Methods For Multiple Attribute Decision Making Problem

Multi-attribute decision-making is the main content of the Modern Decision Analysis, and their assence is to rank the alternatives and choose the best alternative under many criteria. This paper mainly deals with the multi-attribute decision-making problem. It consists of three chapters.In the first chapter, we mainly introduce the development and current research situations of the multiple attribute decision making problem. Furthermore, we list the main contribution of this paper.In the second chapter, a method is proposed to solve the multiple attribute decision making problem, where the decision maker gives his preference on alternatives in a fuzzy relation. In order to reflect the decision maker's subjective preference information, an optimization model that the fuzzy judgment matrix approaches an additive consistency matrix is constructed to assess the attribute weights and rank the alternatives. Finally, an example is used to illustrate the practicality of the proposed method.In the third chapter, methods are proposed from the multiple attribute decision making under linguistic setting and the intuitionistic fuzzy sets. For the multiple attribute decision making under linguistic setting, we conduct this research in which the information about the attribute weights is completely known and the attribute values take the form of linguistic variables and uncertain linguistic variables. Then we realize the alternatives ranking by use of ideal solution and deviation degree, and finally examples are shown to explain this method's practicality and feasibility; And for the multiple attribute decision making problem under intuitionistic fuzzy environment, this paper presents a novel TOPSIS method to deal with this problem, in which the information about attribute weights is partly known. We construct a multi-objective programming model to derive the attribute weights by defining the Hamming Distance between two in-tuitionstic fuzzy sets and the positive-negative ideal solution, then we use the general silimilarity measure to determine the rank of all alternatives. Finally, an example is shown to highlight the validity and feasibility of the proposed method.