Articles

Intelligent Transit: Data Improves Transit Efficiency

Automatic vehicle location and passenger counting systems create data that can fine-tune the running of a transit system.

by / March 27, 2015
The San Diego Metropolitan Transit System used Urban Insights to analyze information from its disparate databases to understand ridership patterns. Flickr/Prayitno

Editor's note: In this five-part series, Government Technology examines the present impact of technology on transit systems and what that can mean for the future of urban transportation.

Mobile ticketing isn’t the only feature available for riders with smartphones. A growing number of agencies (and third-party providers) offer bus and train arrival time via apps. There are different ways agencies can calculate when a bus arrives at a stop, but the most popular and ubiquitous is automatic vehicle location (AVL) technology. 

AVL, part of the constellation of intelligent transportation system (ITS) technologies that have been developed in recent decades, consists of two major components: Onboard GPS that tracks the location of each bus in real time and software that displays the location of the buses on a map. The technology has been a boon for commuters who want to know when the next bus or train will arrive. But it also helps transit managers respond to unplanned service disruptions as well as monitor distance between buses and on-time performance.

Some agencies have coupled AVL data with signal prioritization systems to improve scheduling. In Portland, transit officials decided that in order to make public transportation more reliable (and popular), they would concentrate on the on-time performance of buses that served area schools and hospitals — two places where being on time is critical, according to Esri’s Bills. Signal prioritization technology can synchronize, as well as delay, signal changes for traveling buses to keep them on schedule. By linking AVL data with a signal prioritization system, bus schedules become easier to control, reducing the possibility of a bus arriving too early or too late at a stop next to a school or hospital.

Open Data & Transit

Public transit data is growing exponentially as agencies expand the use of automation to collect information that can improve transit service for both riders and for operations. The problem is that not all agencies are equipped and trained to analyze and use the data to improve performance.
 
One obvious solution is to publish the data so that third-party developers can make use of it for public consumption. More than 200 transit agencies worldwide are publishing their schedules, fares and station locations to Google’s TransitDataFeed in a common data standard and for free, according to Stephen Goldsmith, the former mayor of Indianapolis and professor at Harvard’s Kennedy School of Government (and also a Government Technology columnist). In the United States, a growing number of city transit agencies have set up developer portals and created application programming interfaces to give the public and programmers the ability to create tools and apps that help riders plan trips, receive accurate arrival times for buses and trains, and learn about service updates.  
 
One of the first transit agencies to release its data to the public was Portland’s TriMet, which launched the practice in 2005. Since then, the agency has expanded the number of published data sets and has met with developers several times to explore new types of transit applications. Other open data portal leaders include: Chicago Transit Authority; San Francisco’s Bay Area Rapid Transit; the Metropolitan Transit Authority in New York City; Massachusetts Bay Transit Authority; and the Washington Metropolitan Area Transit Authority.
 
Some members of the American Public Transportation Association believe that open data has triggered more innovation in public transit than any other factor in the last three decades, according to Goldsmith.
 
But open data has its drawbacks. The quality of the tools and applications can diminish considerably if the data isn’t clean, isn’t updated on a timely basis or isn’t available in standardized formats.
Another ITS technology that is helping transit agencies improve service, while helping to control costs, is the automated passenger counting (APC) system. Using infrared sensors, APC records the number of passengers who board and deboard a bus at each stop. APCs are far more accurate than the old-time manual checking systems agencies used to deploy.
 

An APC can help agencies more accurately figure out how many buses they should run on each route, said Chris MacKechnie, a transportation expert and service planner with the Long Beach, Calif., Transit Agency. “The data from APCs can be used to determine schedule adherence and whether bus routes need more or less running time,” he said. 

APC and AVL systems aren’t cheap, however. They can cost between $8,000 and $10,000 — each — per bus, according to MacKechnie. While the feds provide grants to fund ITS technology projects, users tend to be the larger transit agencies, he added.

For those agencies with the funds and resources to automate their fare collection, bus and train routing and passenger monitoring capabilities, the next step is to analyze the reams of data generated by these systems, look for patterns and develop new transit services and strategies. 

Applying analytics to transit data is a new discipline, according to Wade Rosado, director of analytics at Urban Insights, a subsidiary of Cubic Transportation Systems, a global transportation IT company. “You can’t just use existing databases,” he said. “You have to create a new, integrated database.” Done correctly, models can be generated of how people use bus and rail routes and, more importantly, new routes can be designed that best meet the needs of the customers who use transit, Rosado added.

The San Diego Metropolitan Transit System used Urban Insights to analyze information from its disparate databases to understand ridership patterns. The data is complex, involving point-to-point travel times, transfer points, geospatial demographic information, ridership levels and even results from ridership satisfaction surveys. But the deep dive into data and analytics can pay off handsomely. Ridership on San Diego’s bus and trolley lines has increased, service has improved, with improvements in on-time performance and reductions in fare subsidies, according to Esri’s Bills, who has followed the work done by the transit system.

The field of transit data analysis is so leading edge that most agencies rely on firms like Urban Insights, Cambridge Systematics and a handful of others that have the expertise to mine the various types of data. One of the newer entrants to the field is Urban Engines, a firm launched by Stanford University Professor Balaji Prabhakar and Shiva Shivakumar, a former Google executive. Urban Engines emphasizes spatial analysis and behavioral economics to help agencies reduce congestion on transit systems. 

To reduce crowding on bus and train lines, Prabhakar said it’s useful to know how, when and where congestion hot spots occur. For example, a bus route may have a hot spot that affects just 20 percent of the route’s length during rush hour. By analyzing APC information and other data sets, agencies can figure out how to deploy more buses that run along just the most congested portion of a route, rather than the entire route. More efficient use of buses not only helps alleviate congestion, but it can also reduce wear and tear on buses that no longer have to run the entire route, just those sections where there are the most riders.

Next: Look for "Social Media Keeps Transit Agencies Informed About its Riders" on March 30.
Tod Newcombe Senior Editor

With more than 20 years of experience covering state and local government, Tod previously was the editor of Public CIO, e.Republic’s award-winning publication for information technology executives in the public sector. He is now a senior editor for Government Technology and a columnist at Governing magazine.