Modeling the effect of real-time crowding information (RTCI) on passenger distribution in trains

Overcrowding has become a big challenge for public transport systems, affecting passengers' travel experience. At the same time, service supply is often underutilized due to large variations in crowding across services, vehicle trips on the same service and different compartments of the same vehicle. Real-time operational measures, such as information provision, can potentially reduce on-board crowding unevenness and its negative effects. In this study, we extend a dynamic public transport simulation model to provide passengers with predictive real-time crowding information (RTCI) concerning individual train cars. Passengers utilize this information when choosing a specific train car to board. It is demonstrated through a case study for the Stockholm metro network area that in the presence of car-specific crowding information, passengers alter their car boarding choices to avoid on-board crowding, leading to a more even passenger distribution inside trains. We find that passengers' travel experience improves with the provisioning of RTCI, which is a result of the lower on-board crowding unevenness. Moreover, this improvement increases with increased demand levels but only up to a certain point beyond which passengers do not gain from switching train cars.

Automated summary

Overcrowding is a major challenge for public transport systems, affecting passengers' travel experience. Providing real-time crowding information about individual train cars can help passengers make informed decisions to avoid crowded cars, leading to a more even distribution of passengers inside trains. This improvement in travel experience increases with demand levels but reaches a saturation point.