Modeling the Effect of Road Pricing on Traffic Congestion Using Dynamic Travel Demand

Traffic congestion results from a mismatch between travel demand and road capacity. Because the latter depends upon relatively fixed infrastructure, growing metropolitan areas frequently face challenges from traffic congestion. By charging road users a congestion toll, a transport policy of road pricing has the potential to control travel demand. Road pricing is therefore seen as a promising strategy to alleviate traffic congestion. To evaluate the potential effect of implementing a road pricing policy as a congestion management practice, an agent-based simulation model was developed to capture the non-linear relationship between local interactions and global emergent phenomena. The policy mechanism of road pricing has impacts across scale, affecting aggregate traffic performance by changing individual travel behavior. The agent-based model developed in this study simulates the effect of road pricing on a set of autonomous and heterogeneous travelers whose dynamic travel demand is represented for multiple travel modes. Agent behaviors were examined under different spatial environments based on the defined algorithms. Four virtual experiments under different settings were compared to evaluate the effect of road pricing policy on peak-time traffic congestion. These experiments are used to explore the interaction between three traffic factors: travel demand, road capacity and traffic performance. The results demonstrate an improvement of traffic system performance with increasing toll price, underscoring the potential of road pricing as a means of traffic demand control. Applying such a toll on a particular stretch of road leads to a traffic re-distribution along the road network, in use of different transport modes, and in the timing of when travel is undertaken. This analysis points to the important role played by public transit in providing access to the community. This thesis demonstrates how a simulation model can be leveraged as a tool for scenario analysis to support policy makers considering how to implement road pricing in spatially distinct metropolitan contexts.