A behavior-consistent information-based network control approach determines real-time traffic routing strategies by explicitly accounting for drivers likely response to the controller-recommended routes while generating these strategies. This paper proposes paradigms to deploy a behavior-consistent approach previously developed by the authors (1). They seek to enhance deployment effectiveness by analyzing the effects of alternative controller objectives and driver-preferred route sets used to recommend routes. Experiments are conducted using a test network. They analyze: (i) the performance of the behavior-consistent approach under commonly-used controllerobjectives, and (ii) the deployment flexibility enabled by increasing thenumber of driver-preferred routes considered by the controller for routing. The results suggest that the behavior-consistent approach can perform better than standard dynamic traffic assignment models while directing the system towards the desired state. They also illustrate the effectiveness of considering more driver-preferred routes in developing the information strategies. Further, performance is influenced by trade-offs between the number of driver-preferred routes considered by the controller for routing and the quality of routes relative to the controller objective. The resultssuggest that higher compliance rates may not translate to better performance and question the justification of user equilibrium solutions for routeguidance on the ground that a system optimal strategy is not behaviorallysustainable.
Abstract