Moving America on Transit – Innovation in Real-time Transit Information

(Center Identification Number: 79017-00) Real-time transit information has many benefits for new and existing transit riders.  Workshops with 284 participants, including transit riders and non-riders, hosted by the Federal Transit Administration (FTA) at four transit agencies in Seattle (WA), Salt Lake City (UT), Columbus (OH), and Providence/Kingston (RI) concluded that real-time information attracts new riders who are otherwise reluctant to start using the transit system[1].    Recent research demonstrates that transit riders who have access to real-time information perceive their wait time to be around 30% shorter than riders who do NOT have access to real-time information[2].   This same study also found that real-time information users save almost 2 minutes in waiting time, which is used to perform other tasks.  These benefits have a direct impact on the transit rider’s willingness to use public transportation on a per-trip basis, as concluded from interviews with transit riders in San Francisco and Seattle in 2010. The interviews revealed that when the real-time information system was down, several riders elected not to ride the bus[3].   Riders can also use the information to adjust their own use of the transit system, which can benefit all riders.  For example, one user in the interviews stated that she permanently changed her normal commute trip from an earlier crowded bus to a later less-crowded bus based on real-time capacity information[4].

However, offering real-time transit services to transit riders has its challenges.  The cost for a transit agency to implement both vehicle tracking technologies and information dissemination technologies such as electronic signs or mobile phone applications is not trivial, especially in the public sector where budgets are being slashed.  For example, Carnegie Mellon’s research with the Tiramisu crowd-sourcing mobile application cites the estimated cost for an Automatic Vehicle Location system for a mid-size city as $70 million[5].

Transit agencies can cost-effectively increase information dissemination by openly sharing their data with the public.  The public then has the capability to build transit information services using this data, instead of depending on the transit agency to implement these services themselves.  This model has proven successful at several transit agencies in the U.S., including TriMet in Portland, Massachusetts Bay Transportation Authority, and Metropolitan Transportation Authority in New York, who have chosen to openly release their scheduled and real-time transit data.  The sharing of this data has resulted in the broad dissemination of this information via mobile applications to transit riders, often at minimal cost to the agency.

While the open sharing of transit data has many benefits to both the transit agency and transit rider, there are challenges to “freeing” transit data on an industry-wide basis and fostering a global community of innovation for transit applications.  On January 19^th 2012, the White House Office of Science and Technology Policy (OSTP) convened a roundtable meeting with transit industry stakeholders and third party application developers to discuss the current state of real-time transit applications in the United States.  The goal of the meeting was to advance the current state of practice for open transit data in the U.S. by identifying current barriers and creating cost-effective and timely solutions to overcome those barriers.  This roundtable meeting was led by U.S. Chief Technology Officer Aneesh Chopra and U.S. Department of Transportation (USDOT) Deputy Secretary John Porcari. The meeting was attended by representatives from FTA, USDOT, American Public Transportation Association, Google Transit, Trapeze, Clever Devices, Nextbus, OpenPlans, Chicago Transit Authority, Centre Area Transportation Authority, the Metropolitan Atlanta Rapid Transit Authority, the Metropolitan Transportation Authority in New York, and the National Center for Transit Research at Center for Urban Transportation Research at the University of South Florida (i.e., the principal investigator for this proposed project).

Problem Statement

Two key barriers for large-scale proliferation of mobile real-time transit applications identified in the White House roundtable were:

1)      a lack of a consensus on standards for the exchange of real-time transit data, and

2)      a lack of “clinical trials” of cutting-edge technologies in this area.

Consensus on standards is an agreement within the industry, usually occurring during the evolution of new technologies, on the format and protocols used to exchange real-time data.  “Clinical trials” are the evaluation of candidate technologies that demonstrate how a standard would actually work.   These proof-of-concepts, or “reference implementations,” help stakeholders select and strengthen those technologies that benefit the greater good and are feasible for implementation.  Many standards and specifications look great on paper, but when they are actual implemented in software their weaknesses are quickly apparent.

In the area of mobile transit applications, the General Transit Feed Specification (GTFS) has become a consensus standard for static scheduled transit data, and has been adopted by over 110 transit agencies in the United States.  However, despite the demand for real-time transit information, no similar widely agreed upon standard exists for real-time transit data.  As U.S. Secretary of Transportation Ray LaHood stated in his blog following the White House roundtable meeting, “One more obstacle [to riders having  access to real-time transit information] is the variety of standards used to structure transit data.”[6]

OneBusAway is an open-source transit information software system, including several mobile apps, that was originally developed at the University of Washington and deployed in the Puget Sound area of Washington state.  OneBusAway leverages the GTFS data format for schedule transit data, but the original Puget Sound deployment did not use a standardized interface for sharing real-time transit data with mobile apps.

As part of their real-time BusTime API pilot project in early 2011, Metropolitan Transportation Authority (MTA) in New York leveraged the OneBusAway software to build their own transit information system.  MTA also implemented a modified version of the CEN/TS15531 Service Interface for Real Time Information (SIRI) in their OneBusAway server to share their data with mobile app developers.  The first step in a “clinical trial” using OneBusAway and the modified SIRI technology, has been considered a success by MTA[7], and in 2012 MTA is moving on to the second step of deploying the same technology to other NYC boroughs with expected completion in 2013[8].

OneBusAway has significant promise to reduce the cost of transit information system implementations by allowing agencies to re-use open-source software that has already been proven in other regions.  Additionally, the official CEN SIRI committee has expressed interest in adopting modifications to the SIRI standard, such as that deployed by MTA, which would create the first ratified 3^rd party real-time transit API standard for mobile devices. However, three problems exist:

1)      The MTA OneBusAway source code is currently separate from the original OneBusAway source code used in Puget Sound.  MTA has made many improvements to the original source code, and it is expected that the MTA source code will become the new foundation of the OneBusAway project moving forward.

2)      As a result of #1 above, the original Puget Sound OneBusAway native mobile apps are not currently compatible with the new MTA version of OneBusAway software.  Additionally, the OneBusAway mobile apps that are available on the Apple AppStore, Google Play, and Windows Phone store are currently only available for the Puget Sound region.  As a result, the OneBusAway mobile apps, a major feature of the original OneBusAway project, are not currently readily deployable to other transit agencies that set up the OneBusAway software for their system.

3)      As of early 2012, MTA is the only transit agency to deploy a modified version of SIRI as a real-time transit API.  As a result, there are no open-source mobile application proof-of-concept/reference implementations that demonstrate the ability of a mobile application to successfully interact with the modified version of the SIRI standard.  Open-source mobile apps that use the modified SIRI standard would enable developers to rapidly prototype and design apps that leverage this new standard.

Therefore, additional clinical trials with the OneBusAway technology with other transit agencies are needed to fully evaluate the technology.

Two university transportation centers, NCTR at the University of South Florida and Georgia Tech, will combine efforts on this project.  NCTR will lead efforts prior to and including conducing a clinical trial. Georgia Tech will lead the evaluation of the trial to assess the impacts on the transit system.

Objectives and Supporting Tasks

The project objectives are to:

• Positively impact transit system performance through the provision of real-time information delivered via OneBusAway.
• Increase the number of cities offering open real-time transit information via open-source software systems

This project will build on existing work by MTA, OpenPlans, and OneBusAway contributors to test a deployment of OneBusAway with the modified SIRI standard with an open-source reference implementation of a mobile client application.  Additionally, the existing OneBusAway mobile apps will be improved upon so they can easily be deployed to multiple regional OneBusAway deployments.  This work will use the existing free, open-source transit software from the OneBusAway project (http://www.onebusaway.org/), as well as the open-source software developed by MTA and OpenPlans during the deployment of the MTA BusTime API.  OneBusAway is open-source software, which makes it possible for communities to develop and add new features.

Supporting tasks:

1)      Deploy OneBusAway at a transit agency such as Hillsborough Area Regional Transit (HART) in Tampa, Florida.  HART has a proprietary AVL solution developed by Orbital that was deployed in 2008-2009.

2)      Implement an open-source SIRI mobile client reference implementation, based on existing OneBusAway or OpenTripPlanner mobile applications.

3)      Facilitate the regionalization and standardization of the existing OneBusAway mobile apps, to support multiple metropolitan areas as well as the new OneBusAway software based on the MTA project

4)      Dr. Kari Watkins, with funding from the University Transportation Center at Georgia Tech, will lead the effort to evaluate the impact of real-time information on transit system measures such as but not limited to, perceptions of transit safety and reliability, customer satisfaction, and ridership.

Michael Frumin, Systems Engineering Manager for MTA, Jeff Maki, Principal for the Transportation Group at OpenPlans, Dr. Brian Ferris, Software Engineer for Google Transit and creator of OneBusAway have agreed to serve as a peer-review panel for this project on a time-limited basis.  Dr. Kari Watkins, Assistant Professor at Georgia Tech and collaborator with Dr. Ferris in OneBusAway evaluations, will also serve on the peer-review panel in addition to leading Task #4 above.   Additional members may be added.

At completion of this project, the existing OneBusAway mobile apps will support multiple regions as well as the new OneBusAway software based on the MTA project.  Additionally, the lessons learned from a testing of the OneBusAway system with the modified SIRI standard in a mobile app will be available to inform the CEN SIRI committee, as well as SIRI users.  Finally, an open-source mobile application using the modified SIRI standard will be available as a reference implementation for application developers to use when creating mobile apps that conforms to the modified SIRI standard.  The availability of open-source mobile software using the modified SIRI standard will stimulate innovation for mobile transit apps, as mobile developers will be able to build on this existing resource when creating their own apps.