Optimizing Traffic Signal Timing at Urban Intersections: A Simheuristic Approach Using GA and SUMO

This study introduces an innovative simheuristic framework that integrates the Simulation of Urban MObility (SUMO), a detailed microsimulation tool, with the Genetic Algorithm (GA), a robust optimization method, for optimizing traffic signal timing (TST) at signalized intersections. Specifically designed to be applied to typical four-leg intersection phase plans, this framework systematically determines the most effective green signal timings to enhance traffic flow efficiency and reduce environmental impact. By meticulously testing each potential TST solution generated by the GA, using SUMO to simulate its real-world impacts, the framework provides a thorough assessment of various signal timing strategies. Comparative analyses against established methodologies, such as the Particle Swarm Optimization (PSO) algorithm and Webster's traditional method, are conducted during peak traffic demand periods to evaluate the framework's effectiveness in managing congestion and emissions. Our results demonstrate that the proposed simheuristic approach significantly outperforms the benchmarks: it achieves a reduction in CO levels by 4.97% compared to PSO and 11.76% compared to Webster; NOx emissions are reduced by 2.5% and 3.94%, respectively; and PMx levels see a decrease of 3.83% and 6.58%. These improvements underscore the substantial benefits of the framework in both traffic flow efficiency and environmental sustainability, providing critical insights for traffic engineers and urban planners aiming to implement advanced TST strategies in complex urban settings. This study not only enhances understanding of dynamic traffic management but also supports sustainable urban development goals.