Development of Real-Time Transit Delay Map Visualization Tool

The signal system on NYC Transit’s “B” Division (with the exception of CBTC on L line) was installed in phases during the Great Depression and currently in the process of modernization and would take a few more years. All signals operated on relay logic and it was impossible for the dispatchers in the Operations Control Center (OCC) to monitor the whereabouts of each train or their schedule identity in real time. Mr. Reddy led an interdisciplinary team of signal specialists, programmers and analysts to develop and deploy a system to visualize the real-time train locations with a user-friendly interface. The system kept reading the logical state of relay circuits, and sent track occupancy information to a central server where it was combined with the real-time vehicle data generated by terminal dispatchers, and finally the result was displayed in the web interface for the operators and decision makers as a measurement. This system along with track schematic graphical user interface (GUI) was deployed at two Master Towers where it led to an immediate improvement in dispatcher actions and incident management. It was also deployed at the OCC to enable better management monitoring and for the delivery of accurate real-time train arrival information to customers. Currently it has been used by over 100 personnel from service delivery, signal departments, and as well help digital customer communications teams identify where to put out public delay alerts.

Key Features

• Integrates real-time train ID data from electronic train register system, track occupancy data from the signal system, schedule data from the scheduling system, algorithmically infers train locations and displays them on a system map.
• Applies analytics to the incoming data stream and automatically identifies areas where service gaps or traffic congestion exists. Prioritizes display based on severity.
• User-configurable view to show only specific lines, districts, etc.
• Configurable database of track layouts—nothing is hard coded.
• Incorporates the Service Intervention Recommendation Engine (SIRE), which assesses current locations and movement patterns of trains, factors in typical ridership patterns, traincrew and vehicle turns, and recommends dispatcher actions (e.g., holds, skips).
• Provides Time-space diagram view, a real-time graphical representation of train-movements. It compares movements to schedules and linking slowdowns with associated incidents.
• Easy web-based user interface that can run on any web capable computer.  Currently installed on overhead monitors in Master Towers utilizing commodity PCs with corporate Ethernet connection.

In addition to the Transit Viz map visualization tool, Mr. Reddy also led his team to design and implement other innovation tools for the dispatchers to improve the service efficiency. The heat map was developed for the dispatchers to see detailed and tailored train by train activity for various analyses or real time service adjustments. Another tool, the stringline chart, was developed to compare the movements to schedules and linking slowdowns with associated incidents.

Related Project

• Service Intervention Recommendation Engine

Related Publications

• Lehmann, S., A.V. Reddy, C. Samsundar, and T. Huynh.  Automated Train Identification and Train Position Monitoring at New York City Transit. Transportation Research Record (2020), In Press.
• Wood, D., T. Stasko, L. Tarte, D. Jefferson, A.V. Reddy.  A Real-Time Service Management Decision Support System for Train Dispatching at New York City Transit.  Transportation Research Record 2672, pp.327-338 (2018).