Research On Method Of Group Reasoning And Group Ahp In GDSS

Group Decision Support System (GDSS) is one of important research branch of DSS. In the GDSS,techniques in communication and computer are combined with theory and method of Artificial Intelligence and management science as well as other special mathematical tools to promote the group who have different experiences and knowledge structure as well as shared responsibility to solve the semi-structure and non-structure problems in decision making procedure. GDSS has not become a theory system yet in that there is no existed theory or method to go by because of its complexity. Owing to default existing in research of theory and method of group reasoning,integrating of qualitative method and quantitative one and structuring of decision making problems,GDSS is unlikely to solve the semi-structure or non-structure problems perfectly. There lacks sufficiently successful experience in developing and applying GDSS in China,GDSS based on Internet particularly.This dissertation proposes group reasoning and group AHP-based GDSS-architecture and its decision steps to solve the problems existing in research of theory and application of GDSS. Besides,the dissertation also studies the method of group reasoning on rough-set and qualitative reasoning method basis in GDSS. It also studies on constructing method of hierarchy structure of group AHP in GDSS and combinative method of judgment matrixes. Moreover,it studies on method regarding construction of judgment matrixes on rough set based and optimization theory of convex combination parameters of judgment matrixes as well. Finally,it develops an AHP based GDSS to demonstrate the efficiency of these methods.(l)This dissertation also summarizes the developing trend of GDSS and analyses problems as regards the research of GDSS.(2) This dissertation restructures the GDSS and put forward new group reasoning and group AHP-based architecture named GDSSABGRGA by applying the theory of group AHP. It integrates GDSS with the rough set and qualitative reasoning-based group reasoning. Moreover,it combines physical logics-based qualitative reasoning with experience- based group reasoning in GDSS in order to promote the originality and the ability of GDSS to solve complicated problems. Furthermore,it analyzes the function and the relationship among the main components of the architecture. Finally,it illustrates the decision making process of the GDSS on the basis of this architecture.(3)This dissertation makes some researches into the rough-set-based group reasoning method that can be used to acquire the qualitative knowledge and cites an example to illustrate the reasoning process on rough set-based group reasoning as well. It analyzes existing problems applying qualitative reasoning method named QSIM to decision process. It also researches into GDSS-based QSIM that can be used to the group decision of qualitative problems.(4) This dissertation makes a summery of decision thought of AHP and makes a study on structuring method of the complicated problems. It set up two combination ways of judgment matrixes in group AHP. It also proves its theory. It also makes some research into optimal combination theory of judgment matrixes of group AHP. Then it put forward a rough-set based method of constructing judgment matrix.(5) This dissertation develops a group AHP based GDSS based on Internet,proves the efficiency of the optimal combination method and demonstrate that developing GDSS based on Internet is feasible.The achievements made in this dissertation are as follows:(1) This dissertation proposes a theory that integrates GDSS with the theory and method of group reasoning and AHP to solve the problems of the current research in GDSS.(2) By applying the theory of AHP,this dissertation restructures the GDSS and proposes new group reasoning and group AHP-based architecture of GDSS. It integrates the GDSS with rough set-based group reasoning and group decision-based qualitative reasoning. It also merges the physical logic-based qualitative reasoning method and rough-set bas...