Adam Pearce
After years of weekend closures, the MTA announced the completion of the 7 train’s new, computerized signal system yesterday.
Replacing finicky pneumatic electromechanical devices installed in the 50s, the modernized signals increase the number of trains the MTA can run on the 7 line, reduce “signal problems ahead,” and lower maintenance costs. Over the next 10 years, the MTA plans to spend $20 billion upgrading every line’s signals.
Past experience suggests this be will difficult.
In their efforts to fix the subway, the MTA has addressed some of the issues that caused problems in the past. They’re buying train cars that work with computer signals without retrofitting, a previous source of delays. The new head of New York City Transit, Andy Byford, has proposed shutting down all weekend and night travel for years while a line’s signals are upgraded; the intermittent shutdowns on the 7 meant that additional days of necessary trackwork could add a month to the project timeline.
But there are still problems.
Plagued by software issues, the last two years of work on the 7 line took four years to finish. This isn’t a new problem for the MTA. On a previous signal project, ad hoc deployment errors caused interruptions to train service. A vaguely specified voice communications system failed in the field. UIs that had only been prototyped in PowerPoint were hard to use and had to be redone.
Byford says he wants to buy “off-the-shelf” to cut costs and reduce delays. His previous job as CEO of the Toronto Transit Commission provides a clue about what that’ll entail.
When Toronto decided to update its 50 year old signals, they kept the old electromechanical signals in place and overlaid the computerized signals. This is hard to do: the computerized signals need heavily customized software that simulates how the old signals and interlocks work. And the amount of track work increases; instead of just putting down beacons that let the train car know its exact position, the old signals also require extensive modifications.
Faced with mounting delays, Byford simplified. Instead of keeping the old signals, almost everything was computerized.
In its signal modernization efforts, the MTA is still overlaying rebuilt parts of the old signal system as a failsafe. Overlaying old signals accounts for a significant chunk of the project timeline on the signal modernization project currently underway on the Queens Boulevard Line:
A Federal Transit Administration report found that keeping the old signals isn’t necessary and increased capital costs by at least 30%. The safety benefit is limited. In use, two sets of signals can actually decrease reliability because service is interpreted if one of them fails. Keeping the old signals working also requires regularly sending workers onto the tracks. Many of the upsides of computerized signals are negated.
Under the current plan, each line will require years of weekend and overnight closures to overlay the computerized signals on the old signals. Even if the project stays within its scoped budget, this signal modernization will cost a huge amount of money. Already overloaded with debt and weighing service cuts and fare hikes, the MTA has limited fiscal room to maneuver if there are steep cost overruns.
Will Byford continue to overlay two sets of signals or will he decide to simplify? The activation of the new signals on the 7 yesterday could impact the decision. There were extensive delays during the evening commute after a component of the old signal system failed.
Nov. 27, 2018