The simulation-optimization method was adopted to optimize the design of groundwater pollution monitoring network in this paper. The optimization model aimed to maximize the coverage of high-polluting areas and considering the transport of pollutants at all times. In order to reduce the computational load, the Kriging method was used to construct the surrogate model of the simulation model. In the optimization process, the surrogate model can be used to replace the relationship between input and output of simulation model. Finally, to assess the fitting accuracy of the surrogate model and the performance of the optimization, a hypothetical contaminated site was taken as a case study. The results showed that the mean relative error of the output between surrogate model and simulation model was less than 0.5%, which was a high fitting accuracy. The optimal detection rate of the pollutant was 3.37mg/L, and the detection rate was 85%, which was much higher than that of the random layout scheme. It indicated that the method can achieve the target of maximize the coverage of high-polluting areas with a small amount of computation. This study provided a stable and reliable method for the groundwater monitoring wells network design.