The parameter calibration of the micro‐traffic simulation model is a prerequisite for the application of micro‐traffic simulation technology. This paper aims at the average queuing length in the simulation model as an index to evaluate the advantages and disadvantages of the simulation model parameters, taking the evening peak signal intersection of Waijiang Road‐Yongkang Road intersection in Yibin City, and an improved genetic algorithm is proposed to calibrate the parameters of the microscopic traffic simulation model, and analyzes the average queuing length in driving behavior using VISSIM software simulation. The actual traffic operation conditions are analyzed and compared, which proves the feasibility of the parameter calibration method of the micro‐traffic simulation model based on the improved genetic algorithm.