There is a significant correlation between the built environment and the ridership characteristics of urban rail transit. However, while the majority of studies explore the impact of the built environment on ridership at transit stations, limited research explores its effect on station travel time. Ridership reflects the intensity of transit use, whereas travel time indicates its operational efficiency. Using an interpretable machine learning model, this study investigates the nonlinear influence and threshold effects of the built environment on station travel time. The case study of Tianjin reveals that: (1) the spatial distribution of station travel time throughout the entire day and during the morning peak is similar, presenting a U-shaped pattern with higher values at the periphery and lower values in the center; (2) the built environment’s contribution to whole-day and morning peak travel time is similar in magnitude, with closeness centrality, Point of Interest (POI) mix degree, distance from the city center, betweenness centrality, and administrative office facilities POI density being the most influential factors; (3) there exists significant variance in the nonlinear relationships between built environment aspects and station travel time, where closeness centrality, POI mix degree, and betweenness centrality negatively correlate with travel time, while distance from the city center shows a positive correlation. These findings offer strategic guidance for improving built environments through targeted renewal efforts and optimizing rail transit network planning to enhance ridership efficiency comprehensively.