Abstract: Based on the inherent characteristics of intersection,the hybrid optimization model was established through considering both phase control delay and queue length. Without changing conditions of the intersection inherent phase, cycle length and phase in the effective green time real time dynamic optimized and configured in order to advance utilization rate and operation efficiency of the intersection temporal and spatial resources. Then, genetic algorithm was used to achieve numerical simulation of hybrid optimization model. The analysis of the simulation results show that, hybrid optimization model can achieve the real-time dynamic optimization on different traffic demands and states of intersection signal timing under constraint condition, while the optimized cycle length and queue number are relatively stable. Those above prove the rationality of the model and the effectiveness and stability of control. Finally, characteristics of different traffic demand control models are studied, and the effective basis was provided for further improvement of the model.

Key words: engineering of communication and transportation system, signalized intersection, hybrid optimization, delay, genetic algorithm, traffic simulation