An hybrid intelligent approach for real-time traffic control

Abstract

This paper deals with the problem of optimally regulating planned vehicle timetables for public transport when unforeseen events occur in real-time in the network. A multicriteria problem is solved by using an integrated intelligent approach that combines agent-based techniques, Tabu search algorithm and fuzzy preference model. The agents of our model act cooperatively in order to generate efficient solutions that optimize simultaneously and separately the different regulation objectives. This optimization is performed by means of a distributed tabu search algorithm. Efficient solutions are then, classified according to a fuzzy preference model by using an interactive solutions evaluation among agents. In order to assess the distributed approach, an experimental study was carried out on the base of some scenarios of disturbances occurred in a public transportation network.