A Multi-agent Cooperative Negotiation Mechanism in Mobile Agent-based Commerce Environment - The Design and Implementation

• Main Authors: Chien-Nan Kuo, 郭建男
• Other Authors: Fang-Chang Lin
• Format: Others
• Language: zh-TW
• Published: 2001
• Online Access: http://ndltd.ncl.edu.tw/handle/00028398067986162149

碩士 === 朝陽科技大學 === 資訊管理系碩士班 === 89 === Over the past few years, electronic commerce has been developed prosperously. There have been a lot of research and applications on it. In order to reduce the trade cost between buyers and sellers, agent technique has been utilized to develop environment for automatic commerce. Besides, mobile agent technique provided the mobility for agents in the network. On the basis of the above issues, this research developed an e-Marketplace system for mobile agents. In this system, the buyers assigned agents to search the goods information and dispatched agents to supplier sides and traded with supplier agents. In order to work efficiently, this search utilized multiple buyer agents to trade cooperatively and simultaneously. In other words, the buyer assigned multiple buyer agents to trade with supplier agents simultaneously according to the amount of supplier agents. For the negotiation in electronic commerce, it needed to exchange the information between buyers and sellers to reach to an agreement. In order to simulate the negotiation in the real world, this research utilized fuzzy inference to evaluate the proposals of suppliers for one buyer to multiple suppliers. For the application domain, the research took the tourism as an application domain to develop the e-Marketplace system. For example, a travel agency could utilize agents to make room reservations with hotel. This research purposed to provide an architecture and cooperative negotiation in the e-Marketplace for the tourism application. This research had three major achievements: (1) Constructing a B2B e-Marketplace system for mobile agents to providing a flexible and efficient architecture of system; Applying the tourism as an application domain to develop the e-Marketplace system. (2) Developing an asynchronous cooperative negotiation mechanism for buyer to assign multiple agents to trade with supplier agents simultaneously in order to raise the efficiency of system. (3) Utilizing fuzzy inference to evaluate the proposals of suppliers during negotiation in order to fit in with humans.