Transportation systems in New York City are critically important for millions of people. Beyond MTA subways and yellow cabs, commuters and travelers rely on regional commuter rail and the MTA’s bus network to get in and around the city. The past five years has seen bike sharing programs and on-demand transportation burst onto New York’s transportation scene.
These were all topics of conversation at the second Transit Techies NYC Meetup, held at Intersection HQ in Hudson Yards. Approximately 100 Transit Techies gathered to discuss and learn how technology can make these transportation systems better in NYC. After a solid dinner spread featuring dessert from Baked by Melissa, we settled in for four scheduled talks. Here’s a quick recap of the event!
LIRR signal data
Will Fisher from the MTA explained how the Long Island Railroad (LIRR) is working with legacy rail signal systems to better inform staff and customers about current train performance. The LIRR is the busiest commuter rail network in the country and has, over time, accumulated many interconnected systems to manage rail signals. Signals and interlockings are controlled by seven systems, implemented by various vendors at various times.
The LIRR is in a predicament: some of these critical systems are no longer maintained and can’t be modified to add features or extract data. Legacy systems can be difficult to work with and the LIRR has been forced to get creative.
To solve this problem, Will’s team is working on a program that uses optical character recognition (OCR) to glean track data from a real-time graphical user interface (GUI). Then, track segments and interlockings are modeled as nodes in a graph; this graph can then be used to calculate and report performance data, such as expected arrival times.
Destination recommender for on-demand mobility
Assel Dimitriyeva, a graduate student at NYU, presented next about building a dynamic recommendation engine to improve on-demand mobility solutions. Elderly or disabled citizens who cannot access conventional public transit rely on paratransit mobility services (e.g. Access-a-Ride in NYC) to travel around cities. On-demand paratransit services are extremely expensive and making them more efficient could mean affordable mobility for these citizens.
Assel’s project recommends alternative, closer destinations for users of paratransit services using a multi-armed bandit (MAB) approach and Yelp data. Dynamically recommending similar destinations closer to a user can reduce distance traveled and cost. Considering a set of users traveling to a set of destinations, the problem is modeled using MAB. Then, Yelp data provides destinations as input to a contextually-aware recommender engine that solves the MAB problem and recommends relevant alternative destinations.
Bike Inspector by Motivate
Alex Hill, David Bromwich, and Josselin Philippe from Motivate, a major bike share operator in 9 US cities, presented an internal tool called Bike Inspector. Across all programs, including Citi Bike in NYC, Motivate (recently acquired by Lyft) is responsible for 30,000+ bikes, which need to be kept safe and operational for riders. These bikes need to be serviced periodically, or when a rider reports a problem.
Motivate’s bike mechanics then have to service these bikes in the field or their in-house shop; locating, inspecting, and repairing bikes across large cities was a big logistical challenge. Bike Inspector, initially developed by Josselin Philippe while he was a mechanic, provides an intuitive map interface for bike mechanics to use in the field. They can see which docks contain bikes in need of service, diagnose issues, and report repairs all within Bike Inspector.
With Bike Inspector, Motivate can dispatch mechanics more efficiently to docks with bikes that need repairs or inspections. Bike Inspector is in currently in beta in NYC.
Data-driven advocacy for better NYC buses
Mary Buchanan from Transit Center discussed how New Yorkers have joined forces to advocate for better buses in NYC. A quarter of New Yorkers do not live within walking distance of a subway station and rely on buses to get around. Unfortunately, ridership has been steadily declining for years, possibly due to ride sharing services winning customers. As ridership goes down, funding decreases which causes service to worsen, pushing ridership even lower in a vicious cycle.
The Bus Turnaround Coalition, a group of New Yorkers working with Transit Center, are using data to fight against poor bus service. The Coalition uses real-time bus data from the MTA to compute speed, on-time performance and headway for bus lines in NYC. They use these metrics to generate report cards on the bus system, assigning a letter grade (A-F) to each bus line. According to their methodology, 183 of the 246 (74%) bus routes get either a D or F grade.
If you’d like to support the Bus Turnaround Coalition, learn more about helping out the #BusTurnaround campaign here.
If you enjoyed this recap, please 👏 this article below. Please follow my Medium profile for recaps of future transit and sustainability Meetups!
Check out my recap of the first Transit Techies NYC Meetup: