فرآیند تصمیم گیری چند هدفه یکپارچه برای انتخاب تامین کننده با مشکل دسته بندی
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|19162||2009||11 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Expert Systems with Applications, Volume 36, Issue 2, Part 1, March 2009, Pages 2327–2337
When the cost of raw materials or component parts dominates the product cost, supplier selection becomes a crucial process for the company to maintain the cost while holding the quality of the products. At the same time, it is likely that the supplier offers bundling products, strategy to get more orders from the company. In this situation, purchasing manager requires decision-making tool which can deal with these problems simultaneously. This article presents an integrated multi-objective decision-making process by using analytic network process (ANP) and mixed integer programming (MIP) to optimize the selection of supplier. The criteria, which are gathered from experts by using Delphi method, are used to construct an ANP model, and are continued to be used by collecting the data from them. The results indicated that cost per unit and failure product cost are important determinants. Thereafter, the ANP results were used as coefficients of an objective function in MIP to allocate order quantities if the supplier uses bundling strategy. A hypothetical example is presented and the results indicated that the combination of ANP and MIP provided useful tool to select the optimal supplier.
Nowadays, competitive business environment has forced companies to satisfy customers who demand increasing product variety, lower cost, better quality, and faster response (Vondrembse, Uppal, Huang, & Dismukes, 2006). Therefore, offering higher product quality is the main requirement to gain global market share. In addition, companies operate at the lowest possible cost in a competitive market to generate substantial profit (Lau, Pang, & Wong, 2002). These objectives should deal carefully in the supplier selection process, since it enables companies to reduce purchasing cost and improve corporate competitiveness (Demirtas and Ustun, 2008 and Ghodyspour and O’Brien, 2001). However, selecting the right supplier is always a difficult task for many purchasing managers (Liu & Hai, 2005). Managers should realize that no supplier can satisfy all their requirements. Commonly, one supplier satisfies one part of the requirements and another supplier satisfies the other part of the requirements. Therefore, the company has to evaluate and select all possible supplier candidates to various requirements or attributes (Ghodyspour & O’Brien, 1998). These requirements are composed by qualitative as well as quantitative attributes, and the company has to choose the most suitable supplier as its supply chain members. In Dickson (1966), proposed quality, cost and delivery performance as three of the most important attributes. Since then, many conceptual and empirical studies for supplier selection have been reported (Verma & Pullman, 1998). In general, there are two types of supplier selection problem, single sourcing and multiple sourcing (Demirtas & Ustun, 2008). Many studies proposed to deal with single sourcing problem, such as the well-known AHP (Barbarosoglu and Yazgac, 1997, Bhutta and Huq, 2003, Çebi and Bayraktar, 2003, Ghodyspour and O’Brien, 1998, Khurrum and Faizul, 2002, Korpelaa et al., 2001, Liu and Hai, 2005, Mohanty and Deshmukh, 1993, Narasimhan, 1983, Nydick and Hill, 1992, Sarkis and Talluri, 2000, Sarkis and Talluri, 2004, Weber and Current, 1991 and Yahya and Kingsman, 1999) and ANP (Meade and Presley, 2002 and Sarkis and Talluri, 2000), have been used. For multiple sourcing problem, scholars tend to use linear weighting methods (De Boer, Van der Wegen, & Telgen, 1998), mathematical programming (MP) techniques (Akinc, 1993, Barbarosoglu and Yazgac, 1997, Benton, 1991, Bender et al., 1985, Current and Weber, 1994, Degraeve and Roodhooft, 2000, Karpak et al., 1999, Narula and Vassilev, 1994, Rosenthal et al., 1995 and Sadrian and Yoon, 1994), and the combination of ANP and MP techniques (Demirtas & Ustun, 2008). The bundling problem can be done by using MP techniques (Rosenthal et al., 1995 and Sarkis and Semple, 1999), and this study integrates ANP and mixed integer programming to select the best supplier when they use product bundling strategy and defines the optimum quantities among the selected suppliers. The context of this paper is the notebook industry in Taiwan, since it is the 4th largest manufacturer of computer-related products. Seventeen items of IT products made in Taiwan occupy over a half of the world’s market. Among them, notebook computers occupy 61% of the world’s market; main board, 75%; and LCD monitors, 61% (Lu, 2005). Although these facts mainly contributed to Taiwan’s OEM industry, there is a growing sales trend that some of the Taiwanese notebook producers succeed in selling under their own brand. Due to a rapid technology change and high competition among the Taiwanese notebook manufacturers as well as abroad competitors, selecting suppliers is one of the most important steps to offer innovative products with high quality and reasonable price. Relevant literature is reviewed in the section that follows. This article develops the ANP model based on the discussion results among practitioners and experts, which is followed by collecting the data by using Delphi method. Thereafter, the ANP results were used as coefficients of an objective function in MIP to allocate order quantities if the supplier uses bundling strategy. The ANP as well as MIP procedure is illustrated through numerical results based on experts’ interview from Taiwan’s notebook producers. The data were used to demonstrate the ANP and MIP application and examine its effectiveness. Finally, the conclusions and suggestions of the paper are described.
نتیجه گیری انگلیسی
This paper used a two-stage integrated approach to select suppliers when the bundling strategy existed. Being different from other integrated approaches, such as Demirtas and Ustun (Demirtas & Ustun, 2008), an integration of ANP and MIP techniques is suggested to deal with the research question. The first stage uses ANP in supplier selection process of notebook producers in Taiwan. By using relative measurement that deals with different fluctuations of economic condition or the rapid technological change, ANP is a valuable method to control and forecast the dynamics of those conditions toward production side (Saaty, 2006). The results showed that the Taiwanese notebook producers tended to assign cost and quality in the same weights. In addition, the variable cost was the highest influence among other criteria for producers to select suppliers. The results assigned delivery performance, information system capabilities and cost per unit as significant variables in supplier selection process to fit with their condition. The second stage uses the weights computed by ANP as coefficients in the first objective function of MIP model. An example is demonstrated to put it into practice by proposing bundling problem and is solved through mixed integer programming. And the results indicated that the demonstrated method could be valuable for firms to select suppliers and to assign order quantity. Even though some might say that complicated decision-making methods would cause purchasing managers very low intention to adopt, but it can be said that a combination of ANP and MIP does provide a useful tool to deal with qualitative and quantitative issues in supplier selection process, especially when the suppliers offer bundling strategy in their sales term. Future research could involve refining the data collections described in this paper through a series of interviews in purchasing teams from other industries. Once these ideas have been transformed into a research framework, data could be collected and the framework could be tested.