Genetic algorithm-based traffic lights timing optimization and routes definition using Petri net model of urban traffic flow

Abstract

This work presents an algorithm for the optimization of urban traffic flow that computes the vehicle routes and traffic lights timing in real time. The optimization procedure uses a genetic algorithm whose fitness function consists of a high-level Petri net model of the urban traffic flow, which simulation results in the fitness value to be used. The outcome of this work is the optimization of urban traffic flow by simultaneously establishing the best possible routes for each vehicle and the definition of the most appropriate traffic lights timing. According to the tests, the simultaneous optimization of traffic lights time and vehicle routes decreased the total travel time as compared to the optimizations performed considering only the routes.