Modeling and prediction of passenger boardings at transfer nodes of public transport networks

Abstract

Public transport transfer nodes represent challenging elements of a network when it comes to predicting vehicle occupancies, due to the high number of boarding and alighting passengers and the related uncertainties that are associated to these stops. For this reason, understanding the passenger dynamics at these stops can give a contribute in enhancing public transport occupancy predictions. This work presents a methodology to model the passenger boardings at transfer stops, which considers both the passengers that reach the stop on their own from the external (e.g., by foot or bicycle) and passengers that reach the stop on board a public transport vehicle. The proposed methodology is applied to a case study related to one of the major transfer stops of the city of Zurich. Results could be used to improve boarding predictions and to enhance the understanding of the passenger transfers at such stops.