A visual and statistical study of a real world traffic optimization problem

Abstract

In our group we have been dealing with the Traffic Lights programming problem for a couple of years. We have designed a new architecture to cope with this problem. In our system we use a Genetic Algorithm for performing this NP-Complete non-deterministic optimisation. Through this work our aim is to give an inside view -by means of a visual and statistical study - of what is happening during the evolution of the algorithm until it converges. Recently, we have been provided of real traffic data from the Santa Cruz de Tenerife local government for our research. A wide set of tests have been performed with these data, varying some algorithm parameters. In this paper we present some remarkable conclusions from all those tests as well as a new visual approach to the Genetic Algorithm tuning process.