Network-Centric Approach to Real-Time Train Scheduling in Large-Scale Railway Systems

摘要

Railway transport development leads to a rapid increase of passenger and cargo traffic. Train dispatcher often unable to cope with increasing traffic volume. So there is a great need in intelligent real-time systems of railway traffic control. We propose a network-centric approach to creating the system of real-time train scheduling on the basis of multi-agent technologies. The architecture of the network-centric multi-agent system consisting of base planning subsystems is described. Subsystem interaction protocols and protocols of agent interaction within each subsystem are presented. The example of schedule planning is presented. Productive characteristics of the developed system are presented. Good quality of train schedule planning and system performance is shown.