A Space-Time Network Optimization Model for Traffic Coordination and Its Evaluation

In transportation systems, the existing infrastructure can potentially be used more efficiently by deploying intelligent net-centric solutions that coordinate vehicles and traffic signals in real-time. For capacity planning and assessing the cost/benefit tradeoffs of intelligent net-centric coordination infrastructures, it is essential to determine the performance of optimal solutions, i.e., the best possible traffic flow that can be achieved. Given the scale and complexity of transportation systems, it may not be feasible to actually achieve these optimal performances in practice. However, if the results show that substantial improvements are possible by simply using the current physical roadway infrastructures more effectively, then one can justify the cost of deploying intelligent vehicle/traffic-light coordination systems. In this paper, we demonstrate these concepts through a case study of scheduling vehicles on a grid of intersecting roads. We develop heuristic algorithms and an optimization model using the space-time network for this problem, and compare them. Moreover, we also compare the space-time network modeling technique with the integer programming optimization approach and show that the former is better for modeling traffic coordination systems.