Investigating the relationship between built environment and spatiotemporal heterogeneity of metro ridership

A growing body of research highlights the importance of understanding built environment factors influencing metro ridership to shape future transit strategies. However, previous traditional approaches relying on total or average ridership across distinct periods (peak hours and off-peak hours, weekdays and weekends) fail to capture the dynamic temporal variations of metro stations. This study addresses this gap by evaluating metro ridership patterns based on a 24-hour cycle, allowing for a more nuanced understanding of the effect of built environment characteristics on ridership patterns. Focusing on Ankara, Türkiye, the study identified the spatiotemporal heterogeneity of metro ridership using large-scale Google Maps data and analyzed network-based catchment areas (CAs) using street and building vectors and points of interest (POIs). Employing principal component analysis and the k-means algorithm, four distinct daily ridership patterns with unique temporal evolutions were identified: diurnal, nocturnal, low equilibrium, and peaked equilibrium. Subsequently, interactive multinomial logistic regression was utilized to assess the impact of built environment metrics on these patterns. The results revealed significant interactions between built environment features and daily ridership patterns at the metro station level. Notably, the interaction of commercial POIs with entropy was found to increase the likelihood of diurnal and nocturnal ridership patterns. Conversely, the lack of these factors increased the probability of low equilibrium patterns. Residential population characteristics emerged as a more potent determinant of ridership than land use density or diversity, highlighting the importance of demographic considerations in transit planning. Moreover, conventional accessibility metrics, including betweenness and closeness centrality, were found to be insufficient in ensuring consistent ridership. The different influencing mechanisms of various types of metro ridership highlighted the importance of interactive relationships from a micro perspective to create balanced and dynamic metro CAs. This study offers crucial insights for creating more sustainable and efficient urban transportation systems.