AbstractUrban parks have an important impact on the health of urban residents. Normally, the first step is to identify whether the accessibility of urban parks is equitable; however, multiple models that were applied in previous studies have shown different accuracies. In this study, the area within the fifth ring road of Beijing municipality is selected as the location to conduct the accessibility-based equity assessment of urban parks. The urban park accessibility (UPA) of each residential community will be calculated using the Gaussian two-step floating catchment area (G2SFCA) method. Then, random forest regression and partial dependence plots will be carried out to analyze the relationship between UPA and socioeconomic indicators. To identify where the UPA is inequitable, spatial distribution characteristics between the UPA and socioeconomic indicators will be analyzed based on the bivariate treatment of local indicators of spatial association (LISA). The results indicate the close relationships between UPA and four indicators (towns or subdistricts location = LT, bachelor’s degree (or above) = BD, mean house price = HP, and population density = PD). Beijing urban park unfairness exists in low BD with low UPA or low HP with low UPA streets in the southwestern and east areas of Beijing. In addition, unfairness exists in high PD with low UPA streets in the north second to third ring road areas. Therefore, these areas should be prioritized for urban park development. The methods used in this study could be extended to study the relationship between various service facilities accessibility and influencing factors, and the results could guide intervention programs to help reduce urban park inequity.