Unraveling the network effects in station ridership growth patterns under metro network expansion

To address the mobility challenges associated with rapid urbanization, many cities have expanded their metro networks. This requires massive investment, but it does not uniformly translate to ridership growth across stations. The factors influencing station-level ridership dynamics and the underlying network effects are complex and poorly understood. This study thoroughly investigates the station-level ridership growth dynamics under network expansion, utilizing spatiotemporal lag fixed effects models to capture both the supply-side and demand-side network effects as well as temporal dynamics. Utilizing the evolution of Shanghai Metro from 2014 to 2019, we demonstrate that the network effects has a significant and intricate influence on station ridership and its relative growth patterns. On the supply side, neighboring stations show competitive relationships leading to reduced ridership growth, while stations on the same metro line can complement each other. For the whole network, the station betweenness and closeness centralities exhibit divergent impacts on ridership growth rate. On the demand side, we confirm the existence of spatial spillover effects of station ridership, especially at the line level. Additionally, the time-lagged and heterogeneous effects of network expansion are revealed. Our findings provide valuable insights for informed and targeted decisions regarding metro network planning and infrastructure investment.