Algorithms can lead to more efficient public transit. Here’s how you may already be helping. by Greg St. Martin February 7, 2019 Share Mastodon Facebook LinkedIn Twitter The Green Line stops at Northeastern on July 17, 2017. Photo by Adam Glanzman/Northeastern University Riders of Boston’s public transportation system use the Transit app to plan trips, get real-time information on service, and set alerts for departure times. These interactions provide data about how riders use the app and navigate the transit system, known as the MBTA. But little has been done with this treasure trove of information in the two years since the MBTA formed a partnership with Transit in 2016—until now. Daniel T. O’Brien, an associate professor at Northeastern, is leading a team of researchers that will spend the next two years poring over the data, organizing it into a much more accessible database, and leading studies intended to help the MBTA better understand its ridership. “Every interaction a user has with the app gives us a snapshot in time of how someone is choosing to navigate that public transit system,” O’Brien said. The researchers will work to shed greater light on how riders navigate the MBTA system, how satisfied they are with its service, how their travel plans change when routes close or snowstorms strike, and how well-equipped the MBTA is to respond to these disruptions. The researchers will use data from the Transit app to build algorithms and provide analyses that the MBTA can implement to improve service and refine its policies. O’Brien said that his team may also explore whether customer satisfaction surveys from MBTA riders and data records from CharlieCards, which riders use to board trains and buses, could be used to help develop the systems it builds for the MBTA to track and forecast service. Their findings will have implications beyond Boston’s transportation system. The Transit app is used in more than 175 cities worldwide, and analyses resulting from this work can help inform how the app is being used more broadly, O’Brien said. The Massachusetts Department of Transportation funded the project by awarding a two-year, $300,000 grant to the Boston Area Research Initiative, which is based at Northeastern’s School of Public Policy and Urban Affairs. “We’re excited to partner with the Boston Area Research Initiative in analyzing the wealth of data available via the Transit app,” said MBTA Deputy General Manager Jeff Gonneville. “We’re always looking for ways to better understand how our customers use the system. We’ve endorsed the Transit app as the best way for customers to plan their trips, and the opportunity to better understand how our customers get to their homes and jobs and loved ones through analyzing Transit’s data will give us a unique perspective when we consider service issues and future projects.” Daniel T. O’Brien is leading a team of researchers that will spend the next two years conducting studies intended to help the MBTA better understand its ridership. Photo by Adam Glanzman/Northeastern University O’Brien, who co-directs the Boston Area Research Initiative, said the data could reveal much more than is currently known about the habits of MBTA riders. While records from CharlieCards reveal only where riders enter the MBTA system, riders can use the Transit app to plot their full trips, with a starting point and a destination. O’Brien said that the research team could, for example, analyze the data to see how riders reacted each time Red Line service on the MBTA was suspended over the past two years. Did they all flock to a certain bus line, or look up how long it would take for an Uber to arrive to pick them up? O’Brien said the research team could build algorithms that the MBTA could run in real time to proactively respond to, and possibly even anticipate, such disruptions. In order to learn more about customer satisfaction, the Northeastern research team and the MBTA have discussed the idea of surveying riders who use the Transit app and comparing their attitudes toward the MBTA with how transit service performed on those riders’ routes. “This project will offer brand new insight into how people navigate the public transit system, and it’s more than just a one-time academic study,” O’Brien said. “The insights we produce, which themselves are novel, can be repeated and implemented consistently and incorporated into the MBTA system.” For media inquiries, please contact Mike Woeste at firstname.lastname@example.org or 617.373.5718.