A dynamic multi-criteria decision making model with bipolar linguistic term sets. (1st April 2018)

• Record Type: Journal Article
• Authors: Liu, Hongbin; Jiang, Le; Martínez, Luis
• Is Part Of: Expert systems with applications. Volume 95(2018)
• Page Start: 104
• Page End: 112
• Publication Date: 2018-04-01
• Subjects: Dynamic multi-criteria decision making -- Uninorm -- Bipolar scale -- Linguistic 2-tuple -- Membership function; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2017.11.015
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: Novel dynamic multi-criteria decision making (DMCDM) model. DMCDM manages changes of criteria, alternatives and assessments across the time. Experts' assessments can be elicited in a bipolar linguistic scale (negative, neutral and positive). Linguistic information is modeled by the fuzzy linguistic 2-tuple approach. Bipolar linguistic 2-tuple values are aggregated by a full-reinforcement uninorm operator. Abstract: Real world decision making problems under uncertainty face different challenges. These challenges include lack of information, the necessity of quick decisions, and problems may change across time. When the uncertainty involved is due to fuzziness and vagueness, the use of fuzzy linguistic information can facilitate the elicitation of decision makers' preferences of alternatives by allowing assessment of alternatives in unipolar scales. However, in some cases decision makers need to express negative, positive and neutral attitudes that cannot be modeled by unipolar scales. This paper aims at developing a dynamic linguistic multi-criteria decision making model dealing with bipolar linguistic scales in which both alternatives and criteria may vary across time. In order to consider the historical evolution of alternatives leading up to the current assessments, a fusion process based on transformation functions and uninorm aggregation operator is proposed to deal with the membership functions of bipolar linguistic assessments. Finally the performance of this model is compared with a non-dynamic method in a supplier selection problem.