Assessing complementary and competing interactions between transit and shared transportation modes

Abstract

Shared transportation modes have been widely introduced in cities over recent years. However, their interactions with existing transit systems, i.e., whether they complement or compete with them, are still unclear. This paper addresses this issue by exploring the relationships between two shared modes (bikesharing and free-floating carsharing) and traditional public transit (subway and bus) using several passive data streams from Montreal. A novel terminology is first proposed to define different types of interactions based on their impacts on the ridership of each mode in the short or long term. Building on this conceptual framework, a rule-based algorithm is developed to classify individual bikesharing and carsharing trips into distinct groups of potential complementary or competing relationships with transit. The spatial and temporal distributions of trips in different categories are then analyzed. Finally, the causal impacts of competing bikesharing trips on daily route-level transit ridership are assessed using a fixed effects difference-in-differences model. The results reveal that daily ridership on transit lines that would have been used in the absence of bikesharing may have been reduced by 1.3 % for every 100 bikesharing trips. However, the shared modes can also complement transit in time (when the service is closed), space (outside the service area or in the first/last mile connectivity), or when the service is unsatisfying (less direct or slower). Competition is most evident in the city center, at peak times, but can also help to relieve the most congested transit lines, and thus turn into a positive interaction in the long term.