Multi-agent systems as cooperation tools for supply chain management

Supply chain management requires the co-operation of the complete enterprise network to optimize the business process. Software agents and multi-agent systems help to electronically support the resulting co-operation process and thus necessary communications and negotiations between autonomous supply chain enterprises. The presented architecture for an agent system was develop in the context of the DFG project "Interaktive, plan-basierte Softwareagenten im verteilten Supply Chain Management (MAP-SCM)". It was developed to support supply chain enterprises and is before based on real structures and data form enterprises practice. This architecture itself supplies a close-to-reality simulation environment for the further evaluation of different models or protocols for communication and negotiation between the agent system of co-operating supply chain partners. In this context also probabilistic concepts for production plan evaluation will be examined with regard to their suitability for a ATP/CTP functionality in order to support negotiations.