A Parameters Calibration Method in Simulated Complex Traffic Network

Traffic simulation models have been extensively used because of their ability to model the dynamic stochastic nature of transportation systems. Parameter calibration is very complex and does not give optimal results easily. Besides, it is also time-consuming especially for large and complex networks. Initially, the procedure of traffic micro-simulation parameter calibration was put forward. A Vehicle Intelligent Simulation Software Model (VISSIM) models were selected for parameter calibration in complex-network, and, the role of Simultaneous Perturbation Genetic Algorithm (SPGA) was examined in the optimization of component. Moreover an automatic calibration methodology for micro-simulation models was developed in order to select the best parameter set based on the observed Intelligent Transportation Systems (ITS) data which proved effective for different networks. Finally, the methodology was applied to calibrate the Beijing city VISSIM models, followed by the comparison of convergence rate of Genetic Algorithm (GA), Simultaneous Perturbation Stochastic Approximation (SPSA) and SPGA algorithm. The results show that the SPGA was effective and had good performance.