Decision Support for Choosing an Appropriate Simulation Method for ...

Journal of Industrial and Intelligent Information Vol. 3, No. 4, December 2015

Decision Support for Choosing an Appropriate Simulation Method for Dynamic Material Flow Analysis Markus P. Roessler, Joshua Riemer, and Maximilian Mueller Institute of Production Management, Technology and Machine Tools (PTW), Technische Universitaet Darmstadt, 64287 Darmstadt, Germany Email: [email protected]; [email protected]; [email protected]

with minimal stocks, maximal productivity and thus ideally used resources [6]. Using simulations the mentioned parameters as well as highly complex processes can be analyzed and optimized during the whole life cycle from planning to operation [7]. Currently in this context the method of discrete event simulation (DES) is widespread for the simulation and analysis of material flows [8]–[10]. Further the methods of system dynamics (SD) and agent based simulation (ABS) allow the simulation of material flows [4], [11]. In practice none of these methods can be seen as generally superior [12]. In scientific research simulation methods are also compared with each other, but mostly based on the evaluation and experiences of the authors and a focus on distinct areas of application [13], [14]. A decision support that enables a comparison between the three simulation methods based on a predefined list of selection criteria, however, does not exist [14], [15].

Abstract—When improving industrial process chains in terms of productivity it is becoming more and more important to not only consider the static, but also the dynamic behavior of value streams. The effects of intended changes on production systems often are unknown and could underlie vast variations due to the planned outcome. In order to validate such changes before implementation, the use of material flow simulation is widely discussed in research and industrial practice. Such optimization projects usually require the collaboration of production system and simulation experts. Through the increased dissemination of various simulation tools on the market as well as multiple simulation methods available (discrete event, agent based and system dynamics) the actual selection of a suitable simulation setting is a challenging task. To support such optimization teams in the selection of an appropriate simulation method a decision making approach based on comparative criteria is shown and discussed in this article, concluding with a case study.  Index Terms—material flow simulation, production system, decision support, discrete event simulation, agent based simulation, system dynamics

I.

II.

Various authors suggest the use of simulation in value stream optimization projects, where improvements of the material flows of underlying production systems are modeled and the effects analyzed to validate changes before physical implementation [16]–[24]. Usually this is done with systems, which are characterized by a high variability, a high interconnectedness, and a high combinatorial and dynamic complexity. Issues in other systems mostly could be solved using analytical methods. Further information about when simulations are (not) appropriate is available in [25] and [26]. In the next paragraphs the most frequently used simulation methods for material flow simulation will be shortly introduced and discussed.

INTRODUCTION

Due to a constant intensification of international competition through globalization of markets, producing companies are forced to increase their productivity while lowering their costs and improving quality [1]. At the same time producing companies need to react quickly to the customers’ demand. In practice this can be achieved through short production lead times [2]. An optimization of material flow systems in order to lower lead times thus is of crucial importance for sustaining a high performance of production systems. This also includes the reduction of inventories as well as the increase in service levels and utilization. [3], [4] Because of the increasing complexity of material flow systems analytical approaches reach their limits. Instead simulations have to be used. [1], [5] Hence, the prior quantification of possible production system designs, using material flow simulation, is gaining more and more importance. The general objective is to develop systems

A. Discrete Event Simulation In discrete event simulation the dynamic and stochastic behavior of systems is modeled by time variant state variables. Events occur at a limited number of instants of time [27]. Typical events in production systems are the finishing of a task or the arrival of an order at a work station of the production system [10]. Discrete event simulation has long been the most common simulation

Manuscript received December 26, 2014; revised April 28, 2015. © 2015 Journal of Industrial and Intelligent Information doi: 10.12720/jiii.3.4.337-341

STATE OF THE ART IN MATERIAL FLOW SIMULATION

337

Journal of Industrial and Intelligent Information Vol. 3, No. 4, December 2015

technology in the sector of operational research. This also holds true for the special case of material flow simulation. [28] In the context of discrete event simulation a distinction can be made between synchronous and asynchronous simulation. In the asynchronous case, events may occur at any time and will be simulated directly. In the synchronous case, the simulation progresses always after a determined interval. Events, which occur in the meanwhile, will be taken into consideration in the next interval. In industrial applications, asynchronous discrete event simulations are widely used [29]. DES requires less computer resources than continuous simulation methods inasmuch as it can jump temporally from one event to another and has to merely simulate these events. Due to the fast run time, this technology is particularly well suited to the simulation of processes with statistical discrepancy uncertainties. Through a repeated execution of the simulation, statements can be made regarding possible extreme conditions and common scenarios. The discrete event simulation is thereby especially suitable for queuing problems and complex queuing systems [28]. A further advantage is that this method is widespread and so a multitude of highly developed software systems for different applications is available. B. System Dynamic Simulation The method of system dynamic simulation was mentioned for the first time by [30] in the years 1958 and 1961. Afterwards, the method was defined as follows [31]: "Industrial dynamics is the study of the informationfeedback characteristics of industrial activity to show how organizational structure, amplification (in policies), and time delays (in decisions and actions) interact to influence the success of the enterprise.” At this time, the term industrial dynamics was used; today the expression system dynamics is used. The reason for the name change was the expansion of potential applications, which are no longer limited to industrial areas, but also used in the societal area. [32] In the first application of SD a fourstage supply chain was modeled and analyzed [30]. System dynamic simulation uses mathematical models based on differential equations [33], [34]. This approach, however, can also be described as object-oriented [35]. A system dynamic model is composed of a closed system that includes all the required elements and functions. The dynamic behavior of systems is at the center. This is accomplished by buffers having mutually different inand outflows. This leads to accumulations, delays and inertia. [36] Because the components of the system are interdependent, the variables affect each other over time. This leads to a dynamic behavior of the systems, which can be described using feedback loops. [37] The system can thereby selectively be influenced by decisions. An analysis of these decisions and the resulting behavior of the system is thus of great interest [36]. SD is based on macro-modeling; this does not mean that only highly aggregated studies take place. It is dispensed to model single objects, instead a mapping of their flows is performed. [36] Unlike in discrete event © 2015 Journal of Industrial and Intelligent Information

338

applications these variables are represented as continuous parameters [38]. For an analysis of continuous material flows, such as in process industry, the SD approach is suited best [39]. The differential equations used hereby take into account all relevant physical relationships [34]. In order to develop a SD model, it is necessary to identify the causal relations of the organizational behavior [38]. The relations can be represented using a causal graph in a first step. Eventually a system of differential equations is derived. [37] C. Agent-Based Simulation The agent based simulation is built on the principles of the agent technology. The term agent based simulation refers to the modeling and simulation of realistic systems with the help of agents, which interact with each other among a simulation model. [40] This definition makes it quite clear that the main advantage of the agent based simulation lies in the interaction of different agents. Nowadays processes and structures are more and more complex. As a result, an ever deeper specialization of actors takes place to overcome problems. The agent based simulation can be used to simulate such processes [41]. The technology can be understood as an enhancement of object oriented methods and discrete event simulation (conclusion of [42]; based on [43]). Time progresses discontinuously from an event to another. These events are no longer central, but are controlled by agents. For a long time, agent based simulations were exclusively used in a socio-scientific context, such as for the spreading of epidemics or the fall of antique civilizations [42]. In those fields, in which the interaction of a large number of entities is very important, the advantages of the agent based approach are clearly evident. Beyond these application fields, agent based simulations were unknown. This was due to the relatively little simulation software that supported the agent based approach. This software was technologically very demanding and concentrated on specific fields of application, which hindered a faster spread of the agent technology. In contrast to some of the properties of agents like autonomy, social skills, reactivity, and proactivity [44], complex and cognitive properties are usually absent from simulated agents [40]. A definite advantage of an agent based simulation is that the original system must be rarely abstracted in order to reach a simulation model [40]. That means that the actual modeling is much simpler compared to other methods [42]. This is obtained by using a modular setup, which is very effective especially for complex systems. This advantage is particularly relevant in distributed systems [41], in which the complex structures arise from the interaction between different entities [45]. An additional benefit of agent based simulation is found in the higher flexibility compared to the discrete event simulation. On the one hand, the structure of an agent simulation is very modular which makes a possible extension towards additional agents very easy. On the other hand, these agents could be influenced during the

Journal of Industrial and Intelligent Information Vol. 3, No. 4, December 2015

simulation runtime. For example, it is possible whether to create or remove an additional agent (such as in case of a growing population) during runtime. Although today’s computers are constantly upgraded, they still reach their limits in very extensive agent simulations. [5] An additional simplification of complex systems or a distributed execution using various computers is therefore necessary to obtain an effective simulation. Due to the modular structure, this distributed execution is supported. [34]. III.

Criterion

Task Technology

A. Methodology and Study Design The purpose of the further sections is to show a decision making approach for the selection of an appropriate simulation method for material flow simulation projects. Fig. 1 visualizes the focus of this work, embedded in a procedure model for an industrial project.

Selection of a suitable simulation software

Information

Phase

planning (1), realisation (2), operation (3)

Scope

operative (1), strategic (3)

Level of detail

low (1), high (3)

Time resolution

operative (1), tactical (2), strategic (3)

Data demand

low (1), high (3)

Real time ability

yes (1), no (3)

M odeling effort

low (1), high (3)

Effect of failure to stability

low (1), high (3)

Flex. during simulation

yes (1), no (3)

M odeling approach

top-down (1), bottom-up (3)

Accuracy

structural (1), behavioral (2), functional (3)

Computing power

low (1), middle (2), high (3)

Transferability to reality

simple (1), complex (3)

Validation

Degree of automation

simple (1), complex (3) component/individual (1), station/plant (2), corporation/network (3) low (1), high (3)

Delegation

central (1), peripheral (3)

System

Human-machine-interaction low (1), high (3) Coordination (type)

people-oriented (1), technique-oriented (3)

System focus

material flow (1), information flow (3)

Production principle

flow shop (1), job shop (3)

M aterial flow type

continuous (1), discrete (3)

M icro behaviour crucial

yes (1), no (3)

Stochastical environment

yes (1), no (3)

Figure 2. Criteria, characteristics and classification of the three simulation methods.

Modeling, simulation and analysis of material flows

IV.

Concrete actions

Volatile production environment

CASE STUDY

A. Project Definition and Method Selection In this paragraph a case study is introduced to test the decision support tool and its ability to suggest an appropriate simulation method. The case study is conducted at the Center for industrial Productivity at the Technische Universitaet Darmstadt, which represents a real production environment for teaching and consulting purposes [47].

Figure 1. Procedure model for a material flow simulation project and focus of this paper.

B. Criteria Selection and Tool Composition To develop a decision support tool in the first phase the identification and classification of relevant decision criteria is performed [46]. In whole 25 criteria could be identified out of the field of research in simulation, material flow research and practice using literature as well as expert interviews. The criteria could be clustered into three project-relevant groups (task, technology and system) and are shown in Fig. 2. The task group describes the objective and the requirements to the simulation project. The technology group clusters criteria regarding technological requirements of the simulation method as for the hardware during modeling and simulation. The system group describes the material flow system which has to be modeled. Characteristics of all identified criteria could be assigned to the three simulation methods. As can be seen in Fig. 1, in the first phase the person responsible for the material flow simulation project has to characterize his project using the 25 criteria. Hereby multiple characterizations for each criterion are possible and also recommendable. After this characterization he will receive a suggestion out of the developed support tool which simulation methods are suitable. Regarding the criteria there is no claim of completeness for representing the field of material flow simulation; the criteria listed are collected for the purpose of selecting a suitable simulation method only.

© 2015 Journal of Industrial and Intelligent Information

ABS

M odularity during modeling yes (1), no (3)

Level of abstraction

Focus of this paper Tool-based selection of an appropriate simulation method

SD

1 2 3 1 2 3 1 2 3

DEVELOPMENT OF A DECISION SUPPORT TOOL

Definition and characterization of material flow simulation project

DES

Characteristics

14-days forecast

weekly order

Supplier delivery to inventory assembly

Production planning and scheduling

31 pcs daily order

Turning Raw material storage

20 pcs

CT=20 s COT=302 s AV=85 % LS=24 pcs

Inventory turning

DPS=48 pcs CTact=60 s

Outgoing storage

24 pcs CT=40 s COT=309 s AV=85 % LS=24 pcs

Cleaning FIFO

Milling FIFO

Inventory milling

40 pcs

Customer

daily order

weekly order

1 009 pcs

Sawing

90-days forecast

CT=51 s COT=330 s AV=85 % LS=192 pcs

FIFO

CT=20 s COT=0 s AV=100 % LS=4 pcs

Quality control

CT=36 s COT=0 s AV=100 % LS=1 pc

Inventory assembly

540 pcs

Assembly

Final testing

Packing / shipment

CT=55 s COT=0 s AV=100 % LS=3 pcs

CT=32 s COT=15 s AV=100 % LS=3 pcs

CT=14 s COT=0 s AV=100 % LS=3 pcs

10 pcs 181 pcs

Figure 3. Current state map of the production system [23].

The production system produces pneumatic cylinders for various industrial applications in four variants. The plant delivers two times per day to the customer and produces in three shifts per day, each lasting 48 minutes without breaks. The value stream of the production system has to be improved regarding lead time, inventory, flexibility and productivity. Fig. 3 shows a value stream map of the system to be modeled, simulated and improved (CT-cycle time; COT-changeover time; AV-

339

Journal of Industrial and Intelligent Information Vol. 3, No. 4, December 2015

Technology

System

availability; LS-lot size; DPS-demand per shift; CTaktcustomer takt time). The project team for optimizing the material flow in this case consists of three persons, one expert in methods of holistic production systems responsible for the improvement project, a simulation specialist and a programmer. The results of their tool-based evaluation are summarized in Fig. 4.

case study is the production principle. The production system consists of flow shop, as well as job shop elements. Flow shop elements are much easier to implement in practice using the discrete event approach. The complexity resulting out of a mere job shop system can be managed better using an agent based approach. In a similar direction the aspect of human-machineinteraction points, which is an indicator for the amount of local decisions to be made and so the control complexity of the system. In agent based systems control mechanisms are easy to implement into every modeled entity, which supports communication and transparency between the different agents (workers as well as equipment). In the case study the production system contains a variety of characteristics regarding this issue, e. g. fully automated sections, partially automated sections with much interaction as well as mere manual assembly tasks. These four factors (material flow type, flexibility during modeling, production principle and humancomputer-interaction) identified out of the case study are very important especially for material flow projects and the efforts invested in modeling. Regarding the developed decision support tool, further these four criteria are weighted (double), which leads to the following ranking with the ABS approach as the primary recommendation, see Fig. 6.

ABS SD

DES ABS SD

DES

Task

ABS SD DES 0

1

2

3

4 5 6 Sum of compliances

7

8

9

Figure 4. Results of the tool-based evaluation regarding the ability of the three simulation methods to fulfil the requirements to the simulation project.

According to the tool the methods of DES and ABS turn out as being suitable. Each has in whole about 17 compliances between the project characterization and the criteria profile of the three methods from Fig. 2.

Technology

System

B. Simulation and Assessment In the next step two simulation models are implemented by the programmer using identical data out of the value stream analysis and of further system observations regarding variability, personnel and transport resources, distances, work plans, bills of material, etc. For visualization see Fig. 5.

ABS SD

DES ABS SD

DES

Task

ABS SD DES 0

2

4 6 Sum of compliances

8

10

Figure 6. Weighted results of the tool-based evaluation.

V. Figure 5. Exemplarily visualization of DES and ABS models using the software Anylogic 6.

The selection of a suitable simulation method for material flow simulation projects is a challenging task. This article summarizes the three mostly applied simulation techniques in industrial practice when it comes to modeling and analysis of production systems and their material flows. Based on a literature review a decision support tool could be developed which includes 25 criteria for choosing a simulation technique. These could be enriched through practical experiences out of a case study in terms of an appropriate weighting. The decision support tool leads to a suggestion for the most suitable simulation technique. In industrial practice however most production systems consist of very heterogeneous characteristics and subsystems. So the possibility of using different simulation techniques in one simulation model could be focused further in detail. To address the validity of the developed decision support tool a higher number of case studies is recommendable which will be performed also in future.

The method of system dynamics is not used in this case because of the relatively low rating and the continuous material flow type addressed. This is a very important factor, which, according to an expert interview with a simulation specialist is the determining factor for selecting the SD approach. Both implemented simulation models (DES and ABS) are suitable for analyzing the defined indicators of the project, nevertheless after analyzing the results there are still some major differences regarding the implementation efforts. The first issue is the flexibility of the developed models. The agent based approach is more flexible regarding the expansion in terms of adding workers, processes or resources during modeling and also during runtime of the simulation, see also [34]. This results from the modular composition and the bottom-up characteristics of the agent based approach. Another important issue during the © 2015 Journal of Industrial and Intelligent Information

CONCLUSIONS

340

Journal of Industrial and Intelligent Information Vol. 3, No. 4, December 2015

REFERENCES [1]

[2] [3] [4]

[5] [6]

[7]

[8]

[9]

[10]

[11] [12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23] [24] [25]

A. Fink, N. Kliewer, D. Mattfeld, L. Moench, et al., “Model-based decision support in manufacturing service networks,” Business and Information Systems Engineering, vol. 6, pp. 17, 2014. E. Abele and G. Reinhart, Future of Production, Munich: Carl Hanser Verlag, 2011. U. Clausen and C. Geiger, Traffic- and Transport Logistics, 2nd ed., Berlin: Springer-Verlag, 2013. A. Negahban and J. S. Smith, “Simulation for manufacturing system design and operation,” Journal of Manufacturing Systems, vol. 33, pp. 241, 2014. D. Huber, “Controlled simplification of hierarchical partitions of models in material flow simulations,” Diss., Paderborn, 2009. M. Caridi and S. Cavalieri, “Multi-agent systems in production planning and control,” Production Planning & Control, vol. 15, pp. 106, 2004. Verein Deutscher Ingenieure, Simulation of Systems in Materials Handling, Logistics and Production (Fundamentals): VDI 3633 Part 1, 2010. R. Al-Aomar, “Handling multi-lean measures with simulation and simulated annealing,” Journal of the Franklin Institute, vol. 348, p. 1506, 2011. D. Daniluk and R. Chisu, “Simulation and emulation in the internet of things,” in Internet of Things in Intralogistics, W. Guenther and M. ten Hompel, Eds., Berlin: Springer-Verlag, 2010, pp. 147–169. L. Maerz, H. Krug, O. Rose, and G. Weigert, Simulation and Optimization of Production and Logistics, Berlin: Springer-Verlag, 2011. M. Mueller, “Agent-based simulation of material flows as an alternative to classical methods,” Thesis, Darmstadt, 2013. J. D. Morecroft and S. Robinson, “Explaining puzzling dynamics,” in Proc. 23th International Conference of the System Dynamics Society, Boston, USA, July 17-21, 2005, pp. 17. A. A. Tako and S. Robinson, “Model development in discreteevent simulation and system dynamics,” European Journal of Operational Research, vol. 207, pp. 784, 2010. J. Riemer, “Development of a user-oriented decision-making instrument for choosing a suitable simulation method for dynamic analysis of material flows,” Thesis, Darmstadt, 2014. T. Lorenz and A. Jost, “Towards an orientation framework in multi-paradigm modeling,” in Proc. 24th International Conference of the System Dynamics Society, Nijmegen, Netherlands, July 23-27, 2006. F. A. Abdulmalek and J. Rajgopal, “Analyzing the benefits of lean manufacturing and value stream mapping via simulation,” Journal of Production Economics, vol. 107, pp. 223, 2007. A. Azadeh, M. Seifoory, and M. Abbasi, “Integration of simulation and fuzzy multi-attribute decision making for modelling and assessment of fuzzy parameters,” Journal of Industrial and Systems Engineering, vol. 6, pp. 483, 2007. M. Boerkircher and T. Gamber, “Simulation based value stream design - Approach for increasing value added potential in building materials industry,” in Integration Aspects of Simulation, G. Zuelch and P. Stock, Eds., Karlsruhe: KIT, 2010, pp. 405–412. H. Brueggemann and P. Mueller, “Digital value stream mapping,” in Advances in Simulation for Production and Logistics Applications, M. Rabe, Ed., Stuttgart: Fraunhofer IRB Verlag, 2008, pp. 575–584. R. B. Detty and J. C. Yingling, “Quantifying benefits of conversion to lean manufacturing with discrete event simulation,” Journal of Production Research, vol. 38, pp. 429, 2000. Y. H. Lian and H. van Landeghem, “An application of simulation and value stream mapping in lean manufacturing,” in Proc. 14th European Simulation Symposium, A. Verbraeck and W. Krug, Eds., SCS Europe Bvba and SCS International, 2002. T. McDonald, E. M. van Aken, and A. F. Rentes, “Utilising simulation to enhance value stream mapping,” Journal of Logistics Research and Applications, vol. 5, pp. 213, 2002. M. P. Roessler, J. Metternich, and E. Abele, “Learning to see clear,” Business and Management Research, vol. 3, pp. 93, 2014. C. R. Standridge and J. H. Marvel, “Why lean needs simulation,” in Proc. Winter Simulation Conference, 2006, pp. 1907–1913. J. Banks, J. S. Carson, B. L. Nelson, and D. M. Nicol, Discrete-

© 2015 Journal of Industrial and Intelligent Information

341

Event System Simulation, New Jersey: Pearson Education, 2010. [26] S. Robinson, Simulation: The Practice of Model Development and Use, Chichester: John Wiley & Sons Ltd, 2004. [27] W. Domschke and A. Drexel, Introduction to Operations Research, 7th edn., Berlin: Springer-Verlag, 2007. [28] P. Siebers, C. M. Macal, J. Garnett, D. Buxton, and M. Pidd, “Discrete-event simulation is dead, long live agent-based simulation,” Journal of Simulation, vol. 4, pp. 204, 2010. [29] R. Herrler, “Agent based simulation of processes in hospitals and other distributed, dynamic environments,” Diss., Wuerzburg, 2007. [30] J. W. Forrester, “Industrial dynamics,” Harvard Business Review, vol. 36, pp. 37, 1958. [31] J. W. Forrester, Industrial Dynamics, Waltham: Pegasus Communications, 1961. [32] T. S. Baines, D. K. Harrison, J. M. Kay, and D. J. Hamblin, “A consideration of modelling techniques that can be used to evaluate manufacturing strategies,” International Journal of Advanced Manufacturing Technology, vol. 14, pp. 369, 1998. [33] A. Deckert and R. Klein, “Agent-based simulation for analysis and solution of business related decision problems,” Journal Fuer Betriebswirtschaft, vol. 60, pp. 89, 2010. [34] H. Parunak, R. Savit, and R. L. Riolo, “Agent-based modeling vs equation-based modeling,” Lecture Notes in Computer Science, vol. 1534, pp. 10, 1998. [35] H. Pierreval, R. Bruniaux, and C. Caux, “A continuous simulation approach for supply chains in the automotive industry,” International Journal of the Federation of European Simulation Societies, vol. 15, pp. 185, 2007. [36] N. Schieritz and P. M. Milling, “Modeling the forest or modeling the trees,” in Proc. 21st Conference of the System Dynamics Society, New York, USA, July 20-24, 2003. [37] L. Rabelo, M. Helal, A. Jones, and H. S. Min, “Enterprise simulation,” in Proc. Winter Simulation Conference, New Orleans, USA, December 07-10, 2003, pp. 1225. [38] S. C. Brailsford, S .M. Desai, and J. Viana, “Towards the holy grail,” in Proc. Winter Simulation Conference, Baltimore, USA, December 05-08, 2010, pp. 2293. [39] A. Borshchev and A. Filippov, “From system dynamics and discrete event to practical agent based modeling,” in Proc. 22nd International Conference of the System Dynamics Society, Oxford, England, July 25-29, 2004. [40] D. Pawlaszczyk, “Scalable agent-based simulation,” Diss., Ilmenau, 2009. [41] F. Kluegl, Multi-Agent-Simulation: Concepts, Tools, Application, 1st edn., Addison-Wesley, 2001. [42] P. Davidsson, “Multi agent based simulation,” Lecture Notes in Computer Science, vol. 1979, pp. 97, 2001. [43] J. Misra, “Distributed discrete-event simulation,” Computing Surveys, vol. 18, pp. 39, 1986. [44] M. Wooldridge, “Intelligent agents,” in Multiagent systems: A modern approach to distributed artificial intelligence, G. Weiss, Ed, 1999, pp. 25–73. [45] J. Ferber, Multi-Agent Systems: An Introduction to Distributed Artificial Intelligence, Addison-Wesley, 1999. [46] V. Belton and T. Stewart, Multiple Criteria Decision Making: An Integrated Approach, Boston: Kluwer Academic Publishers, 2002. [47] E. Abele, N. Eichhorn, and S. Kuhn, Increase of Productivity based on Capability Building in a Learning Factory, Biograd, Croatia, 2007. Markus Philipp Roessler, born 1986 in Kuenzelsau, Germany, studied Production and Logistics Management at Heilbronn University and Business Engineering at the Karlsruhe Institute of Technology, both in Germany. Since 2011 he works as a research associate for the Institute of Production Management, Technology and Machine Tools (PTW) at Technische Universitaet Darmstadt, where he is responsible for research on approaches and enhancements of value stream mapping techniques as well as their application in various industrial projects. Supplementary he is working as training instructor for specialist and executive courses on topics of holistic production systems.