research will have to focus on KPI that close this gap. 6

ACKNOWLEDGMENTS

The presented work summarizes outcomes of the research projects InnoCaT® and eniPROD®. The pre-competetive joint research project "Innovation Alliance Green Car Body Technologies" is funded by the "Bundesministerium für Bildung und Forschung (BMBF)" (funding mark 02PO2700 ff) and supervised by "Projektträger Karlsruhe (PTKA)". The authors are responsible for the content of the publication. The Cluster of Excellence "Energy-Efficient Product and Process Innovation in Production Engineering" (eniPROD®) is funded by the European Union (European Regional Development Fund) and the Free State of Saxony. 7

REFERENCES

[8]

Huppler, K. R., 2013, Performance Per Watt Benchmarking Ways to Get More for Less, in: Proceedings of the 4th TPC Technology Conference, Istanbul, pp. 60-74.

[9]

Tereci, A.; Ozkan, S. T. E.; Eicker, U., 2013, Energy benchmarking for residential buildings, in: Energy and Buildings, Vol. 60, pp. 92-99.

[10] Andersen, B., 1998, Industrial Benchmarking for Competitive Advantage, in: Proceedings of the ESPRITCOPERNICUS Symposium on Distributed Enterprise, Intelligent Automation and Industrial Benchmarking Putting Information and Communication Technologies into Practice, Wrocław. [11] Reinhart, G.; Karl, F.; Krebs, P.; Reinhardt, S., 2010, Energiewertstrom – Eine Methode zur ganzheitlichen Erhöhung der Energieproduktivität, in: Zeitschrift für wirtschaftlichen Fabrikbetrieb – ZWF, 2010/10, pp. 870875. [12] Reitz, A., 2008, LEAN TPM – in 12 Schritten zum schlanken Managementsystem, MI Wirtschaftsbuch, München. [13] Ohno, T., 1988, Toyota Production System: Beyond Large-Scale Production, Productivity Press, New York. [14] Sterzing, A.; Neugebauer, R.; Drossel, W.-G., 2013, Metal Forming – Challenges from a Green Perspective, in: Proceedings of the International Conference on Competitive Manufacturing, Stellenbosch, pp. 19-24.

[1]

Bogetoft, P., 2012, Performance Springer-Verlag, New York.

[2]

Camp, R. C., 1989, Benchmarking. The search for industry best practices that lead to superior performance, Quality Press; Quality Resources, Milwaukee, Wis, White Plains, N.Y.

[15] Mayer, G.; Pöge, C., 2010: The Road to Standardization: From the Idea to Realization of the VDA Automotive Toolkit, in: Zülch, G. (Ed.): Integration Aspects of Simulation: Equipment, Organization and Personnel, KIT Scientific Publishing, Karlsruhe, pp. 2936.

[3]

Gutowski, T.; Murphy, C.; Allen, D.; Bauer, D.; Bras, B.; Piwonka, T.; Scheng, P.; Sutherland, J.; Thurston, D.; Wolf, E., 2005: Environmentally benign manufacturing: Observations from Japan, Europe and the United States, in: Journal of Cleaner Production, Vol. 13, No. 1, pp. 1-17.

[16] Stoldt, J.; Schlegel, A.; Franz, E.; Langer, T.; Putz, M., 2013, Generic energy-enhancement module for consumption analysis of manufacturing processes in discrete event simulation, in: Proceedings of the 20th CIRP International Conference on Life Cycle Engineering, Singapore, pp.162-170.

[4]

Kristof, Kora; Türk, Volker, 2006, Ressourceneffizienzsteigerungen durch unternehmensübergreifende Instrumente. Status-Quo Analyse, Kritik, Politikempfehlungen, Wuppertal.

[17] VDI, 2011, VDI-Richtlinie 5600: Fertigungsmanagementsysteme (Manufacturing Execution Systems – MES), Beuth Verlag, Berlin.

[5]

Stich, V.; Brandenburg, U.; Kropp, S., 2012, Benchmarking Concept for Energy Efficiency in the Manufacturing Industry – A Holistic Energy Efficiency Model, in: Advances in Production Management Systems. Value Networks: Innovation, Technologies, and Management, pp. 390-395.

[6]

[7]

Benchamrking,

the "Triple Bottom Line of Sustainability" for long-term Success in Manufacturing Industries, in: Proceedings of the 10th Global Conference on Sustainable Manufacturing, Istanbul, pp. 594-601.

Mertins, K.; Kohl, H.; Riebartsch, O., 2011, Sustainable Key-Figure Benchmarking for Small and Medium Sized Enterprises, in: Proceedings of the 9th Global Conference on Sustainable Manufacturing, St. Petersburg, pp. 309-314. Mertins, K.; Kohl, H.; Riebartsch, O., 2012, Benchmarking: The ideal Management Tool to exploit

[18] Neugebauer, R.; Kreppenhofer, D.; Langer, T., 2010, Determination of the overall equipment effectiveness for assembly systems on the base of product data, in: 43 rd CIRP International Conference on Manufacturing Systems, Vienna, pp. 991-999. [19] Putz, M., 2012, Determination of extended availability and productivity statements for assembly systems using existing data base, in: Annals of the CIRP, Vol. 61, No. 1, pp. 17-20. [20] Putz, M.; Schreiber, A.; Firlej, A.; Langer, T., 2013, Linked Factory – Datenintegration zur mobilen Informationsbereitstellung in der Produktion, in: Proceedings of the 19th Industriearbeitskreis “Kooperation im Anlagenbau”, Magdeburg.

478