Abstract
The most popular tools for traffic simulation and signal timing optimization are macroscopic and deterministic in structure. Microscopic tools are also available and are gaining popularity because they simulate the stochastic nature of traffic flow. Performance of the optimized signal timing developed by any tool depends on the fidelity of the traffic models embedded in it. This study evaluates the performance of signal timing developed by the most popular signal-timing optimization tools. It examines signal timing from each tool in various macroscopic and microscopic simulation environments and includes statistical tests to find which tool produces signal timing of the highest quality. Each of the four tools selected (two macroscopic and two microscopic) can both optimize and evaluate signal timing. A real-world arterial with 12 signalized intersections serves as the test bed. To eliminate skewness and bias in the results, all experiments were performed under the same geometric and traffic characteristics in each tool. Saturation flow rates from each calibrated model were assessed to ensure that all the tools processed traffic demand consistently. The results indicate that, overall, VISSIM-based genetic algorithm optimization of signal timing and the Synchro programs produce signal timing of the highest quality and provide extremely similar performance.