Multi-objective Optimization Method for Signal Control of Single Intersection

Currently, in China, the single-objective optimization methods are commonly used for road intersection signal control, which has one single objective such as minimizing the number of stops, average delays, or queuing length. When the intersection traffic varies large with the time of day, using this method for intersection traffic control will always cause poor adaptability and low efficiency. To solve these problems, a multi-objective optimization method for signal control of single intersection is proposed, with the composite optimization objectives to minimize the average delay, the least number of stops, and the shortest vehicle queue length, which can set preferences of various optimization objectives to adapt to different traffic conditions. The method uses the Pareto multi-objective evolutionary algorithm to optimize signal timing parameters to get the best signal control plan. Simulation results show that this method can significantly improve the adaptability and efficiency of the intersection traffic control.