Earth observation satellite system (EOSS) is the main space platform collecting ground information. Optimization of EOSS is difficult, as it is a complex system referring a great deal of design variables and uncertain factors. Therefore, an optimization framework based on design of experiment and surrogate model is proposed. Design of experiment is used to generate simulation plan, which will greatly cut down cost of simulation. Then surrogate model is built to analyze simulation data and approximate real EOSS. Genetic algorithm and improved general pattern search method are adopted to solve the model. According to the framework, a case study is carried out. The final results illustrate the framework is useful and effective for the problem of EOSS optimization.