Determinating The Priority Vectors Of Alternatives In Fuzzy Multiple Attribute Decision Making

Fuzzy multi-attribute decision making (FMADM) is a primary research aspect in modern decision-making analysis, and their main task is focus on how fuzzy information gathering, election to determine the priority vectors of the alternatives. The main researches of this thesis are problems of prioriy vectors of alternatives and its application based on fuzzy judgment matrix in fuzzy multi-attribute decision making, achievements are as follows:1. In view of the question of analytic hierarchy process(AHP), new logically relation of elements of fuzzy positive consistent reciprocal and complementary judgment matrix with priorities are determinated based on the relation of elements of fuzzy positive judgment matrix with priorities and the consistency of human thinking. A new formular of priority vectors is proposed, in which the parameter is included. Decision maker would choice appropriate parameters and acquire priority vectors with better discrimination which makes the ranking alternatives with persuasion. Several important conclusions are obtained, therefore, Saaty's AHP is extended based on this proposed method. Based on new parameter prioritization method of positive complementary judgment matrics andα-levels and its ordering and possibility, a new method of FMADM is proposed, which deals effectively with the elements in attribute values matrix and attribute weights are supposed as L-R fuzzy numbers.2. To the question of fuzzy weighted average (FWA) and its application of limitations, a method of developing FWA is proposed based on theories ofα-levels and the arithmetics of interval. Then interval synthesis values of alternatives underα-level are obtained based on the theories ofα-levels by using this development method. Complementary judgment matrix is built by employing possibility of interval, then a priority vector is found by using new parameter prioritization method based on positive complementary judgment matrics. Total priority vectors of alternatives are obtained by gathering the synthesis values at differentα-levels, and then the ranking or chosing the best of alternatives.