Nonlinear, threshold and synergistic effects of first/last-mile facilities on metro ridership

Understanding the effects of intermodal connectivity on metro ridership is essential for policymakers and urban planners to facilitate metro usage. Although many studies have examined the associations between the built environment and metro ridership, few investigated the effects of first/last-mile facilities. Using smartcard data and multi-source big data in Shanghai, this study applies LightGBM approach to estimate the nonlinear, threshold and synergistic effects of last-mile facilities on metro ridership, while controlling for built environment characteristics and metro network attributes. Results show that last-mile facility variables collectively contribute 43.57% for predicting metro ridership. Nonlinear associations illustrate that most last-mile facilities need to reach certain thresholds to maximize metro ridership. Furthermore, bus supply and built environment factors may moderate the effects of last-mile facilities on metro ridership in a multiplex way. These research findings help urban planners identify effective ranges of different last-mile facilities and prioritize intermodal connectivity strategies in planning practice