Dynamic integrated freeway/traffic signal networks: A routing-based modelling approach

While the potential benefits of integrating freeway and traffic signal control have long been recognized, fundamental incompatibilities between their respective modelling approaches have prevented the development of a suitable simulation model. Such a model is not only required to evaluate the potential of externally generated control strategies, but it should ultimately also form the basis for a comprehensive optimization model. This paper presents a new modelling approach, called INTEGRATION, which has been designed specifically to evaluate such integrated networks and their controls. The approach considers the behavior of traffic flows in terms of individual vehicles that have self-assignment capabilities. This capability serves as a traffic assignment function within an integrated network of traffic signals and freeways, and circumvents the need to use either an explicit time slice or iterations during the traffic assignment. Consequently, one can consider continuously variable traffic demands and controls, both freeway and signalized networks, as well as any links that joint them. While the proposed approach requires further field testing and validation, it does represent a critical step towards the development of an ultimate tool for evaluating and optimizing integrated traffic networks and their controls.