A model for distributed multi-agent traffic control

Abstract

Railway traffic is not always being controlled optimally. Using the current approach the Dutch Railways can only use up to 70% of the capacity of the infrastructure and adjustments often lead to delays. It has been investigated whether the use of distributed multi-agent traffic control can enlarge the use of the capacity of the infrastructure, improve the ability to make adjustments when disturbances occur and reduce delays. It seems that the approach is successful in reducing delays. More important, however, is that when delays occur, travelers are less bothered by the delays, because their interests are taken into account when adjustments are made.