Group Decision Making Model Based On Dynamic Fuzzy Logic And Its Application

GDM(Group Decision Making) focuses on selecting the best alternative(s), which stand(s) for the ideal one(s) decided by the group, from a limited set of solution alternatives. The complex social activities lead to more influence factors which make decision-making problems much more complicated. GDM can get more fair and reasonable results by brain-storming. Therefore, it shows incomparable superiority compared to individual decision-making and has been the hotspot and frontier of decisionmaking science. And GDM has already formed a relatively complete theoretical system.Although GDM has made considerable progress, there are still some issues to be studied in the GDM process. So far, most of GDM methods use static approaches to solve problems. They scarcely consider the dynamic information existing in individual preference information as well as the dynamic feature of decision-making process.Therefore, this thesis builds a GDM Model based on DFL(Dynamic Fuzzy Logic). The preference information can be represented by five common formats, namely: 1) utility values; 2) preference orderings; 3) reciprocal preference relations; 4) fuzzy preference relations; 5) linguistic evaluation information. Utilizing some functions provided by the model, we transform these different formats into dynamic fuzzy preference representation. Meanwhile, this thesis studies two typical GDM problems, including MAGDM(Multi-attribute Group Decision Making) and dynamic GDM, and gets some achievements as follows:Firstly, we propose two methods for the dynamic problems with respect to MAGDM.(1) Dynamic fuzzy MAGDM method based on entropy weight. By calculating the overall attribute entropy weight values for each alternative, we rank the alternatives and select the best one(s).(2) Dynamic fuzzy MAGDM method based on DFOWA(Dynamic Fuzzy Ordered Weighted Averaging) operator. We use DFOWA operator to gather individual preference information and build collective preference information. Then, it ranks and obtains the best alternative(s) according to the collective preference information.Secondly, for the dynamic group decision situation, in which the set of solution alternatives would change throughout the process, we propose a dynamic GDM method based on dynamic fuzzy preference relations. In the consistency/consensus reaching process, it uses the consistency level of individual preference and consensus degree of the agreement of all the decision makers simultaneously. Thus, it can not only avoid inconsistent individual opinions, but also reach an appropriate satisfaction value. During the selection process, we extend the IOWA(Induced Ordered Weighted Averaging) operator to IDFOWA(Induced Dynamic Fuzzy Ordered Weighted Averaging) operator, and adopt the concept of fuzzy majority to build dynamic fuzzy preference information.