Part 1: Strengthening Rural Transportation Through Community and Regional Strategies
Speaker: Abhay Lidbe
Abstract: Rural communities across the Southeastern United States face persistent challenges in accessing reliable, and efficient transportation. Dispersed populations, limited public transit, and growing economic and healthcare needs make mobility a pressing issue that affects quality of life, workforce participation, and community well-being. This presentation introduces a set of innovative approaches designed to address these challenges through both community-driven and regionally coordinated strategies. One approach explores how local residents, organizations, and institutions can leverage their own resources—such as volunteer drivers, shared vehicles, and nonprofit partnerships—to build grassroots mobility systems that are sustainable and responsive to local needs. The other examines how technology, collaboration, and system integration can connect existing services into a seamless regional network, reducing inefficiencies and expanding access across county lines. By combining local participation with system-wide coordination, these approaches move beyond incremental improvements and offer a vision of rural mobility that is adaptable, comprehensive, and future-ready. The goal is to ensure that residents with limited transportation options are not left behind but instead become active participants in shaping solutions that expand access and livable opportunities.
Part 2: Modeling Ride-Hailing Demand for Any Census Tract in the United States Using Open Data
Speaker: Greg Erhardt
Abstract: In this research, we assessed the expected efficiency and affordability of automated vehicle (AV) passenger services, such as Waymo and Cruise, in rural communities. Our methodology involved developing and applying an innovative model capable of predicting the human-driven ride-hailing demand (Uber and Lyft) for every Census tract in the United States. We assessed the potential for AVs to reduce the cost of providing equivalent service, and predicting how the demand would change in response to cost reductions. We previously developed the predictive model, and via a CR2C2 grant, we quickly scaled it up for large-scale application and tested its spatial transferability against observed data in rural Massachusetts. Validating the model against observed ride-hailing data in Massachusetts shows that it reasonably estimates demand in rural counties, although under-estimates demand in urban counties. These results suggest that the model is capable of providing a reasonable first-order estimate of ride-hailing demand in the vast majority of the United States where data are not available.
By applying the model to a series of scenarios in Kentucky, we estimated that only about 2% of ride-hailing trips in Kentucky originate in rural areas. Further, we found that cutting the fares to half of their current level would result in approximately 50% more trips, and cutting the fares to a quarter of their current level would result in approximately twice as many trips. The modest number of rural trips also opens an opportunity to directly subsidize rural ride-hailing trips by drawing revenue from the much larger pool of urban trips, which are already well-served by ride-hailing. Such a scheme applied to human-driven vehicles could be more cost-effective than fully developing AV technology for rural areas.