Research On Extended Model For Multiple Attribute Decision Making Based On Fuzzy Information

Multiple attribute decision-making (MADM) is an important part of modern decision science. It always involves multiple decision attributes and multiple decision alternatives. The purpose of the decision-making is finding the most desirable alternative(s) from a discrete set of feasible alternatives with respect to a finite set of attributes. It has been extensively applied to various areas such as society, economics, military, management, etc. Therefore, the research on MADM has important theoretic significance and higher practical application. This paper studies extended model of multiple attribute decision-making based on fuzzy information, and the primary coverage is as follows:1. A new classification model based on fuzzy associative rules is proposed. The results on five data sets show that the proposed model has high accuracy rate, and better interpretability compared with homogeneous classification model.2. Two improved PROMETHEE II methods are proposed to solve the MADM problems with completely unknown weight information. Firstly, inspired by ensemble methods, an ensemble PROMETHEE II model is proposed. One example of competitiveness evaluation of the ports of Circum-Bohai-Sea is included to illustrate the proposed method. Secondly, an improved PROMETHEE II method based on AFS is proposed, in which a new kind of preference functions is defined and a subjective and objective integrated approach is used to determine weights of the attributes. With AFS, the preference functions are defined by the ranking and difference degrees of performance value on each attribute. Furthermore, the proposed preference functions do not depend on parameters, which reduce the subjectivity of decision makers.3. A ranking method for MADM based on AFS clustering is proposed, the weights of decision makers for multi-attribute and multi-alternative large group decision-making are determined by using AFS clustering. Then, the ordering vectors of alternatives are obtained by synthesizing the decision makers’ weight vectors and decision matrices. Furthermore, a ranking method based on dominant relation and FCM clustering is proposed, and a method to determine the priority level in an uncertain MADM is given, in which FCM clustering theory is employed to deal with partition of the original data. This method can reduce the influence of subjective factor of equal interval partition, and it is suitable for MADM of large scale samples cases.4. Study on gray relational analysis model based on hesitant fuzzy linguistic term sets. The weights of attribute are determined by the group decision-making models based on AHP in this model. Futhermore, concepts of expected value, standard variance and distance measure with respect to uncertainty hesitant fuzzy linguistic term sets are proposed. We extend the model to uncertain linguistic environments, i.e., present the gray relational analysis model based on uncertainty hesitant fuzzy linguistic term sets. The illustrative examples show the effectiveness and feasibility of the proposed model.