Design and integration of intelligent agents to implement sustainable production systems

Many works are in progress for sustaining the global environment, reducing wasted products, decreasing lead times and work in process (WIP), in order to improve acceptability, effectiveness and service of modern production systems. This is particularly true when considering ecology and new e-business constraints. Main problem comes from the structural complexity of such production systems. Multi agent systems (MAS) are enabled to break and manage complexity, offering modularity, flexibility and reactivity to production systems. Intelligent agents constitute the basement of the project PABADIS (Plant Automation Based on Distributed Systems). They allow for decentralization of execution tasks and dynamic reconfiguration of machines associated in a networked community. To support this project, a simulation tool has been developed: VFDCS (Virtual Factory Dynamics Configuration System) is intended to implement new concepts relevant to supply chain management (SCM) and demand flow technology (DFT).