Holonic self-organization of multi-agent systems by fuzzy modeling with application to intelligent manufacturing

Holonic manufacturing aims to design standardized, modular manufacturing systems made of interchangeable parts, to enable flexibility, online reconfigurability and self-organizing capabilities for the production systems. Recent advances in distributed artificial intelligence and networking technologies have proven that theoretical multi-agent systems (MAS) concepts are very suitable for the real life implementation of holonic concepts. Building on our recent results in the design and implementation of holonic reconfigurable architectures, the paper introduces a novel approach to the online self-organization of distributed systems. By using fuzzy set and uncertainty theoretical concepts, we construct a mathematical foundation for modeling MAS, where appropriate holonic structures are identified for each particular application. This approach opens new possibilities for the design of any distributed system that needs self-organization as an intrinsic property.

[1]  David M. Dilts,et al.  The evolution of control architectures for automated manufacturing systems , 1991 .

[2]  George J. Klir,et al.  Fuzzy sets, uncertainty and information , 1988 .

[3]  A Koestler,et al.  Ghost in the Machine , 1970 .

[4]  Peter F. Patel-Schneider,et al.  The DARPA Knowledge Sharing Effort: A Progress Report , 1997, KR.

[5]  Lihui Wang,et al.  Design of Distributed Real-time Control Agents for Intelligent Manufacturing Systems , 1999 .

[6]  Nicholas R. Jennings,et al.  Intelligent agents: theory and practice , 1995, The Knowledge Engineering Review.

[7]  Xiaokun Zhang,et al.  Dynamic reconfiguration of holonic lower level control , 1999, Proceedings of the Second International Conference on Intelligent Processing and Manufacturing of Materials. IPMM'99 (Cat. No.99EX296).