| The principal focus of this research is on the combined time-dependent signal control and System Optimal (SO) Traffic Assignment problem faced by a central controller with full information on time-dependent O-D trip desires as well as current network conditions. The controller seeks the optimal time-dependent routing policy and optimal signal setting plans such that the total travel time in the network will be minimized.;A mathematical formulation and an iterative simulation-based algorithm, along with a variant of it, are developed to find the solution to the problem. Also, a simulation-assignment modeling framework is developed to assess the efficiency and robustness properties of the algorithm as well as its variant. Experimental tests are conducted under different network loading levels generating various levels of congestion in the network ranging from virtually uncongested to very highly congested conditions nearing gridlock.;A number of simulation-based experiments are performed to test alternative approaches for calculating the time-dependent signal settings and to gain insights on several aspects of the problem. In particular, the effect of information availability, when calculating the signal settings for individual intersections, on the overall performance is examined. |
