The scientists tracked anonymized information from more than 680,000 Boston-area drivers over three weeks. The engineers used commuters’ cellphone signals to locate the position of individual drivers. The study identifies the origin or “home” districts of the drivers according to the hours and locations of the calls. For the purposes of the study, a trip is occurring when the same mobile phone user is observed in two distinct zones within one hour.
MIT and UC Berkeley researchers used cellphone signals to create what they claim is the most detailed study of urban traffic patterns ever conducted. They found that commuters from a handful of specific areas—including Everett, Marlborough, and Lowell—have an outsized effect on congestion across Greater Boston.
In rush-hour traffic, a single driver blocking the box can ruin everyone’s afternoon. But when you’re sitting on a backed-up road, cars stretching into the distance, the source of the gridlock is more of a mystery: Where’s the problem coming from?
What’s true at a single intersection turns out to apply on a larger scale: It’s just a few drivers, relatively speaking, who jam things up for the rest of us.
That’s the conclusion of a team of engineers from the Massachusetts Institute of Technology and the University of California, Berkeley, who have produced one of the most detailed maps of urban traffic patterns ever constructed. They did it by analyzing the cellphone records of 680,000 Boston-area commuters (they perform a similar analysis for the Bay Area). The call logs—which identify the towers used to transmit calls—allowed the researchers to trace each individual’s commute, anonymously, from origin to destination.
The new technique represents a leap forward in traffic engineering. In the past, transportation officials in cities have used pen and paper surveys to study traffic patterns. These surveys are hard to administer and typically are only carried out once every few years. And while the data they produce are useful, they only capture the driving patterns of a small number of people over a short period of time. The cellphone records provided the engineers with a much more comprehensive picture of mass commuting patterns.
What they found, perhaps surprisingly, is that during rush hour, 98 percent of roads in the Boston area were in fact below traffic capacity, while just 2 percent of roads had more cars on them than they could handle. These congested roads included short stretches of I-495 southbound and Route 128 southbound, a number of downtown streets, and a wide scattering of suburban arteries, such as Bridge Street in Lowell (southbound) and Water Street in Haverhill (northbound). Each of these roads has what the engineers term a high degree of “betweenness”—that is, they’re essential for connecting one part of the metropolitan area to the others.
The backups on these roads ripple outward, causing traffic to snarl across the Hub. Marta Gonzalez of MIT, one of the lead engineers on the study, explains the effect this way. “The analogy we make is of your circulatory system,” she says. “When you have one artery that is blocked, it will affect your entire circulation.”
By tracking the cell records, they found that it’s just a small number of drivers from a small number of neighborhoods who are responsible for tying up the key roads. Specifically, they identified 15 census tracts (out of the 750 in Greater Boston) located in Everett, Marlborough, Lawrence, Lowell, and Waltham as the heart of the problem, because drivers from those areas make particularly intensive use of the problematic roads in the system.
Besides letting us point the finger of blame, the research suggests a potential new approach to alleviating traffic: focus on those problem neighborhoods. The engineers calculate that if you were to reduce the number of car commuters from those 15 areas just 1 percent, all drivers across Boston would get home about 18 percent faster each night.