روش تصمیم گیری چند صفتی با نقاط مرجع دوگانه
Abstract
I. Introduction
II. Theoretical Analysis of DRPs Problems
III. Decision-Making Methiod
IV. Application of Decision-Making Method to the Location of an Agricultural Product Logistics Center
V. Conclusion
Authors
Figures
References
Abstract
A single reference point (SRP) is typically employed in traditional studies on the multi-attribute decision making (MADM) method. However, SRP lacks the advantage of the multiplicity of reference points and is thus unable to adequately describe loss aversion actions. With the goal of determining the loss aversion characteristics of decision makers, this work proposes an MADM method that is based on double reference points (DRPs) to solve the MADM problem using crisp and interval fuzzy numbers. First, this work describes the universality of decision-making problems with DRPs and then evaluates the characteristics of DRPs and their effects on decision making. Second, attitude and utility functions are established on the basis of the requirements of loss aversion. Third, considering the ‘‘one-vote veto’’ characteristic of a ‘‘dissatisfaction’’ attitude, a binary utility value is created by integrating the occurrence probability p, and the utility value u. The fourth, the main characteristics of and the aggregation method for the binary utility value are analyzed and established. Finally, a new decision-making method is applied to resolve the location decision of an agricultural product logistics center. Results indicate that the relationship between attribute and reference point significantly influences decision behavior. Compared with the traditional decision-making method, the proposed decision-making method can effectively identify the feasibility levels of alternatives. The ranking of the proposed decision-making method for feasible alternatives is basically the same as that of the traditional decision-making method. These results can effectively help solve agricultural engineering problems and broaden the coverage of and provide references for MADM research.
Introduction
Multi-attribute decision-making (MADM) problem refers to the ranking and selection of finite alternatives with multiple attributes [1], [2]. Owing to the MADM problems generally occurring in daily production and activities [1]–[5], the academic community has carried out comprehensive analyses. Today, studies on MADM problems come in two forms: those focused on perfect rational decision making and those focused on bounded rational decision making. The former is typically based on expected utility theory and is frequently adopted to solve MADM problems. The recent studies on the theory and method of MADM suggest that the research on MADM based on complete rationality has achieved substantial results [6], [7], [8]. The latter type is based on loss aversion theory. According to this theory, loss, relative to a reference point, produces greater psychological utility than the same amount of gain does [9], [10]. Loss aversion is a common phenomenon that is rooted in people’s decisionmaking actions and occurs in the areas of politics, economics, athletics, and so on [10]–[12]. Tom et al. even published a thesis on the science that proves loss aversion as a basic mechanism of human beings or animals [13].