Bilateral Price Negotiations by Software Agents - A Model for Measuring Marketplace-Related Negotiation Behavior

Software agents are seen as a promising technology for automating economic transactions as they can act autonomously on behalf of their human owners.In open, decentralized markets with incomplete information and possibly non-cooperative behavior, bilateral price negotiations have been shown as an adequate mechanism for balancing out the individual interests and for achieving coordination.Automating bilateral price negotiations and embarking on a strategy require inherent knowledge of the environment, e.g. of the typical negotiation leeway or behavior on the respective marketplace.Based on findings on negotiation theory, this contribution introduces a measuring system for quantifying marketplace-related negotiation behavior which can be used as a base for automated decision-finding in price negotiations and realizing a concrete strategy.The measuring system has been implemented in software agents and evaluated within a prototypical electronic marketplace.Results show the effectiveness of the measuring system and that coordination of the market can be achieved.