Causality between multi-scale built environment and rail transit ridership in Beijing and Tokyo

Abstract

Recent years have witnessed researchers' academic interests in the relationship between the built environment and rail transit ridership. However, few studies have analyzed it from a beyond-station scale and causal perspective. To fill this gap, this study uses Beijing and Tokyo as cases and utilizes Bayesian networks and generalized propensity score matching to achieve causal discovery and causal inference of built environment and ridership. Moreover, we innovatively consider the line-scale built environment factors. Evidence suggests that transit agencies should promote the overall accessibility of stations along the line to employment concentration areas more than single-station accessibility in Beijing. Meanwhile, the positive impact of bus-rail transit line cooperation on ridership is limited in Beijing. Planners should consider the capacity of rail transit lines. The causal inference results are mainly in line with previous correlation studies, but some interesting differences exist. For example, the net effect of dense pedestrian roads on ridership is negative, reminding policymakers that they should avoid expanding pedestrian paths without developing ridership-attracting resources. This study uses causal theories to emphasize the necessity of corridor-based and modest development.