Using cell phone data to beat rush hour

Scientists at MIT recently used cell phone data to track traffic in congested areas of the San Francisco bay area and Boston, Massachusetts. 

Using cell data from 1 million anonymous users, the researchers were able to reconstruct traffic information for the areas over a period of 3 weeks. They found that a surprisingly small number of traffic sources contribute to make the congestion worse at rush hour. They estimate that cancelling only 1 percent of trips in these zones could reduce time spent in traffic jams by other road users by up to 18 percent. 

“This has an analogy in many other flows in networks,” says Marta González, who led the study at MIT’s Department of Civil and Environmental Engineering. 

“Being able to detect and then release the congestion in the most affected arteries improves the functioning of the entire coronary system.”

In Boston, cancelling 1 percent of trips by some drivers travelling from the neighbourhoods of Everett, Marlborough, Lawrence, Lowell and Waltham would cut all drivers’ time spent waiting in traffic by 18 percent. In the San Francisco area, cancelling trips by drivers from Dublin, Hayward, San Jose, San Rafael and parts of San Ramon would cut 14 percent from the travel time of other drivers.

González and her partner Pu Wang, worked out drivers’ home neighbourhoods based on the regularity they travelled on particular routes and from the location of cell phone towers that handled calls made between 9 pm and 6 am. From this and estimated traffic volume and speed on traffic routes they were able to pinpoint the specific neighbourhoods that contributed most to traffic jams at rush hour. 

“These percentages are averages based on a one-hour commute with additional minutes caused by congestion,” says Wang. 

“The drivers stuck in the roads with worst congestion would see the greatest percentage of time savings, because the selective strategy can more efficiently decrease the traffic flows in congested roads.”

This is the first time cell phone data has been used in a large scale traffic study. Previous studies relied on survey data which is much less accurate because of the time lag between trips and when they are reported by users.

As almost 5 billion people are expected to live in urban areas by 2013, using cell phone data to help solve traffic congestion problems appears to be the way forward. González and Wang are currently looking at anonymous cell data from other countries, including Dominican Republic, France, Portugal, Rwanda and Spain. Their current study is published in Nature’s Scientific Reports.