Learning from user's behaviour of some well-known congested traffic networks

Intelligent transport systems planning is a research area that offers a variety of problems that are interesting from a mathematical point of view. In the last decades, due to technology evolution, one could argue that such planning might be enhanced and optimized by the use of satellite navigation devices that offer real-time information. However, as expressed in [9] and analyzed in [6], the use of such technologies generates the displacement of congestion from one zone to another. It is believed that such effects are a consequence of not knowing the travel choice patterns of users and making no behavior prediction. Hence, it is still very useful and valuable to study a deterministic and static version of the problem.