AI-driven method to reduce traffic delays and improve road safety

Dr. Abolfazl Karimpour, Assistant Professor of Transportation Engineering at SUNY Poly, has developed an innovative framework that estimates the length and duration of traffic queues and delays caused by crashes, without relying on physical roadside sensors. Dr. Karimpour is lead author of the new article published in Case Studies on Transport Policy.

By integrating real-time vehicle speed and location data from widely available crowdsourced sources, this method enables consistent, statewide monitoring of crash impacts at a fraction of the cost of traditional approaches. In practical terms, this research equips transportation agencies with a powerful tool to detect and respond to incidents more quickly, better manage congestion, and improve roadway safety for drivers.

This recent publication was co-authored with recent SUNY Poly graduate Anthony Alteri, Adrian Cottam from Auburn University’s Transportation Research Institute, and Ellwood Hanrahan II from the New York State Department of Transportation (NYSDOT).

Conducted through SUNY Poly’s Transportation AI Research Lab (TRAIL), where Dr. Karimpour serves as director, the project benefited greatly from NYSDOT’s collaboration. The agency provided critical transportation data, contributed to brainstorming sessions, and offered key insights that helped shape the research direction and outcomes.