Multi-agent Environment for Dynamic Transport Planning and Scheduling

Nowadays, transport requests of the society are becoming more and more common and important. Computer systems may be a useful tool for transport companies. This paper is aimed at presenting a system which simulates the functioning of a transport company. The theoretical problem, which needs to be solved, is a dynamic Pickup and Delivery Problem with Time Windows and capacity constraints (PDPTW). The nature of the problem, its distribution and the possibility of using a lot of autonomous planning modules, predestines to apply a multi-agent approach. Multi-agent approach facilitates the introduction of elements which do not appear in classical PDPTW as company organisation, different strategies of requests acceptation by different vehicles or communication among vehicles.