Simulation Optimization of Traffic Light Signal Timings via Perturbation Analysis

In this paper, we develop a simulation optimization algorithm for determining the traffic light signal timings for an intersection of two one-way street traffic flows modeled as single-server queues. The system performance is estimated via stochastic discreteevent simulation, and gradient-based search based on stochastic approximation is applied. In particular, we use smoothed perturbation analysis to derive both left-hand and right-hand gradient estimators of the queue lengths with respect to the green/red light lengths within a signal cycle. Numerical experiments compare the performance of the gradient estimators with finite-difference estimators and illustrate the optimization algorithm.