Study on the distribution of passenger ridership intensity and network efficiency at metro stations considering land factors

Currently, research in rail transit passenger flow primarily focuses on prediction, with limited attention to the mechanisms of how land use affects passenger flow. Furthermore, studies on network efficiency in this context rarely consider both passenger-weighted and land-weighted factors. This paper integrates subway passenger Automatic Fare Collection (AFC) data and collects Point of Interest (POI) data for the areas around stations. It defines multi-level impact zones centered around stations and utilizes station clustering algorithms along with a multivariate logistic regression model to investigate the influencing mechanisms o f passenger flow distribution about land use. Additionally, the study constructs a Space L-weighted subway network that takes into account land, passenger, and locational factors. It proposes network efficiency assessment metrics that balance physical topological features and the impact of land use. Through complex network attack simulations, 14 critical stations with lower weighted network efficiency and higher passenger loss rates are identified. This research offers insights into urban rail transit network land use planning and station layout construction.