摘要：In order to widen individual preference information extracting way and enhance practical values of the traditional DS/AHP method, multiplicative preference relations and fuzzy preference relations are reviewed as key extracting decision information way of pairwise comparison. After that, a basic model for deriving basic probability assignment (BPA) functions is constructed by means of minimizing the value of total deviation degree, which is able to attach importance to efficient elements and neglect inefficient elements of knowledge matrices in the process of calculating BPA functions. Based on the proposed basic model, an integrated method for DS/AHP is established to gradually integrate evidence pieces over all of criteria and that over all of group members by fusing ambidextrous decision information with criteria priority values (CPVs) and member priority values (MPVs). A numerical example about textbook choice is utilized to illustrate the whole procedure of the proposed method finally.
关键词：DS/AHP;multiplicative preference relation;fuzzy preference relation;integrated method;multiple criteria decision making
