AbstractThe digitalization of logistics processes is often based on distributed models and decentralized control. As these logistics models constitute an important part of Industrie 4.0 concepts they must be powerful enough to cover dynamic processes and must enable a host of functions such as goal-oriented, reactive, pro-active, communicative, cooperative, competitive, and learning behaviors. In addition, these distributed models must allow for simulating, planning, allocating, scheduling, and optimizing logistics tasks. This implies that they must be able to act through communication channels with each other thus establishing logistics social communities.Multiagent Systems (MAS) have been around for more than 30 years and lend themselves to the implementation of these distributed models needed for autonomous and cooperating logistics processes. It will be described and also demonstrated by three case studies why MAS are well suited for social and learning logistics systems. It will be shown how the resulting distributed MAS models provide the required functionalities for production and transportation logistics including the handling of dynamic local events as an essential feature for the successful planning, scheduling, optimizing, monitoring, and control of global logistics processes.
Price-Setting Combinatorial Auctions for Coordination and Control of Manufacturing Multiagent Systems: Updating Prices Methods
