Exploring the correlation between ride-hailing and multimodal transit ridership in toronto

Document Type

Journal Article

Publication Date


Subject Area

mode - bus, mode - subway/metro, mode - tram/light rail, place - north america, ridership - behaviour, ridership - demand, planning - service improvement, planning - service level


Ride-hailing, Public transit, Ridership, Temporal analysis, Level of service


Ride-hailing (RH) services have been growing rapidly and gaining popularity worldwide. However, many transit agencies are experiencing ridership stagnation or even decline. Understanding the correlation between RH trips and transit ridership has become an urgently important matter for transit agencies. This study aimed to explore the relationship between RH and public transit ridership and provide a starting point for future studies. This study benefitted from having access to detailed data on trip-level RH trips, transit supply and transit ridership in Toronto for three years (2016–2018). With this dataset, the study utilized random-effects panel data models and log–log regression models to estimate the correlation of RH pickup/drop-off counts with subway station and surface transit route (buses and streetcars) ridership within transit catchment areas, broken down into five different periods of a non-summer weekday. The results show that RH services generally have a positive association with subway station ridership while negatively correlating with surface transit route ridership. The positive relationship between RH and subway station ridership is the strongest during the mid-day and early evening. In contrast, the negative relationship between surface transit routes and RH ridership is the highest during peak commuting hours. Additionally, RH trip volume is more positively related to ridership at terminal/transfer subway stations in Toronto’s city centre while more negatively associated with routes with relatively poor services (e.g., low on-time performance, low vehicle running speed and low frequency) in the city centre where traffic congestion can be severe. According to the above findings, the degree of the relationship between RH and public transit demand tends to be mixed, varying by transit mode, time of day and transit level-of-service. The gained knowledge about RH and transit can provide insights for transit agencies to improve transit services, which are discussed in this paper.


Permission to publish the abstract has been given by SpringerLink, copyright remains with them.