The U.S. has nearly 19,000 cities and towns with populations of 50,000 or less. In many of these small communities, bus systems are sparse, infrequent and inefficient — if they exist at all.
Microtransit — a shared, technology-enabled, public transit system with flexible routing and pick-up/drop-off locations that accommodates on-demand trip requests — has emerged as a promising solution for connecting carless, disadvantaged individuals to employment, health care and other important destinations. Demand for microtransit is rising in small communities, but adding more vehicles and drivers to meet the demand is cost-prohibitive for public agencies.
A team of researchers from NC State, Cornell University and the Georgia Institute of Technology, led by CCEE Assistant Professor Eleni Bardaka, was awarded a $1.8 million grant from the National Science Foundation (NSF) Smart & Connected Communities Program to develop, test and evaluate smart, technologically enabled, and community-supported solutions for distributing travel demand over time and increasing ridesharing and efficiency in public microtransit systems in an equitable manner.
The research team will develop a Cooperative Adaptive Ride-Sharing system (CARS) that creates empathy-building messaging based on real-time user information and powered by artificial intelligence (AI), to motivate prosocial behavior for the benefit of other users and the system as a whole, such as encouraging users to shift the time of more flexible trips (e.g. errands) to accommodate other users’ urgent and critical trips (e.g. work), and walking farther, when able, to share rides.
The envisioned microtransit system will be piloted in Wilson, North Carolina, a city with a high proportion of economically disadvantaged and carless households. Wilson instituted a microtransit system three years ago, but the volume of trip requests is so high that not all trips can be accommodated.
“A disabled woman from Wilson described trying to book a ride after her doctor’s appointment unsuccessfully for more than an hour; she ended up traveling five miles in her wheelchair to get home late in the evening,” Bardaka said. “Traditionally, pricing would be used to regulate demand for travel. But when most users are carless and low-income, increasing the fare would hurt those that need microtransit the most. We therefore had to come up with an approach that moves away from pricing and monetary incentives.”
In Wilson, researchers will work with local governmental and nongovernmental organizations to help build community engagement and understanding of the program and associate technology.
“Small cities and towns like Wilson have a strong sense of community and empathy between their members, something that we witnessed in our interactions with the community,” Bardaka said. “We hope that our research will lead to improved and more equitable microtransit systems.”
The research team is led by PI Bardaka. Co-PIs include Munindar Singh (Department of Computer Science), Christopher Mayhorn (Department of Psychology), Crystal Lee (College of Education) and Samitha Samaranayake (Cornell University). Pascal Van Hentenryck (Georgia Tech) and Kai Monast (NC State Institute of Transportation Research and Education) serve as Senior Personnel.