Abstract: Park green space is an important part of urban public service facilities, whose spatial distribution is closely related to the life of urban residents. According to the status of the layout, this paper takes 2020 as the optimal time node and takes into account both the efficiency and the equilibrium of layout. First, this paper takes the minimization of residents travel cost as the efficiency goal, and constructs the location optimization model to locate the newly added park green spaces. Then we take the minimization of accessibility difference coefficient as the equilibrium goal, and construct the area optimization model to determine the appropriate scale of park green spaces. The two models have jointly realized the layout optimization of the park green spaces. The conclusions of this study show that: (1) Under the present conditions, the layout of the park green spaces cannot meet the basic recreation needs of the residents, and the spatial efficiency and equilibrium are quite low.(2) To achieve the goal of the optimization, we need to add 28 urban park green spaces（1 100 hectares）and 34 regional park green spaces（868.01 hectares）.(3)The location model can significantly reduce travel cost of the residents to park green spaces, and the per capita travel cost can be reduced from 1 311.24 to 915.93 m. (4) The location and area optimization model can significantly improve the equilibrium of park green spaces’ layout,  making it balanced with the population of residents.The accessibility difference coefficient of the residents in different regions to the park green spaces has been reduced from 0.71 to 0.54.