Crews installed speed restriction signs at Metro-North’s Port Chester station on December 12, 2013. Photo: MTA / Patrick Cashin
When the MTA is in crisis mode, it has shown the ability to respond quickly and efficiently to address issues. For an agency not known for finishing projects on time, adapting new technologies within any reasonable timeframe or generally seeing through innovation, time and again, the MTA tackles major problems better than just about any other governmental agency. Most subway service was restored less than a week after Sandy in 2012, and, more recently, Metro-North and the LIRR instituted sweeping safety improvements barely a week after the first crash with passenger fatalities in Metro-North history.
For this, the MTA deserves some praise. The agency has learned how to make the most out of potentially catastrophic scenarios and has become adept at responding. But the speed with which the MTA addressed some major underlying safety concerns over the past few weeks has raised some eyebrows. Why did it take a fatal crash to implement basic safety upgrades? If these problems were so easy to fix, why weren’t they implemented years ago?
WNYC’s Kate Hinds raised this question on her Twitter account last week, and today, Matt Flegenheimer tackles it in The Times. Here’s an excerpt of his story, and it is not particularly kind:
Since a Metro-North Railroad train derailed in the Bronx on Dec. 1, killing four people and injuring more than 70, the Metropolitan Transportation Authority has not had to look far for safety remedies that might have prevented the crash. Recently ordered improvements, delivered in response to the derailment, have been borrowed from Metro-North’s sister agency, the Long Island Rail Road, and at times from the Metro-North system itself.
Changes significant enough to have thwarted the crash, according to rail experts, were simple enough to have been completed within days. Others are so straightforward that some of the authority’s board members assumed they had been in place for years. While the authority and federal investigators have cautioned that a full accounting of the derailment is not yet complete, many transit officials have arrived at a troubling conclusion since the crash: The authority could have — and in many cases should have — installed a series of protections long before the train’s operator apparently became dazed at the controls early that Sunday morning, racing into a sharp curve at nearly three times the allowable speed…
Asked why broader changes to Metro-North’s signal system were not made sooner — particularly at well-known curves like the one at Spuyten Duyvil — Marjorie Anders, a spokeswoman for the authority, said it was the job responsibility of train operators to “know all the physical characteristics, including the speed limits” in all areas where they were qualified to work. For more than 30 years, she said, “that system worked fine,” with no accident-related passenger fatalities since Metro-North was created in 1983. The recent changes were “a result of the intense introspection currently underway at Metro-North,” she said.
The bulk of Flegenheimer’s explores the reactions to the MTA’s changes. Board members were surprised by the speed at which Metro-North implemented its fixes and were floored to find out that many were rooted in common sense and parallel best-practices in place at the Long Island Rail Road. “The fact that some of the stuff was done in the rebuilding of track that occurred over a couple of days, it does lead you to believe that it could have been done earlier,” William Henderson, head of the MTA’s Permanent Citizens Advisory Committee, said.
Other MTA Board members wondered if an intense focus on on-time performance was to blame while others wondered if the federal government’s singular approach to safety via a federally mandated positive train control system forced the MTA to examine safety issues through a single-lens approach. “I think it may have slowed down the process, actually,” Ira Greenberg said to The Times. “Why put in a system on top of a system that does virtually the same thing, when you can wait for the better one?”
Ultimately, though, the MTA’s response to the crash and the response to the response brings up a tried-and-true problem with the agency — and really any government agency of its size and breadth. The MTA has been essentially reactive for decades now as it has struggled to overcome decades of deferred maintenance and disinvestment in the system. It is not a proactive leader in the global field of transit, and it may never get there. From an area as basic as the fare payment technology to a realm as important as safety, the MTA has not been ahead of the curve, and two weeks ago, it proved quite costly indeed.
11 comments
I think blaming this on run-of-the-mill government inefficiency is letting the MTA off the hook too easy. Americans are used to their government being poorly-run, this is not an inherent characteristic of government-run transit. There are plenty of large government bureaucracies that manage not to be incompetent on safety matters – like, say, any first-world public transit agency outside of the US. Even within the US, airplane safety regulation is quite good.
Re the focus on-time performance, this was also blamed for the Amagasaki crash: Japanese railroads are fascists to their employees about the timetable, so the train driver oversped and led to mass death. However, despite this, Japanese railroads overall are extremely safe. The MTA does nothing of this sort – even at Metro-North, the punctuality is only good by American standards, and is about average by French ones and terrible by Swiss or Japanese ones. The outer New Haven Line trains pad the nonstop Stamford-New York segment by 7-10 minutes, which is 15-24% of the unpadded travel time on that segment; Swiss trains pad 7% and Japanese trains don’t pad at all.
Instead of the operation being spelled “deep dive”, it should be spelled and called “deep duyvil”
Relatively amazing that, in 2013, it took people dying to install speed limit signs. SIGNS. Amazing.
The subways have had “speed limit signs” for decades, it was just that most riders really did not have a clue about reading such signs, or knew where they were, or would pay attention to such signs. The riders for the most part have to depend upon the skill, fitness, and capabilities of the train operators.
My point is that some of the posted speed limit signs are really not for the employees, but rather to re-assure the riding public that “something has been done”. Which is not to say that more steps to make the railroad safer for both riders and workers are not needed.
The basic mantra over the past couple of weeks, has been on the order of, “Our transit personnel are professionals who know the routes, know the speed limits, and they know how to run a railroad”, etc. However recently in public discourse – over time there has been the effort to denigrate and criticize the skills and capabilities of the transit personnel, and other public employees. From efforts and statements such as, “making the running of trains idiot proof”, statements concerning, “over-paid”, “feather-bedding”, etc. Then are there efforts to characterize the “Positive Train Control” technology as “off-the-shelf”, or “an obviously needed improvement”, etc.
While a part of this discussion has an aspect of, “re-assuring the public that rail transit is SAFE”, versus any kind of public message, “We Don’t Know What To Do To Fix This Problem” – which would be a really bad message to send out in any circumstance. The efforts to reassure the public, can lead to statements such as, “Well if the remedy is so obvious, why did you not do it before!” The answer the public might now want to hear is that the remedies – cost a great deal of money, take a good deal of time to install, involve changes to several operations, or might require changes on the part of the public, etc.
A part of the quest to reassure of the public is to offer instant analysis, instant answers to questions – because there is the thought that the public simply can not handle complex issues, have a limited attention span, can not actually wait until the proper analysis has been done, and needs “some kind of display of action NOW!” If possible, let’s create a “villain” somebody to blame for what went wrong. And while we’re at it, blow up the “villain’s” actions as much as possible to direct blame away from the “system” that allowed this “villain” to operate. Then, as usual promise more “oversight” and “management solutions” to reduce the problem, to tide things over. Again, we have to re-assure the public that riding the rails is safe.
Plus the folks who chatter, really need “something” to talk about, and instant news sells newspapers and internet headlines. The history of the aftermath of airplane accidents, train accidents, and other events says that the predictable steps are being followed.
Just my thoughts.
Mike
Despite the fact that MNR and LIRR are under a single agency, the MTA, they do not share any institutional knowledge/safety guidelines. That’s got to change, and this deadly accident may spur further improvements.
But the MTA needs to do more to better integrate that institutional knowledge. The MNR and LIRR continue operating like fiefdoms, and that’s part of the problem.
Deadly accidents unfortunately are what drives safety improvements. Happens in engineering/architecture/construction, the auto industry, and in the railroad business too. Heck, NYC’s DOT is struggling to deal with the deadly curve coming off the Queensboro Bridge that has been the scene of numerous deadly accidents in recent years. To date, the DOT has made some changes, but they’ve been insufficient and the accidents continue to occur with alarming results.
Some of us might argue that Metro-North is, in fact, very proactive when it comes to customer service ingenuity and service offerings, way ahead of most regional railroads (most of which are far more accurately defined as “commuter” railroads, which Metro-North has graduated from/raised itself above.
Concur with lawhawk, though, that institutional knowledge — and, yes, safety guidelines — should be shared between LIRR and Metro-North, and agree further that events like this help spur such changes.
Respectfully disagree with Stephen Smith; I don’t recall NJ Transit being whipped quite so energetically in 1996 by FRA, or anyone else, when three people died in the swamp. NJT managed to go roughly 14 years without a (non-trespasser) fatality before that took place; Metro-North managed to extend the equivalent non-fatality run to 30 years. One more wrinkle to throw in if we’re heaping the blame for things like Con Ed power failures fully upon Metro-North, per convenience, as 2013 ends.
In 1996, the FRA was too busy sabotaging the Acela with Tier II regulations. In 2013, the FRA is about to modernize rail safety regulations and permit EMUs that are lighter than an Abrams tank. 17 years ago, it was the biggest obstacle to rail modernization. Today it’s farther-looking than the agencies that actually run the trains.
Also, Metro-North only looks good when you’re comparing it to other US commuter rail operators. By any non-US standards, it’s deeply deficient, in terms of off-peak frequency on inner stations, stopping patterns, and fare integration. Why do inner New Haven Line trains not make local stops in the Bronx? Why do express trains run nonstop from Harlem to Stamford or White Plains instead of stopping at major stations, like New Rochelle, the way express subways do? Why does it cost more to ride the Harlem Line in the city than to ride the subway, and why do transfers require paying a separate fare? Why is weekend and midday frequency hourly or at best half-hourly? Just because it has a small population of reverse commuters toward Stamford, Greenwich, and White Plains doesn’t mean Metro-North is actually concerned for the needs of off-peak travelers.
” Why does it cost more to ride the Harlem Line in the city than to ride the subway”
Compare the misery and filth of riding the subway with the almost pleasant conditions on MNR and it’s clear where the cost difference comes from.
“… express trains run nonstop…” Because they’re express.
Trains from way out somewhere d-mn well should run express from Stamford. They should also run faster in that segment (I suggest 90 actual with a 100 hard limit, and bypass 125th), but with recent events that’s not likely to happen.
Re the curve at Spuytin Duyvil, the train made it most of the way around at 82 mph. That’s not a 30mph curve.
I say increase the turn banking a couple inches, raise the speed on the curve to 35 at least, (40 is more reasonable) and install some type of signal control which forces a train traveling over 45mph about 1/4 mi before the curve to brake. Adjust in leaf season if need be.
As far as freights, they’ll be fine. There’s no triple stack, not even double on that line. Just to be sure, forbid freights to stop on that curve.
Do the same at all similar locations, and instead of longer run times several minutes per run are saved. Once other track work gets done, a significant number of minutes per run are saved, which might allow the crew a long enough lunch break to get a bit of exercise.
2c on the root cause:
The sitting/standing still required by the job is probably what caused this and many other zone-out moments, with the others not resulting in accidents. This is not the engineer’s fault: ask a human body to do something against it’s design and eventually it will skip a beat, no matter how forcefully your mind pushes it.
Trains already run 90 mph on the New Haven Line, in New York State. In Connecticut the tracks are maintained to 75. None of this changes anything about stopping patterns; these are EMUs, not mile-long freight trains with shit acceleration. Express trains on the subway do make intermediate stops.
As for “the train made it most of the way at 82 mph,” sorry, but no. The train derailed. Speed limits aren’t based on what the train might possibly make. They’re deliberately set well below the safety limit. It doesn’t have to be a 30 mph curve, but even 50 mph may not be possible given issues such as turnout geometry and freight-friendly superelevation. With vanilla superelevation and cant deficiency it’s a 48 mph curve, but again the actual superelevation may have to be a bit lower because of turnout considerations as well as freight needs.
The broader issue I’m complaining about is that Metro-North and the LIRR make no effort to serve city transportation needs. They have too many seats and too little standing space, even on branches that don’t go far into the suburbs, like Hempstead, Far Rockaway, and Long Beach. Even given what they have, which is better than what e.g. the MBTA has, their city schedules suck, and the fares are prohibitive. Zurich’s S-Bahn trains are more comfortable to ride than the buses; the fares are still the same, and transfers are free. The trams remind you of it with the printed slogan, “ein Ticket für alles.” Likewise, in Vancouver, SkyTrain is so much more comfortable than buses, but the fares are still the same and transfers are free. In New York, for all the condescension to the subways, they too are more comfortable than the buses, and several times as fast; again, everything costs the same and transfers are free.
The stop spacing is part of it, and what you’re proposing – skipping East Harlem – would make it worse. Just because commuter rail currently only serves peak-hour CBD commuters, and a few reverse commuters on Metro-North, doesn’t mean that there aren’t other travel markets. People from New Haven travel to Greenwich, New Rochelle, the Bronx, and Harlem, and vice versa. If you ride commuter rail both on weekdays at the peak and on weekends, you’ll be able to notice major differences in which stations are the busiest. Right now, urban service is so crappy nobody bothers, but the intercity service is passable, so New Haven is busier on weekends than any other stop, while commuter suburbs become less important. I contend, based e.g. on observing what times of day I’m the only white person on transit in Philadelphia and what times of day I’m one of several, that inner-urban neighborhoods are also going to be generate much more travel than commuter suburbs on weekends and in the midday off-peak.
Specifically with express service, the point here is that the way the commuter rail operators run it is based on getting people from the suburbs to Grand Central or Penn Station the fastest. The way they should run it, as on the subway, is based on getting people from any station to any station the fastest, with cross-platform transfers at busy stations like Stamford and New Rochelle (which, on weekdays, is the second busiest New Haven Line stop, after Stamford). If Amtrak’s shit-acceleration Regionals can stop at New Rochelle, so can Metro-North EMUs.
These emergencies are useful in puncturing the myth that change is impossible and that the MTA can’t improve or do things more efficiently. Every time someone in Capital said that something is impossible or will take years and billions of dollars, an unfortunate opportunity came up where Operations then went and did this thing that supposedly couldn’t be done, and for much cheaper than projected. This includes the trip-hazard staircase fix in the 36th St station, Sandy recovery, and now Metro-North.
In fact, I would even argue that the MTA shouldn’t have a Capital Budget at all – it’s just a way to launder taxpayer money into contractors’ pockets. As much as possible should be done in-house via the Operations Budget. Large expansion projects can be funded by the city or a city-state Infrastructure Bank with hopefully more transparency and better cost control.