Road transportation contributes significantly to global and urban greenhouse gas emissions, with intersections in cities posing a particular challenge. At these junctions, pollution levels can be as much as 29 times higher than on open roads. Half of the emissions at intersections result from vehicles accelerating after coming to a stop. While some stop-and-go traffic is inevitable, a portion of it can be prevented by optimizing the timing of traffic lights.

Enter Green Light, an initiative from Google Research, leverages artificial intelligence and Google Maps driving trends to model traffic patterns and offer recommendations for enhancing existing traffic light plans.

Traditional approaches to improving traffic light timing involve expensive hardware installations or manual vehicle counts, both of which are costly and often do not provide comprehensive data. Green Light harnesses AI and Google Maps driving data, benefiting from a deep understanding of global road networks, to create intelligent suggestions for city traffic engineers to improve traffic flow. Initial findings suggest the potential for up to a 30% reduction in stops and a 10% decrease in greenhouse gas emissions.

City engineers can implement these recommendations in as little as five minutes using the existing infrastructure. By optimizing not only individual intersections but also coordinating across multiple nearby intersections to create synchronized flows of green lights, cities can enhance traffic flow and further reduce emissions resulting from stop-and-go traffic.

This technology is currently operational in 70 intersections across 12 cities on four continents, spanning from Haifa, Israel, to Bangalore, India, to Hamburg, Germany. In these intersections, Green Light has the capacity to save fuel and reduce emissions for up to 30 million car journeys every month.