Analysis of the Impact of Small-Scale Operational Disruptions on the Travel Behavior of Subway Commuters

Urban rail transit, which carries a huge volume of passenger transport in megacities, is the main artery of urban transportation. However, disruptions, especially those lasting more than 30 min, can lead to severe operational disruptions. Currently, large crowds of passengers gather, trains are frequently delayed, and the quality of service drops significantly. However, few studies have explored the impact of disruptions on passenger behavior, with most relying on manually collected survey data. This study aims to explore the potential for a comprehensive analysis of the impact of disruptions using automated data collection. As a result, a systematic approach was proposed to assess the impact of disruptions in urban rail transit systems on passenger responses using AFC data. We have proposed an inference method for quantifying commuter passengers' responses to interruption events, depending on their travel characteristics and the location and time when the interruption events occurred. The method proposed in this paper is demonstrated based on AFC data from the Beijing urban rail transit interruption event and normal AFC data. The case study reveals that local disruptions in the Beijing metro have a significant spillover effect on passenger flows across the entire network, and that commuters exhibit behavior responses with clear spatiotemporal dependence.