How will passengers make the joint choice of departure time, trip-chaining, and travel mode under disruption of metro service?

Understanding passengers’ behaviour during metro disruption can help operators adopt remedial strategies. Passengers will re-plan the departure time, trip-chaining, and travel modes upon getting disruption information. Previous literature has limited research on this behaviour. Hence, we conducted a two-week survey during the COVID-19 pandemic, collecting 3,400 samples to analyze the joint choice behaviour of passengers in Chongqing, China, regarding departure time, trip-chaining, and travel modes. The 12 Cross-Nested Logit (CNL) models were established to explore the passengers’ joint choice behaviour in 12 different disruption scenarios. Analysis results reveal that on weekdays, passengers tend to choose immediate departure, simple trip-chaining, and bridging buses. On weekends and holidays, passengers prefer immediate departure when the disruption duration is less than 60 min. Whereas when the disruption duration exceeds 60 min, passengers are more inclined to postpone departure time, which suggests that operators need to pay more attention to the potential risk of an increased number of hindered passengers in these scenarios. The CNL structure analysis shows that the nest-specific coefficient for “trip-chaining” is the smallest in the weekday scenarios, indicating a high correlation between simple trip-chaining and complex trip-chaining. The results also imply that when exogenous variables change, passengers primarily adjust their departure times and travel modes. Notably, in the 12 models, the frequency of travel cost and waiting time being significant variables is higher than that of transfer counts, indicating that passengers are more sensitive to travel cost and waiting time. And elasticity analysis also shows that passengers are more sensitive to the waiting time for bridging buses. This study can help planners adopt effective strategies to maintain the reliability and sustainability of transportation systems.