Congestion Management in Urban Traffic Networks based on Macroscopic Fundamental Diagram

With the expeditious development of urban modernization, traffic congestion, air pollution, and road safety problems are necessary to be paid attention to. Therefore, the crucial research field of Intelligent Transportation System (ITS) is heavily employed and explored to monitor traffic congestion in real-time and apply responsive control schemes. However, effective control and management approaches for large-scale urban traffic networks still remain a big challenge, due to the complicated modeling of traffic dynamics in large networks (e.g. large number of road links, signalized intersections) and unpredictable traveler patterns (such as route and mode choices, and departure times). This thesis employs Macroscopic Fundamental Diagram (MFD) to improve mobility and decrease vehicles total time spent in the urban network. Five main chapters included in this thesis. Chapter 1 introduces the background and motivation in studying network-level modeling and control strategies based on MFD of large-scale urban networks. Chapter 2 develops a perimeter flow control strategy employing Sliding Mode Control (SMC) theory to diminish the influence of heterogeneous inter-region transfer flows and cordon queues on traffic congestion of the two-region network. Chapter 3 extends the perimeter control strategy and MFD modeling to a multi-region heterogeneous network to provide a more complex control framework considering location-varying region boundaries. Chapter 4 introduces a congestion pricing scheme grounded on Model Predictive Control (MPC) for a multi-region urban network as an elegant control approach to simultaneously achieve reducing vehicles' total time spent as well as maintaining revenue neutrality of the whole network. Chapter 5 closes with summary and future research directions