Decentralizing decision making process of simultaneous bilateral negotiations of large multiagent systems in e-commerce applications using holonic structures

Abstract

Automated strategic negotiation is a key technology which facilitates to reach agreements between self-interested agents. Existing mechanisms mainly practice principles of auction mechanisms in which a single seller handles many buyers. The key advantage of that approach is the facility for the guarantee to reach the optimal deal because of global knowledge the seller has regarding the demand at hand. However, the main limitation of such an approach is the difficulty of exercising different negotiation strategies with different opponents. Although concurrent multiagent negotiation approaches resolve many of these limitations, the existing frameworks and protocols are too simplistic to handle dynamicity of larger multi-agent negotiation systems situated in very dynamic environments. The dynamicity of a large number of simultaneous bilateral negotiations can be appropriately coordinated through a topological structure that decentralises the autonomy of decision-making process under centralised supervision with the help of appropriate scheduling and planning mechanisms. This paper presents an application of a framework which coordinates the dynamicity of simultaneous bi-lateral negotiations by using Holonic structures. The framework possesses a top-down hierarchical structure with a decentralised control which takes the bottom-up perspective for effective decision making. The framework was simulated in an e-commerce application and showed how practically our approach could be utilised to model real-world scenarios to handle simultaneous bi-lateral negotiations of large multiagent systems.