Automated Task Negotiation in Multi-agent Based Virtual Enterprise

To cooperate effectively in virtual enterprise (VE), principal enterprise and member enterprises usually employ negotiation. With regard to the feature of VE, this paper presented a formal negotiation model for multi-agent negotiation for task allocation in VE. Then, constructed probability function as offer and counter offer strategies of agents and taken the expected price of both sides in the next turns and their reserved price into account, a contract-net based price negotiation algorithm is proposed. The algorithm considered the preference of agents and the restriction of time etc. By employing this algorithm, agent can modify its bid zone dynamically and learn about the opponent's bid price. In this way, the algorithm can reflect the attitude of the two negotiation parties and simulate the iteration of offering price very well in reality, which may promote the efficiency of negotiation and attain better negotiation result. Finally, a simple example is used to validate the approach's availability