Examining the determinants of microtransit use and ridesharing based on trip microdata

Microtransit is a technology-driven public transportation service that accommodates on-demand trip requests and dynamically adjusts its routing based on real-time origin–destination patterns. It provides shared rides with flexible pick-up and drop-off locations and incorporates trip scheduling and fare payment in a smartphone application. Although many jurisdictions in North America have initiated microtransit pilots in recent years, our understanding of these systems remains limited due to the lack of empirical research. This study identifies a methodological framework to investigate the neighborhood sociodemographic, built environment, and other spatial and temporal attributes associated with higher public microtransit use based on trip microdata. As microtransit demand tends to be sparse, resulting in excess zeros in the dependent variable, hurdle negative binomial models are selected as a suitable approach for explaining microtransit passenger pick-ups and drop-offs at a small spatiotemporal resolution. This research also empirically explores the factors significantly influencing ridesharing in public microtransit. Although microtransit vehicles are designed to serve multiple passengers simultaneously, a substantial portion of microtransit trips is single-passenger in practice. A binary logit model is estimated to identify how system-level and local microtransit demand, trip characteristics, and the built environment contribute to the probability of a microtransit trip being shared with another booking. The microtransit system with the highest ridership in North Carolina, operating in the City of Wilson—a community with a large transit-dependent population—serves as our empirical example. Our findings indicate that neighborhoods with lower-income, carless, African American, and younger populations generate greater demand for microtransit. We also find that a larger female population is associated with more microtransit trips, potentially because women are less likely to have a driver’s license or access to personal vehicles, especially in vehicle-deficit households. On average, the likelihood of ridesharing increases for longer trips and when the service experiences high demand but decreases for trips that require a wheelchair-accessible vehicle. Urban areas and local employment and commercial centers are associated with more trips as well as a higher probability of ridesharing.