Distributed Fuzzy Product Configuration Using a Multi-Agent Approach

This paper proposes an agent-based approach to assist the product configuration in conceptual design. Based on the distributed fuzzy models, fuzziness of interactions during the collaborative and distributed design for configuration, a computational approach for product configuration is developed. The modelling and the implementation of an agent-based system, called APIC (Agents for Product Integrated Configuration), is presented. In this agent-based system, there are four communities of agents: specification community of agents, function community of agents, physical solution community of agents and production constraint community of agents. Through intra and inter-communities interactions of agents emerge the consensual physical solutions agents, a sub community of physical solution agents. The optimal product configuration is searched based on the affinities of the consensual physical solutions agents. A case study is presented to demonstrate the potential of this approach.