Research On The Deadlock Elimination Of Negotiation Based On Particle Swarm Optimization

With the rapid development of Electronic Commerce and the popularization of Internet, we are much stricter with Electronic Commerce. In order to meet this request, the Agent technology is used in this field to improve Electronic Commerce capacity and practicability. In recent years, Agent-based autonomy negotiation has been widely used in the field of the Electronic commerce.Autonomy negotiation model is the base of dual Agent achieving multi-issues autonomy negotiation. But practically, in the process of dual multi-issues autonomy negotiation, the negotiation deadlock often occurs because one of the negotiation topics did not achieve the balance point. To improve the successful rate of negotiation and negotiation effect, the study of negotiation deadlock mechanism is necessary. Besides, it is significative to promote the intelligent electron commerce to be more practical. At present, how to eliminate deadlock is one of the most important parts in the research of multi-issue negotiation.Based on the bilateral and multi-issue negotiation model and the principle of the vector's equivalent replacement, a mathematic model of deadlock-eliminated has been proposed on the condition that the integrated-utility of different negotiators are equal. We use this model to optimize the automated negotiation model in electronical commerce. Aiming at the problem of equivalent replacement the evolution equation of the PSO is improved to insure the particles moving in a hyper-plane. In order to validate the validity of the model, we design a dealing process of electronical commerce. The improved PSO is used to optimize the deadlock-eliminated mathematical model and the reserved value vector of the negotiation issues which is optimized to eliminate the deadlock. The result shows that this method has enhanced the successful rate and the negotiation utility. We improved that the system is valid.The paper contains seven parts:(1)Expound the study background and the signification of the paper, and summarize the study actuality of the deadlock elimination and the particle swarm optimization algorithm.(2)Introduce the definitions and correlation concepts of Electronic Commerce, Multi-Agent technique and automatic negotiation technique.(3)An automated negotiation model in electronic commerce based on Agent is proposed in this paper. First, pertinent knowledge of the negotiation environment is introduced; and then the concept and theory of negotiation issue, suggest policy, and negotiation agreement is described; lastly formal definition of the automated negotiation is presented.(4) Introduce the theory of deadlock elimination. A mathematical model of deadlock elimination is proposed and the alternation process is described.(5) Introduce both the Particle Swarm Optimization and several improved Particle Swarm Optimization. Particle Swarm Optimization has been used in the process of deadlock elimination. Based on the deadlock elimination model we improved the basic Particle Swarm Optimization, and use it to solve the problem of deadlock in negotiation.(6) Introduce the design of deadlock elimination, and use particle swarm optimization to eliminate the deadlock in automated negotiation model. The validity of this method is validated in the experiment.(7) Summarize the work generally and point out the shortage of the paper. Bring forward the possible study direction.