Understanding in-station mobility gaps in metro systems: Evidence from smart card data on passengers with reduced mobility

As urban populations age and accessibility concerns intensify, understanding the in-station experience of passengers with reduced mobility (PRM) has become increasingly important for equitable mass rapid transit (MRT) planning and operations. While prior studies have largely focused on station access, limited attention has been paid to mobility within station buildings, where design and operational factors directly affect passenger movement. This study addresses that gap by analyzing in-station dwell time using smart card transaction data from the Taipei Metro, with a particular focus on PRM passengers. A modified passenger-to-train assignment model (PTAM) is developed to decompose total dwell time into access, transfer, and egress components. Applied to the Green Line of Taipei Metro and validated using holdout data, the model is followed by multilevel regression analysis to assess how dwell time varies across traveler groups and station paths. Results show that, on average, PRM passengers experience 26.0 % longer access times, 12.9 % longer egress times, and 10.7 % longer transfer times compared with regular passengers, after controlling for station physical size, gate passenger volume, station design and accessibility features, and train headways. The probability of boarding the first arriving train was also consistently lower for PRM passengers, particularly during peak periods at congested stations. These disparities are most pronounced in spatially complex or crowded stations, suggesting targeted interventions such as optimizing transfer routes, managing peak-period passenger flows, and adjusting train dispatching strategies. By linking detailed behavioral insights to station- and system-level interventions, the study contributes a scalable, data-driven framework for inclusive transit planning and policy development.