In 2013, with real-time train arrival data becoming widely available through its general transit feed specification real-time (GTFS-RT) feed, New York City Transit (NYCT) recognized the potential for new service management tools capable of on-the-fly performance visualization and reporting. This paper describes the process taken by NYCT to develop and evaluate a transit analysis tool that uses information from automatic train supervision countdown clocks, transmitted as GTFS-RT data, to visualize train spacing, movement, and related parameters. The main part of this tool is a new web application that uses stringline (time-distance) charts for monitoring operations in the system as they happen. Operational problems (e.g., delays, train bunching, or gaps in service) can be identified more easily on stringline charts than on the traditional model board display to help managers and operating personnel deliver continuous service performance improvements. The application was built in-house with the use of existing data resources (the GTFS-RT feed of predicted train arrivals) and open-source tools. In-house development minimized cost and allowed for maximum flexibility to add features in the future. The development team successfully employed a methodology of iterative development to incrementally review, add features, and test the application without having to go through very long design and requirements cycles. As more transit agencies release GTFS-RT data to the public, they may see opportunities to take a similar approach to develop new tools for internal use quickly and with little additional cost.
Transportation Research Record Journal of the Transportation Research Board
2538(1):19-29
DOI:10.3141/2538-03
