A Multiclass Simulation-Based Dynamic Traffic Assignment Model for Mixed Traffic Flow of Connected and Autonomous Vehicles and Human-Driven Vehicles

One of the potential capabilities of Connected and Autonomous Vehicles (CAVs) is that they can have different route choice behavior and driving behavior compared to human Driven Vehicles (HDVs). This will lead to mixed traffic flow with multiple classes of route choice behavior. Therefore, it is crucial to solve the multiclass Traffic Assignment Problem (TAP) in mixed traffic of CAVs and HDVs. Few studies have tried to solve this problem; however, most used analytical solutions, which are challenging to implement in real and large networks (especially in dynamic cases). Also, studies in implementing simulation-based methods have not considered all of CAVs' potential capabilities. On the other hand, several different (conflicting) assumptions are made about the CAV's route choice behavior in these studies. So, providing a tool that can solve the multiclass TAP of mixed traffic under different assumptions can help researchers to understand the impacts of CAVs better. To fill these gaps, this study provides an open-source solution framework of the multiclass simulation-based traffic assignment problem for mixed traffic of CAVs and HDVs. This model assumes that CAVs follow system optimal principles with rerouting capability, while HDVs follow user equilibrium principles. Moreover, this model can capture the impacts of CAVs on road capacity by considering distinct driving behavioral models in both micro and meso scales traffic simulation. This proposed model is tested in two case studies which shows that as the penetration rate of CAVs increases, the total travel time of all vehicles decreases.