Research On Group Decision Support System And Coordinator Based On Multi-Agent In The Network Environment

With the development and improvement of computer,communication,decision support and artificial intelligence technology, the theory and application of Group Decision Support System have become an advanced topic of the cross-research between artificial intelligence and many other subjects. It is one most important development direction of Decision Support System.In order to solve dynamic,complex and uncertain problems,GroupDecision Support System Based on Multi-Agent System and coordinating mechanism are elaborated in this paper.This paper firstly introduces the concept and development history of Group Decision Support System.It analyses and summarizes the related research work and existing problems.Research content and outline are also introduced in it.After explaining the limit of centralized structure with very complex coordination commonly adopted in current GDSS research based on Multi-Agent System and application field of Group Decision Support System, it secondly proposes an innovative generic structure framework of Group Decision Support System based on Multi-Agent System and analyzing formally and description.Then ,the paper sets forth that decision Agent can take evidence composing of Evidence Theory as the fusion strategy in coordinator. The states occur probably in the process of decision fusing in coordinator of GDSS are analyzed in detail.Research on the related attribute of evidence, in fluence and method of dealing with is carried out. Aiming at conflicting problem of decision making,fusions trategy of BPA reassign is adopted in coordinator of GDSS. Aiming at contradictory of decision-making, the process of decision-making is divided into two stages that are learning and decision-making. Each Agent is trained by proper information given by training example to eliminate contradictories among decision-making Agents and correct mistakes that ensure Agents cooperate effectively in the stage of cooperating decision.The paper finally concludes the work and discusses its future work.