iCoMAS: An agent-based system for cooperative transportation planning in the food industry

Abstract

In this paper, we describe a cooperative transportation planning problem that is motivated by a real-world setting found in the German food industry. Several manufacturers with joint customers but complementary food products share their fleets to deliver the customers. After an appropriate hierarchical decomposition of the transportation planning problem into subproblems, a set of rich vehicle routing problems with time windows for the delivery of the orders, capacity constraints, maximum operating times for the vehicles, and outsourcing options is obtained. These subproblems are solved by an Ant Colony System (ACS). A multi-agent-system (MAS) is proposed that differentiates between decision-making and staff agents. It improves solutions by exchanging appropriate orders between subproblems. Furthermore, it allows working with local data. Some results of simulation experiments with the MAS are presented. © 2011 Springer-Verlag.