Abstract: Sprocket system is a power transmission device widely used on a variety of industrial equipment, it has importantly theoretical and practical significance to optimize and improve the performance of such system. In this paper, an optimization method with optimization process of sprocket system based on experimental design of response surface model is proposed to overcome the shortcomings of both optimization efficiency and optimization accuracy in current sprocket system optimization. Based on the finite element analysis results of the sprocket system, the optimum hypercube design for the optimal control parameters affecting the sprocket system is presented. The design control parameters with significant influence are obtained and the interactive responses between the control parameters are established. The response surface approximation model of the sprocket system is constructed based on the parameters of the control samples. The multi-island genetic algorithm is used to globally optimize the response surface approximation model and optimize the sprocket system. Compared with the existing optimization methods, this method further improves the overall optimization performance of the sprocket system, and shows that the method has good feasibility and practicability.