The connection between the built environment and the passenger volume at urban rail transit stations has been extensively explored in previous research; however, limited research addresses the factors influencing station ridership across different time periods. Using multi-source geospatial data, we constructed a “5D + N” built environment indicator system. Ordinary Least Squares regression was applied to identify significant factors affecting station traffic at different times, while Quantile Regression was utilized to examine how the impact of these built environment factors varies. A case study of Tianjin reveals that significant factors influencing station traffic vary by time period, with fluctuations in the degree of built environment impact. This methodology offers insights for developing time-specific and zone-based guidance strategies. The findings hold strategic value for improving the operational efficiency of urban rail transit and for revitalizing and optimizing the built environments surrounding stations.