Back in May, at a talk with reporters, MTA CEO and Chairman Jay Walder talked, as he often does, about his experiences in London. To accomplish work efficiently, Transport for London often shuts down service across entire lines for much of a weekend, and the MTA has been debating doing that here. Walder mentioned it to me again during my interview with him, and at yesterday’s Transit Committee meeting, NYCT head Tom Prendergast broached the subject as well.
The MTA, it seems, may be revamping the way they approach weekend work. “Maybe for some of the more difficult tests, that take a long time to set up, pick one Saturday a month and do it at night, starting after five or six o’clock at night, look at things like that,” Prendergast said.
On the one hand, this will inconvenience thousands of subway riders, but on the other hand, weekend service changes already do that. If the MTA can minimize the frequency of the changes, blitzing a line could pay off in the long run. Meanwhile, in response to the findings that crews falsified signal inspections, Transit may need to increase the time and manpower required to conduct a thorough review of the system’s numerous signals. Weekend work and changes just make get worse before it gets better.
28 comments
I don’t understand how these proposed changes are any different from what happens every non-December weekend anyway. Forget five or six on Saturday: lines go entire weekends with no service. What is Walder talking about?
As I understand it, no reroutings and no partial shuttering of lines. So instead of “The F is running on the A line from High Street to 34th Street” or whatever, you’d get “No F train this Saturday. Take a bus.”
But what does that accomplish? In most of Queens, the F shares tracks with the E, and where it doesn’t, it’s a four track line, where two tracks (presumably either the two express tracks or one express an done local track) can be inspected while trains operate on the other two. The route through 63rd has 53rd as an alternate (again, preferably only one direction at a time). Once on 6th, the D also runs, and, again, closing the two local tracks doesn’t make sense where closing one local and one express track is an option. Between West 4th and Jay, trains can run via the A in one direction. South of Jay, especially with the work currently taking place there, shutdowns are sometimes required, but even they usually only go as far as Church. South of Church, there’s a third track, so shutdowns can usually be handled by one-way bypasses or, if necessary, by diversions to one of the other Coney Island lines.
Remember that operating buses is much more expensive, on a per-rider basis, than operating trains. It only makes sense to operate buses where there’s no feasible alternative. In particular, four-track lines – which tend to be the busiest lines, where very frequent bus service would be required – don’t usually require complete shutdowns. Also, unless NYCT is willing to purchase additional buses and hire additional bus operators to support weekend shutdowns, there’s an absolute limit on how many shuttle buses can run at any one time.
A full shutdown is only necessary on two-track lines, or on three-track lines where work has to take place on at least two tracks simultaneously, or on four-track lines where work has to take place on at least three tracks simultaneously, AND there is no alternative route.
I assume that they would only consider full shutdowns in the situations you describe – not when there’s another set of tracks available that they could easily use.
Also, by “take a bus” I think he meant one of the normal buses – there would be no special replacement shuttle. So no added expense there.
But full shutdowns are already done in those situations.
The normal buses wouldn’t be able to handle more than a small fraction of the regular riders of the closed subway line (and forget about the regular bus riders!).
I don’t know what it accomplishes. This is simply the impression I get from what the MTA has been saying on the matter.
The CTA did that on the Green line for a total rehab job & it was a disaster.
CTA’s problem was they shut the entire line down for 2 whole years, cut a few stops out altogether all the while some of the adjacent neighborhoods on the near south and west sides of Chicago continued to empty out, now at a faster rate with easy transit access having been removed. Others got used to new routings. I used to live out near the Austin Blvd stop and wound up car pooling for a full year after the line reopened mostly out of convenience. A two year closure is highly disruptive. The CTA learned that and rehabbed the Douglas Branch of the Blue Line (now Pink) with just weekend closures and did the Brown line expansion with intermittent extended station closures and occasional weekend work.
Sadly, reforming the worst-run major transit agency in the world by learning from the second worst-run agency would be an improvement.
How does PATH maintain 24/7 service? I cannot recall of any bus substitutions for PATH ever.
Same way the floors in Path stations are kept clean. It’s known in other places as doing one’s job.
Any transit agency can be improved by not being the MTA.
Virtually every transit agency – PATH and NYCT being two of the few exceptions – shuts its entire rail system every night.
Would that be an improvement?
“Would that be an improvement?”
Absolutely not.
PATH, NYCT, PATCO, two lines of the Chicago L, and the Copenhagen Metro. The Copenhagen Metro uses automatic train operation to operate on single track at night, allowing regular maintenance.
The Copenhagen Metro is brand new, and it was presumably built with enough crossovers to allow for single-tracking at a reasonable headway.
That doesn’t apply on most of the two-track lines here. (It will apply on SAS, I believe.)
I’ve read it’s more the automation than the crossovers.
It’s both. The travel time through the single-track segment has to be short enough to maintain a reasonable headway. Automation can help by running the trains quickly and consistently, but the crossovers still have to be close enough.
If you have any connection to construction work the story is pretty obvious. Catch a eyeful of the overnight workers where visible to the public (such as from a train on the opposite track) then figure what’s going on when you can’t see.
Work done between trains is incredibly inefficient – not by accident, it’s a way of increasing employment. That made sense in the ’30s, but not now. Work done where live tracks encumber the process is also slow, though the details can vary greatly.
There’s an obvious workaround which MTA does not use, perhaps due to a few fatal accidents a while back and perhaps due to nearby trains generating enough sound to interfere with work. Still, it is not hard to:
Shut one side of line for several miles.
Put up a temporary barrier between the live tracks and work area. (I’ve seen this done where work is expected to continue for several weeks). Trains run normally in one direction and express (middle track) in the other.
You now cannot lean/trip/be pulled by a heavy object you are wrestling with into the path of a train or onto its 3rd rail.
Work could, in theory, now proceed at a normal pace rather than having 2 people watching a 3rd work. A lot of people will need to be fired, and the city might have to manage through another strike for this to happen.
The typical overnight GO runs from midnight to 5 AM. How long does it take to put up these barriers and then to take them back down again? How much time is left for the actual work? What sort of flagging is required while the barrier is going up and coming down?
Doing the shutdowns on weekends in Manhattan really shouldn’t be a major hassle south of 59th on all but the Lex, since the other lines are bunched pretty closely together (and even the Lex for most of its run north of Grand Central is bi-level, meaning one level could be shut and worked on while the other continues operations without endangering the workers). The bigger hassles will be shutting the branch lines in the outer boroughs, especially the Queens Blvd. line.
The M shuttle was shutdown for at least 3 months worth of weekends this past year just to replace the station platform edges. It was a originally supposed to be a 1 month of weekend closures. And then that seemed to have done — or they discovered the platform edges were in much worse shape. So then we endured it again in October and November. Striking to me how long that project took and how little we got of it. The stations along the M in Brooklyn are in horrible shape. It would have been nice to use those closings to do some needed leaky roof repairs for instance.
I’d like to suggest that if MTA pursues this they think of combining projects into more extensive renewal efforts both on the tracks and inside the stations.
NYCT shifted from full-blown station rehabs to component replacements a year or two ago.
I guess you preferred it the old way.
I can see both sides of this. No one wants “their” service impacted.
But I would honestly think that shutting down completely would be much easier to finish what ever project in a more timely manner than what we have been getting.
The system is still in pretty bad shape. And a lot of the improvements will require overhauls. So a way to get this down quickly is of high importance. And if there can be some cost savings by shutting down as opposed to rerouting or slow order or the like, i think i’d be all for it.
No one would have a problem with repeated closures of entire lines if:
1) The lines were replaced with bus services that provided replacement service for THE ENTIRE LINE. When the L was out of commission every weekend a couple years ago, you often had to take a bus to the J, take the J to Essex, then transfer to the M14A for the rest of your trip. That’s absurd, but it was often the only option. L train riders would have been very happy had the bus stopped at every L station, then went over the Williamsburg Bridge, up to 14th and over to 8th Ave, without the time consuming and annoying double transfer to ridiculously slow local services. You can see examples every weekend of how the MTA only provides the shuttle buses
2) They use enough buses that each bus is not overflowing with people. There is nothing worse than getting on a nasty shuttle bus in the middle of winter when everyone is in their big coats covered with snow, the bus gets wet, and you have to rub up against everyone. It’s very unpleasant.
That’s completely unnecessary!
Whenever the L is out of commission into Manhattan, the M runs into Manhattan (instead of being cut back to Myrtle, as it usually is on weekends). That’s especially useful now that the M runs uptown: you can take the M directly to 14th and 6th, or you can transfer from either the M or the J to another line which will get you closer to your destination.
The only people who need the M14A are people going to 3rd and to 1st, including those who were planning to transfer to the M14A at 1st anyway.
Running a bus over the bridge and back up to 14th would be extremely costly, and it wouldn’t be very popular, since it would be slower than taking the train.
I agree that shuttle buses are often overcrowded. Sometimes I think it’s a matter of inadequate dispatching – buses sent out empty just as a train is pulling in, or buses being stacked up for blocks and blocks at the terminal instead of being sent out on a short headway – rather than inadequate scheduling.
I remember the 7 line being shut down with no service whatsoever on weekends between 61 St and Flushing. They installed an interlocking at Roosevelt and other things.
The Port Washington Branch paralleled the line and shuttle service was added to it between Flushing and Penn with no fares being checked on trains between those two stations.
http://www.mta.info/lirr/News/2008/Shuttle.htm
They also finished one weekend early, as I recall.
The general rule of thumb is that a line has to be shut down if work is taking place on all or on all but one of its tracks. (The exceptions are if the interlocking and signal layout can accommodate single-tracking at a suitable headway or if there’s an alternate route that trains can use to bypass the closure in one direction, such as Sea Beach vs. West End.)
An interlocking connects at least two tracks, and some of the work on installing a new interlocking will typically require a simultaneous shutdown of both tracks. Since that part of the 7 is a three-track line, only one track is left, at best. Therefore, it was shut down.