CONAN: A heuristic strategy for COncurrent Negotiating AgeNts

We develop CONAN, a heuristic agent for concurrent bilateral negotiations in electronic markets that are open, dynamic and complex. Existing strategies often omit the factors determining when a market environment is open or how an agent evaluates progress in bilateral negotiations. Such omissions in turn damage the offer-making ability of an agent and consequently the number of successful negotiations that this agent can achieve. Negotiation experiments indicate that CONAN outperforms other agents that rely on the current state-of-the-art by a significant amount in terms of the utility gained during a negotiation.