Exploring spatiotemporal dynamic of metro ridership and the influence of built environment factors at the station level: A case study of Nanjing, China

Abstract

While previous research has examined the impact of built environment factors on urban rail transit (URT) station ridership, there has been limited focus on the spatiotemporal heterogeneity of these effects and the variation in ridership patterns. To address this gap, this study investigates the spatiotemporal relationship between URT station ridership and the built environment by integrating smart card data (SCD) with point of interest (POI) data. Using the Nanjing rail transit system in China as a case study, we apply the Gaussian Mixture Model (GMM) to classify URT stations based on hourly ridership data over ten consecutive weekdays. The Multiscale Geographically Weighted Regression (MGWR) model is then used to explore how the built environment influences various ridership patterns, including daily totals, peak hours (AM/PM), and boarding/alighting across different station types. The results show that GMM outperforms the K-means clustering algorithm, identifying six distinct station types. Furthermore, the MGWR model demonstrates greater reliability than both the Ordinary Least Squares (OLS) and Geographically Weighted Regression (GWR) models. It reveals that the variables influencing ridership differ between MGWR models, with single-function stations being affected by distinct factors, while mixed-use stations share common influencing variables. Moreover, the relative explanatory power analysis reveals that land use typically demonstrates the strongest explanatory power across most station types, while specific factors, such as residential amenities, office facilities, and shopping facilities, play dominant roles across different station types and time periods. This study contributes to understanding spatial variations in the relationship between station-level ridership and the built environment, offering empirical evidence for context-specific transit and land-use planning.