Congestion-Aware Genetic Algorithm for Urban Mobility Optimization in Intelligent Transportation Systems

Urban traffic congestion increases travel time, fuel consumption, and emissions. Different traditional traffic routing algorithms, including that of Dijkstra, are prone to failure in dynamic congestion as they use the static shortest paths. The proposed congestion aware Genetic Algorithm (GA) routing model in Intelligent Transportation System (ITS) proposed in this research paper is in which the route fitness is a compounded measurement of the travel time, the intersection delay, and the amount of congestion. In SUMO, the strategy is implemented and experimented using traffic records of Bethlehem City. The proposed method decreases the average travel time and intersection waiting time by up to 25 and 20 percent respectively relative to the Dijkstra routing conditions on the same demand conditions. Also, a sensitivity analysis analyzes the strength of the model when there are different levels of congestion. The results indicate the usefulness of AI-based traffic optimization in generating scalable, adaptive, and effective solutions to transportation in urban areas.