Optimization of Dynamic Congestion Pricing Strategy and its Effects on Traffic Flow – A Case Study of Athens City

With increasing urbanization, a future vision of urban mobility focuses on traffic conditions and congestion mitigation on the road network. Traffic congestion, particularly during peak hours, needs to be addressed in car-oriented cities; and tolling strategies are one of the critical tools that can mitigate the traffic. As a result, the goal of this research is to formulate the optimal congestion pricing strategy by incorporating the system's dynamic congestion into an optimization model. This study introduces a dynamic congestion pricing strategy to improve system efficiency, which also encourages users to make more effective travel decisions, which helps in traffic management. To formulate the optimization, model this research employs a dynamic traffic assignment model that incorporates the dynamic user equilibrium assignment for the route choice. An objective function is formulated in an optimization model. The objective function considered the dynamic congestion, parameters related to dynamic traffic demand, and traffic emissions. Further, the optimization model is validated in mesoscopic simulation model in the network of Athens city in AIMSUN Next software (a traffic simulation software). We incorporated the results of the simulation in the optimization model. We have utilized a state-of-art model-based optimization paradigm named as Bayesian optimization which tries to model the underlying black-box function based on prior-posterior causality. The output of simulation and optimization model showed the valuable insights in the traffic parameters like traffic delay, and emissions in the congested areas during the peak hours. The evolution in overall traffic performance with dynamic congestion toll is assess with macroscopic fundamental diagram.