Supplier selection with different risk preferences and attribute sets: An innovative study based on generalized linguistic term sets. (October 2021)

• Record Type: Journal Article
• Authors: Lin, Hui; You, Jianxin; Zhang, Xueyang
• Is Part Of: Advanced engineering informatics. Volume 50(2021)
• Publication Date: 2021-10
• Subjects: Supplier selection -- Multi-granular linguistic MAGDM -- Risk preferences -- Several attribute sets -- Incomplete attribute weight information -- GLTS; Computer-aided engineering -- Periodicals; Engineering -- Data processing -- Periodicals; 620.00285
• DOI: 10.1016/j.aei.2021.101424
• Languages: English
• ISSNs: 1474-0346

Abstract

Abstract: Selecting the optimal supplier is crucial to the management of the company's supply chain, which has received widespread attention from academia and business circles. Generally, a variety of suppliers and multiple attributes are usually involved in this selection proceeding which can be comparatively regarded as a linguistic multiple attribute group decision making (MAGDM) problem. However, common linguistic MAGDM problems may not take the following characteristics into consideration. Due to the limited knowledge, distinctive interests, and different semantic value expectations of decision makers (DMs), it is necessary for them to consider their diverse risk preferences and use multi-granular linguistic term sets (LTSs) to assess suppliers on individual attribute sets independently. Meanwhile, the complex decision environment may have influences on the integrity of the attribute weight information, such that it is always incompletely known. To deal with the afore-mentioned situations, this paper presents a procedure based on risk preferences and several attribute sets with incomplete weight information for choosing the desirable supplier. Firstly, a new multi-granular fuzzy linguistic transformation model is constructed to normalize linguistic domains of multi-granular generalized linguistic term sets (GLTSs). Multi-granular GLTSs are introduced to describe semantic values of multi-granular LTSs given by DMs with risk preferences. Secondly, according to the maximizing deviation method with incomplete attribute weight information, an optimization model is also established to determine attribute weight vectors of individual attribute sets. Thirdly, a novel method that comprises of the aforesaid models is presented to handle supplier selection problems with risk preferences and several attribute sets. Finally, an illustrative example on supplier selection and comparative analyses are provided to clarify the validity and feasibility of our proposed method. Significantly, the initiation of the proposed method and its application could afford to the theoretical development of linguistic MAGDM, as well as the practical expansion in the domain of supplier selection.