Cornell researchers are helping to transform portions of the Chattanooga, Tennessee, transit system into a seamless, artificial intelligence-powered network where buses, shuttles, electric cars and bikes work together to provide the most efficient routes - all at the push of a button.
The project, "AI-Powered Autonomy-Aware Neighborhood Mobility Zones: Equitable Solutions and Business Models for Revamping Transportation," is being led by the Chattanooga Area Regional Transportation Authority and funded with a $3.2 million grant from the U.S. Department of Energy.
The goal is to create neighborhood "mobility zones" - designated areas where residents will have access to a variety of transportation options, including fixed-route buses, on-demand shuttles, electric vehicle shares and bike shares. Advanced AI will dynamically recommend the most efficient routes and modes of travel through a mobile app, reducing reliance on personal vehicles.
With only 1.6% of trips in Chattanooga currently made via public transit, according to project officials, the mobility zones seek to increase this figure to 5%, while also reducing per-person transit energy consumption by 10%.
Cornell will contribute to the project on both the demand side (by studying user behavior) and supply side (through operational planning and optimization). On the demand side, Cornell is leading the development of an advanced choice-based recommender system designed to optimize transportation decisions. The research is led by Ricardo Daziano, professor of civil and environmental engineering in Cornell Engineering and an expert in transportation economics.
The system will analyze traveler preferences, constraints and real-time transit conditions to provide personalized route and mode recommendations, enhancing efficiency and user experience.
"By integrating advanced economic modeling with AI-driven planning, we can systematically identify the right set of incentives to drive engagement and align transit design with individuals' preferences and needs," said Daziano, a principal investigator on the project. "Truly understanding what motivates users enables us to implement targeted recommendations that create a more seamless, equitable and energy-efficient mobility system, ultimately enhancing public transit adoption."
On the supply side, service design and operational algorithms will be developed jointly by researchers at Cornell, Vanderbilt University and Pennsylvania State University. The Cornell team is led by Samitha Samaranayake, associate professor of civil and environmental engineering (Cornell Engineering), whose research focuses on optimizing urban mobility networks and is an expert on multi-modal transit systems.
By combining AI, economic modeling and transit optimization, the team aims to create a scalable model that can be replicated in other cities facing similar transportation challenges.
"The transportation sector has seen significant innovation in the past decade, but the primary focus has been on personal vehicles. Unfortunately, this alone will not lead us to cities that have less congestion and are more sustainable and equitable," said Samaranayake, co-principal investigator on the project. "Achieving this goal requires a fundamental shift towards more transit-oriented technologies."
The initiative will launch in select Chattanooga neighborhoods, while actively engaging the community to ensure local residents have a voice in shaping a transit system that meets their needs.
Consulting firm Spark is also a collaborator on the project.
Syl Kacapyr is associate director of marketing and communications for Cornell Engineering.
