Print Email Facebook Twitter Strategic Supply Chain Decisions through Discrete Event Simulation. Development of DrESSeD Framework: An alignment of supplier selection criteria with customers’ preferences Title Strategic Supply Chain Decisions through Discrete Event Simulation. Development of DrESSeD Framework: An alignment of supplier selection criteria with customers’ preferences Author Kokkotas, G. Contributor Verbraeck, A. (mentor) Kortmann, L.J. (mentor) Cunningham, S. (mentor) Wenzler, I. (mentor) Faculty Technology, Policy and Management Department Technology, Policy, and Management Programme Engineering and Policy Analysis/Systems Engineering Section Date 2014-01-30 Abstract This research is based on the development of a supply chain method that could help firms/organizations to take strategic decisions. DrESSeD framework is a simulation-based framework for strategic decision making support. The proposed framework tries to directly relate supplier selection criteria, customers’ preferences and manufacturer’s performance. The acronym DrESSeD stands for (D)isc(r)ete (E)vent (S)imulation for (S)trat(e)gic (D)ecisions. The need for more coordination and control on supply chain activities, processes, and uncertainties within and between organisations caused the creation of the concept of Supply Chain Management (SCM). Strategic supply chain management can provide companies with the appropriate strategic knowledge in order to obtain competitive advantage and improve their performance (Hult et.al., 2007). Companies make high level strategic supply chain decisions considering issues in the fields of product development, customers, manufacturing, and suppliers (Murray, 2012). Simulation approaches have been used as decision support systems for supply chain networks. Two most popular simulation approaches are systems dynamics (SD) and discrete event simulation (DES) (Tako & Robinson, 2012). Although, strategic issues are commonly modeled by SD, only a few models exist for solving strategic issues using discrete event simulation. Supplier selection simulation is an unexplored area in the field of supply chain networks (Tako & Robinson, 2012). Although, strategic issues are commonly modeled by SD, few models stand for solving strategic issues using discrete event simulation (Tako & Robinson, 2012). This finding creates a need for exploring the area of supplier selection processes using discrete event simulation. In addition, companies do care about the reaction of customers in the market. Moreover, strategic supply chain decisions are related with issues in suppliers’ and customers’ tier (Murray, 2012). Finally, it can be useful for companies to take into consideration customers’ preferences for the decisions in their supplier selection processes. As a conclusion, the use of discrete event simulation for studying supply chain strategic issues, focusing on supplier selection processes and customers’ preferences, creates the knowledge gap of this research. This master thesis report is expected to scientifically contribute by answering the question: How can we support strategic decision making of firms/organizations in the dynamic supplier selection process, using discrete event simulation, taking into consideration specific selection criteria, customer’s preferences and market characteristics? Considering the research question, there is need for an appropriate method that can explore supply chain strategic issues, using discrete event simulation in order to support companies in strategic decision making. A simulation based framework is the method that used in this report. Framework is a conceptual structure, intended to serve as a support tool. In other words, it creates something more generic, which it can be expanded in the future. Existing supplier selection frameworks have not considered customers’ needs and/or simulation approaches (Ng, 2008), (Jahani et.al, 2011), (Vijayvagy, 2012). With the increased competition in global supply chain networks, the significance of vendor selection based on supply chain risks is high (Liu et.al, 2011). There is need for a new framework, which has simple structure, can be easily used in different environments, and can bring together firm’s performance, suppliers’ performance and customers’ preferences. DrESSeD framework is a new proposal in the field of supplier selection using discrete event simulation. The framework consists of three main parts. Part I helps the end user to apply the Design Code Method in order to analyse the performance/behavior of suppliers. The Design Code is a method that based on Boolean theory and uses different attributes of suppliers for the determination of suppliers’ performance. Part II is proposing to companies how to conduct an analysis, regarding suppliers’ selection criteria and customers’ preferences. Both the supplier selection criteria and customer’s choices will be studied by discrete choice theory. Part III teaches the user how to use a simulation model in order to calculate market share. The generic discrete event simulation model is showing the basic processes of the supply chain network and includes two main functions: the supplier selection processes, and the customer choice process. The supply chain network consists of two major suppliers, one manufacturer and N customers. This is the case for the moment, used to validate the framework. In future, this number may be extended to model any number of competing suppliers. Customers’ choice are determined by cost, time, quality, and innovation (Cohen & Roussel 2005). Based on these four decision variables, customers make their final choice for purchasing a product/service. Considering the appropriate evaluation factors for supplier selection, the author selected four factors (Strategic, Operational, Financial, and Risk) that proposed in Accenture’s supplier selection report (Accenture, 2011). DrESSeD framework was evaluated through a case study. From practical perspective, case studies help companies and organizations evaluate the advantages of methods and tools. The case study for this report focuses on the Dutch smart phone market. Since the needed supplied evaluation factors could not be found in scientific sources or Accenture’s reports, the information from an empirical study presenting generic evaluation factors for supplier selection ( Verma & Pullman, 1998) was used instead. On the other hand, the author conducted a statistical research in order to obtain responses about customers’ preferences. The results of the statistical research are collected through a questionnaire. Framework’s evaluation through a case study wanted to show Apart from framework’s evaluation through a case study, DrESSeD framework was evaluated in order to identify the scientific added value that contributes to the research field and the business added value that contributes in Accenture. Therefore, the evaluation process of the framework was based on interviews with supply chain professionals both from academia and industry. Regarding the aspect of adding value in a business context, the author evaluated DrESSeD framework using 3U framework (Verbraeck, 2010). In the 3Uframework the main evaluation factors are: Usability: Alignment of the application to the user Usefulness: Alignment of the application to the task Usage: Alignment of the application to the organizational goals Professionals believe that the framework is not usable by companies’ employees. However, all the experts agreed that the framework could be very useful in the supplier selection task. In terms of usage, the framework cannot satisfy organizational purposes. Accenture is the main client of this particular research. Accenture’s clients are potential users of DrESSeD framework. The author proposes a plan that Accenture can use in order to overcome usability and usage shortcomings of DrESSeD and apply it on its projects. This plan is based on Accenture’s report about supplier selection evaluation factors (Accenture, 2011) and on its access to data within organisations. DrESSeD framework has two main advantages. The first is related with framework’s structure. DrESSeD framework has simple structure and development; characteristic that makes it easy to be adapted by other frameworks and/or applied in different case studies. The second positive aspect of the framework presents the contribution of the study. DrESSeD framework brings under one umbrella, a simulation approach (DES), a statistical analysis (discrete choice analysis), and the observation of firm’s and suppliers’ performance. On the other hand, the two major problems of DrESSeD framework are low validity and low applicability. Giving some details about framework’s development, there are two important aspects (one positive and one negative) that have to be mentioned: From the positive point of view, the author used an econometric model called Multinomial Logit (MNL) in order to conduct the discrete choice analysis. MNL model uses a maximum likelihood estimation scheme to maximize the probability of choosing an alternative with given attribute levels (Verma & Pullman, 1998). MNL model used properly in DrESSeD framework, as it presented perfectly the attribute levels of suppliers’ profiles. From the negative point of view, DrESSeD framework used four specific evaluation factors for supplier selection and four decision variables as purchasing criteria. This choice was not effective, because it increased the complexity of the model. Supplier selection has to be based on three factors: Cost, Product Leadership, and Service (van Houten, 2013). DrESSeD framework succeeded to make a difference in the field of supplier selection process. The use of discrete event simulation for facing the supply chain issue of supplier selection, it is not a new idea. But, the implementation of supplier profiles, which indicate supplier’s performance, and the interaction of supplier with customer preferences under the umbrella of DES, is a new approach for supplier selection. That difference can benefit Accenture’s clients, because it can provide any company with multi-dimension results (decision about appropriate supplier, customer choices, market share value) without big costs (simulation model). Unfortunately, based on evaluation (Section 4.5) and simulation (Section 4.6) results, DrESSeD framework is characterized by low validity. As a result, the framework does not work for Accenture and it cannot be applied in current projects. Subject Discrete Event SimulationStrategic DecisionSupplier SelectionCustomer Preference To reference this document use: http://resolver.tudelft.nl/uuid:eed2c6d1-76a6-4540-80c6-4678df16ca75 Embargo date 2015-02-02 Access restriction Campus only Part of collection Student theses Document type master thesis Rights (c) 2014 Kokkotas, G.