A new algorithm developed by researchers at MIT greatly improves upon existing traffic systems that cities currently have in place and could lead to a future with no traffic jams, better fuel economy for cars, buses and other vehicles and, as a result, less air pollution.
The system uses both city-level data and individual driver data to create traffic light patterns that keep traffic moving. Right now, traffic systems in cities mainly focus on major roads and intersections. That leaves out important information like how drivers behave when traffic jams occur. If drivers take alternate routes, that affects the flow on other roads where new jams can happen without smart traffic systems extending to them as well.
"What we do," Professor of Civil and Environmental Engineering Carolina Osorio says, "is develop algorithms that allow major transportation agencies to use high-resolution models of traffic to solve optimization problems."
MIT reports, "Typically, such timing determinations are set to optimize travel times along selected major arteries, but are not sophisticated enough to take into account the complex interactions among all streets in a city. In addition, current models do not assess the mix of vehicles on the road at a given time—so they can't predict how changes in traffic flow may affect overall fuel use and emissions."
This new system takes into account all the vehicle types from cars to buses and also calculates the impact on fuel efficiency and pollution.
The team has been testing their technology using the city of Lausanne, Switzerland, simulating the driving behavior of the thousands of vehicles that hit the road each day. Through testing they were able to find the right balance of city-level data and more fine driver-level data that could actually lead to better suggestions from the software. Those suggestions lead to improved travel times across the entire city in the models, with 17 key intersections and 12,000 vehicles included in the study.
Having the fuel efficiency and emissions data calculated by the software means that agencies using the program can clearly show the benefits to new traffic management ideas before any changes are made. This means new ideas will have a greater likelihood of being implemented and a greater chance at success.
The researchers are now focusing their energies on Manhattan and other traffic-plagued cities to see how the system works on a large-scale. Ultimately, the software could help city planners with more than just traffic lights. It could also find optimal spots for bike sharing racks, car-sharing hubs, better bike lanes and pedestrian bridges.