FCM یکپارچه و مجموعه نرم فازی برای مسئله انتخاب تامین کننده بر اساس ارزیابی ریسک

Supplier selection problem, considered as a multi-criteria decision-making (MCDM) problem, is one of the most important issues for firms. Lots of literatures about it have been emitted since 1960s. However, research on supplier selection under operational risks is limited. What’s more, the criteria used by most of them are independent, which usually does not correspond with the real world. Although the analytic network process (ANP) has been proposed to deal with the problems above, several problems make the method impractical. This study first integrates the fuzzy cognitive map (FCM) and fuzzy soft set model for solving the supplier selection problem. This method not only considers the dependent and feedback effect among criteria, but also considers the uncertainties on decision making process. Finally, a case study of supplier selection considering risk factors is given to demonstrate the proposed method’s effectiveness.

Supply chains have become major elements in the global economy. Nowadays, the competition among enterprises has evolved into the competition among the supply chains. But supply chains expose to different kinds of risks that increase along with increasing globalization. Therefore, supply chains risk management (SCRM) is a field of escalating importance and is aimed at developing approaches to the identification, assessment, analysis and treatment of areas of vulnerability and risk in SC [1]. The research on supplier risk is one of important areas of supply chains risk. An effective supplier assessment and selection process is essential for improving the performance of a focal company and its supply chains [2]. So the supplier selection problem taking risk evaluation into consideration is greatly meaningful. Although the evaluation of supplier risk has begun to draw considerable attention, research on supplier selection under operational risks is limited. Huang and Chen [3] discussed a possible Risk Breakdown Structure (RBS) for virtual enterprises and suggested a risk evaluation method. Then they applied the proposed risk evaluation method to the partner selection problem. Wu and Olson [4] considered three types of vendor selection methodologies in supply chains with risk: chance constrained programming (CCP), data envelopment analysis (DEA) and multi-objective programming (MOP) models. Wu et al. [5] proposed a fuzzy multi-objective programming model to decide on supplier selection taking risk factors into consideration. Various models are available to select supply chain partners. In the existing research on supplier selection models, the criteria used by most of them are independent, but the dependent and feedback effects are often overlook. The expert system developed by Vokurka et al. [6] captured the previous supplier selection process in a knowledge base, which can be used to suggest selection criterion for future supplier selection process. Kokangul and Susuz [7] applied an integration of analytical hierarchy process and non-linear integer and multi-objective programming under some constraints to determine the best suppliers. Cakravastia et al. [8] developed an analytical model of mixed-integer programming for the supplier selection process in designing a supply chain network. Choy et al. [9] described a knowledge-based supplier selection and evaluation system, which was a case-based reasoning decision support system for outsourcing operations at Honeywell Consumer Products (Hong Kong) limited in China. Bevilacqua et al. [10] proposed a fuzzy quality function deployment (QFD) approach to support supplier selection. Chen et al. [11] presented a fuzzy model to determine the ranking order of all suppliers according to the concept of the TOPSIS. Seydel [12] used DEA to tackle the supplier selection problem. Ha and Krishnan [13] applied an integrated approach in an auto parts manufacturing company for supplier selection. Based on the assumption of preferential independence, above the methods can be seen that the dependence and the feedback effects cannot be considered. However, the real-life situation usually emerges the dependence and the feedback effects simultaneously while making decision. Agarwal and Shankar [14] proposed an analytic network process (ANP) to evaluate alternatives that provided the route of performance improvement in supply chain. Gencer and Gurpinar [15] proposed an ANP model to tackle the supplier selection problem. Although the interrelationships among criteria were considered in the selection process, two main problems should be highlighted as follows [16]. The first is the problem of comparison. Sometimes it is hard for experts to compare the important degree of an index to another. Furthermore, the key for the ANP is to determine the relationship structure among features in advance [17]. The different structure results in the different priorities. However, it is usually hard for the decision maker to give the true relationship structure by considering many criteria. As we know, due to the problem with compound and interaction effects, it is hard for decision makers to make a good decision using the simple weighted method. In order to deal with the problem, we use FCM to find the weights of criteria, which not only can overcome the preferential independent and but also can overcome the shortcomings of ANP. Moreover, in practice decision-making on supplier selection problem includes a high degree of fuzziness and uncertainties. Molodtsove [18] initiated the concept of soft set theory, which is a new mathematical tool for dealing with uncertainties. Fuzzy soft set has rich potential for applications in several directions. In the present paper, we firstly apply fuzzy soft sets in supply chains. This paper attempts to develop a novel evaluation framework to select supplier considering risk. We first integrate FCM and fuzzy soft set for solving supplier selection problem. The structure of integrated method is shown in Fig. 2. First, the weights of criteria/attributes can be effectively captured by FCM, which not only considers the dependence and the feedback effects among criteria but also can overcome the shortcomings of ANP. Second, in order to compensate for FCM method’s dependence for expert advice in the reasoning process, we use PSO algorithm to train fuzzy cognitive maps and obtain the weight of each criterion. Finally, fuzzy soft set is formulated and solved to identify the best partner. The major advantages of combining FCM with fuzzy soft set are that the evaluation can account for the interdependency of criteria/attributes and the uncertainties in decision making process. Such a combination was rarely seen in literature before. The remainder of the paper is organized as follows. Section 2 introduces the basic principles of FCM, PSO and fuzzy soft set. The proposed model is presented in Section 3. Section 4 illustrates the procedures in the proposed system using a numerical example. Finally, conclusions are drawn in Section 5.

A process for evaluating potential suppliers must consider the risk of disruption to the manufacturer’s assembly operation associated with the characteristics of the potential supplier [42]. Our paper differs from the previous studies in that we take risk into consideration. What’s more, we focus on the dependence and the feedback effects among criteria and uncertainties on the decision-making process. This study first presents a multi-criteria decision making model for evaluation supplier using FCM and fuzzy soft sets. FCM is used to depict the dependence and feedback among criteria which can overcome the preferential independent and the shortcomings of ANP. In order to compensate for FCM method’s dependence for expert advice in the reasoning process, this article introduces the PSO learning algorithm for training FCM. Fuzzy soft sets as a new mathematical tool can deal with uncertainties. A numerical example for selecting the most appropriate supplier is presented to examine the practicality of the proposed model. This proposed approach is not only novel but sufficiently general to be applied under various settings, thus it can help firms to measure and select the optimal suppliers in supply chains management. Also the integrated method can be extended to the analysis of other decision problems. Hybrid methods integrated with fuzzy soft set theory approach can be considered as a topic for future research.