Open gangways are a standard way to increase capacity without running more trains. (Photo by Benjamin Kabak)
To say the subways are crowded these days is to state the obvious. Average weekday ridership hit 5,650,610 last year, up by 9.5 percent since 2010, and despite constant service changes and complicated re-routes, combined weekend ridership is up by nearly 11 percent over the same time period. As daily rides where we all stand shoved against people and doors and poles trying to find some amount of space attest, the trains are bursting at the seams.
In today’s Times, Emma Fitzsimmons explores the overcrowded subway system. Her focus is generally on safety concerns, and although the overcrowding is a symptom of larger funding issues and lack of general support for transit investment, we shouldn’t be so quick to dismiss safety concerns. After all, even if people aren’t falling off subway platforms, if riders feel unsafe, it will affect how they view and use the transit system. Thus, at key choke points — on the Lexington Ave. line, narrow platforms at Bleecker St. and Union Square and an overall lack of space at Grand Central come to mind, people worry about safety whether or not reality reflects those fears. (The numbers do not show any uptick in crowd-related injuries.)
In terms of solutions, Fitzsimmons offers up a glimpse at an agency struggling for more space:
But the subway infrastructure has not kept pace [with increasing ridership], and that has left the system with a litany of needs, many of them essential to maintaining current service or accommodating the increased ridership. The authority’s board recently approved $14.2 billion for the subways as part of a $29.5 billion, five-year capital spending plan.
On the busiest lines, like the 7, L and Q, officials say the agency is already running as many trains as it can during the morning rush. Crowds are appearing on nights and weekends, too, and the authority is adding more trains at those times. The long-awaited opening of the Second Avenue subway on the Upper East Side this year will ease congestion on the Lexington Avenue line. Installing a modern signal system, which would allow more trains to run, is many years away for most lines.
When the MTA’s timelines are put forward in terms of “years” and “decades” and even something as simple as a 2-mile subway extension takes 10 years to build, relief is not exactly on the horizon. Yet, from where I sit, there are at least three steps the MTA should take immediately to address capacity concerns.
1. Open Gangways. One of the biggest missed opportunities of the last 15 years involves the MTA’s rolling stock designs. Since 2000, the MTA has seen nearly 4000 new subway cars enter service, and none of them were designed with open gangways, a feature standard in subway rolling stock throughout the world. As I wrote last year, open gangways can lead to a 10 percent increase in capacity without adding a single extra trainset, and while the MTA in 2013 acknowledged the need for articulated trains, the upcoming R211 order includes just one ten-car prototype. It’s not clear if the MTA has the option to add more open gangway trainsets to the R211 order, but not doing so would be a costly mistake for decades to come. This generation of rolling stock is likely to be in service until the late 2060s or early 2070s, and missing the opportunity to expand capacity now will burn us for generations to come.
2. Speed up CBTC installation. This is of course easier said that done, but recent reports have shown how it will take the MTA decades to fully modernize the signal system. CBTC would allow for modest increases in capacity, and prioritizing these efforts — whether through full line shutdowns over concentrated periods of time or other initiatives — should be an agency priority.
3. Just run more trains. As Fitzsimmons detailed, the MTA says it can’t run more trains on perennially crowded lines. For some, that’s due to routing choices — the Q is chock full of choke-points — and for others, such as the L, terminal capacity constraints come into play. Part of the MTA’s capital plan should involve expanding capacity through investments such as tail tracks at 8th Ave. and other minor upgrades that can net big results. For lines that aren’t maxed out, the MTA should just run more trains. But there’s a catch: An aggressive rolling stock retirement plan and a delayed Bombardier order has left the agency tight on available trainsets. Thus, just running more trains, in the short term, isn’t a practical solution even if it is the more obvious answer. Meanwhile, trains are operating at slower speeds, especially along crowded routes, and that too limits the agency’s ability to run more trains and clear out crowds.
Where we go from here isn’t particularly clear. You’re not in danger of falling into the tracks due to crowded platforms, and the MTA doesn’t need to resort to temporary platform closures as London does or subway pushers as Tokyo does. But relief isn’t exactly around the corner. Crowded commutes with packed cars running later into the evening and earlier in the morning are just a way of life until the MTA has the funds available to engage in an aggressive push to increase capacity. For now, though, we ride as we always do: crammed into a subway car, hoping for the best.
85 comments
Open gangways throughout the R211 order seems like such a no brainer to achieve a 10% capacity bump decades ahead of any other solution on the horizon. Any idea what the hesitation is on the part of the MTA?
Also, has any system had experience retrofitting open gangways to existing rolling stock?
Agency officials have expressed some concerns about tight curves and engineering difficulties. It’s not an intractable problem though.
You’d think they could customize a work train to simulate a couple cars of a open gangways train and test some of the tightest curves in the system.
That’s what they did when it came to the 75-footers! They cut up an old R9, added 15 feet to it, outfitted it with “whisker” sensors to detect clearance issues and let it roam the system.
Is not the extra weight on the shared trucks under the articulated joints a factor? These interior trucks would have twice the weight on them compared to similar cars on non articulated trains. Even if somehow one could really lighten up the cars themselves, I doubt that half the weight could be shaved off.
That’s not exactly true. The R142s (for example) are based on the Bombardier Movia Metro platform. The Movia Metro platform, in it’s standard configuration (if any particular configuration can properly be called standard), incorporates cars with the same two truck layout as the R142s AND open gangways. In other words, it’s not necessary (and possibly not even usual) to have shared trucks at the articulated joints, and there are other cities using rolling stock that is little more than articulated versions of subway rolling stock.
Have a look at the Berlin U-Bahn H Type for comparison. Same platform (Movia Metro) as the R142, same car length as the R142, about the same car height and width as the R142, same track gauge and similar two truck per car layout. The principle difference is that the R142 do not have open gangways.
That’s not to say that H Types could be substituted for R142 rolling stock. But weight issues arising from truck layout are not any type of problem.
If the trucks are not set up underneath the articulation point doesn’t the amount of required accordion action (to address the ends of the cars swinging out away from each other on the curves) become much greater – to the point where the necessary interior structural accommodations (i.e. movable wall panels and sliding plates on the floor kind of thing) may have to go beyond what can be stashed behind the interior surfaces? Wouldn’t things like exposed accordion folds and moving plates be kind of dangerous – at least with the kind of statistics dealing with millions of passengers?
I’m not an expert, so I won’t speculate on how various articulation designs work in practise.
However, I will summarise my earlier point. The R142 cars were designed by Bombardier at about the same time as the Berlin H-Type. The two cars are variants of the same design (the Movia Metro railcar), are the same size and are even superficially similar. Both are two truck per car designs. The H-Type is articulated, without requiring a truck under the articulation point. It works just fine. In fact, the H-Type is currently the main type in use on the Grosprofil division of the U-Bahn (which is analogous to B-Division on the subway).
“I’m not an expert, so I won’t speculate on how various articulation designs work in practice.”
I am not an expert or an engineer.
One of the problems in designing / creating / using 75-foot cars in the BMT/IND subway system was that the cars had to act like 67-foot cars when it came to the turning radius in the tunnels. Basically the BMT/IND subway tunnels were designed for the maximum length of a subway car to be 67-feet. This point is nothing new – it has been well established in plenty of NYC subway history books, etc.
It has been well described that on the 75-foot cars used on the NYC subway on very sharp curves that the end doors do not meet. That is why the end doors on 75-foot cars are often locked to prevent riders from falling to the tracks as the train rounds a curve.
The point about open gangways is really about to accommodate and negotiate such sharp curves on the BMT/IND subway lines.
The Berlin H-Type and same design Movia Metro railcars that you’ve noted may indeed have open gangways installed and used on certain transit systems. These same transit systems may not have a same types of sharp curves that exist on the BMT/IND NYC subway system.
A similar car design feature may not work well in another place, simply because of the features of the other place.
It could simply be the case that if “we” want open gangways and connected subway cars – then the 60-foot type car designs might be the better way to go. Just a thought.
Mike
Mike
600 feet of 60 foot cars has ten cars and 40 doors if there are 4 sets of doors per car. 600 feet of 75 foot cars has 8 cars and 32 doors.
Your point makes perfect sense. Obviously just because two subway systems use Movia Metro railcars doesn’t mean that the rolling stock on both systems is interchangeable. According to the manufacturer, Movia cars are customised to the requirements of each customer, which I assume includes modifications to suit the loading gauge, structure gauge, turn radius etc etc of each system. (Again, I stress, I am no expert, I don’t know any of this for certain, I’m just making guesses that seem reasonable to me).
What I was really commenting on though was the idea that open gangway trainsets have inherently higher axle loads because they inherently have fewer trucks due to the requirement that trucks be shared between adjacent cars and placed underneath the articulation points. (And that these higher axle loads exceed the design of the subway and are the real reason open gangway designs are not used).
Clearly, by looking at international variants of existing subway rollingstock (such as the R142), it can be seen that open gangways versions are done all the time on car designs that do not have shared trucks under the articulation point. Therefore open gangways do not have inherently higher axle loads. Therefore, axle loads are not the reason open gangway rollingstock is not used in the subway. There may of course be other very good dimensional reasons (as you outlined).
I think the best arguments against open gangways are hygienic (stink car becomes stink trains).
I’ve never been in a stink car. Only in a stink half-a-car. The smell doesn’t travel that far.
I most certainly was, last week one evening on the F Train. I could only bear it until the next station when I and the other remaining passengers who bathed during the past month fled into the adjacent cars. And over time, I’m convinced the A/C units, or filters absorb these odors. I often think many cars have an odor akin to yesterday’s diapers even when there’s no homeless person present.
I’ll say it once again, the Q wouldn’t be so crowded on weekends is the B ran
The Q was perfectly fine with 8-minute weekend headways before the 2010 cuts. If the B were restored, it (and the Q) would operate no more frequently than every 15 minutes (giving 7.5 min combined headways on Brighton) or else trains would be running half-empty all the time. One frequent service is better than several infrequent ones.
I would rather they at least run the B on the local track on weekends.
If I wanted to go to a Yankee game from Ave H / Newkirk Plaza on a weekday, 1-seat ride on the B. On a weekend, It’s the B to either the D or 4 & no cross-platform transfer.
It’s too bad that there’s no way for 4th Avenue Express trains leaving Atlantic to stop at DeKalb. That would allow the Brighton to 6th Ave weekend transfer without having to use Herald Square.
I suppose they could switch to the local after Atlantic and then switch again to the Brighton platform before DeKalb…there’s certainly not enough frequency on the R train to cause problems here. But it would only provide a same-platform transfer, not a cross-platform transfer.
Fantasy, perhaps, but hear me out:
If the R tracks at Dekalb were moved to a lower level, directly under where they are now, a platform could be built over what are now the R tracks, which would allow trains on the bypass to stop at Dekalb. Two switches would need to be added north of Pacific Street, but no routing flexibility would be lost.
This would also make de-interlining the B/D/N/Q more palatable, as people wouldn’t have to walk the entire passageway at Atlantic anymore.
I would argue that the Q every 10 minutes (let alone every 8) is better than the B (local or exp) and Q, each every 15. On the way back from Manhattan, passengers would have to choose a platform in advance and then wait up to 15 minutes for a train.
This would also help out the C on CPW, a needed improvement as well.
How about providing more accountability for unexplained gaps in service? I waited 11 minutes for the uptown 6 train at Astor Place at 8pm last night. The scheduled headway is half that. The train just didn’t leave Brooklyn Bridge on time, which is typical. It’s easy to blame delays on crowding, but sometimes the MTA just aren’t doing their jobs and no one seems to care.
Everyone at the MTA should be forced to go visit other cities’ transit systems around the world and see everything they do so much better. I’d be happy to pay for these junkets with my tax dollars. We need to destroy the “not invented here” and “we’re doing so much better than the 70s when your train caught fire and you’d get mugged” mentality that pervades the system.
Do crews get a break at Brooklyn Bridge? I’ve wondered about this since it’s a looong run from Pelham Bay. If there’s not a relief crew and someone needs a restroom, that will delay the train.
It seems hard to believe that restroom breaks are the reason for the delays, given that they’ve had approximately a century to figure out how to deal with the distance between Pelham Bay and Brooklyn Bridge, and the 6 isn’t even the longest line…
They could do what London calls “stepping back” where the arriving crew gets a break and then takes a train that another crew brought in. It’s not exactly rocket science.
Not quite – ‘stepping back’ is taking the next train out from the same platform, at places with intensive service and limited number of platforms at terminal stations. EG. Driver brings in train. New driver is already at the end of platform, gets into other cab and train leaves within 1-2mins. The driver from train 1 is then waiting at end of platform for the following train to come in. This allows the train to leave the platform quicker than it takes for a human/driver to walk the whole length of the platform and get in at the other end.
Stepping back, in its true sense does not allow the driver from train 1 to have a toilet/other break. Of course this can be included in rostering to give a e.g: 10 min comfort break between train/driving duties, but that is different to stepping back.
Not quite – ‘stepping back’ is taking the next train out from the same platform, at places with intensive service and limited number of platforms at terminal stations. EG. Driver brings in train. New driver is already at the other end of platform and gets straight into this cab, and train leaves within 1-2mins. The driver from train 1 is meanwhile walking back along the platform so he/she is then waiting at the correct end to drive the following train out. This allows the train to leave the platform quicker than it takes for a human/driver to walk the whole length of the platform and get in at the other end.
Stepping back, in its true sense does not allow the driver from train 1 to have a toilet/other break. Of course this can be included in rostering to give a e.g: 10 min comfort break between train/driving duties, but that is different to stepping back.
I’ve seen this done during weekend service changes where trains need to turn quickly. When 1 trains turn at 96th/Bway, for example, they’ll pick up an operator on the uptown platform to reverse the train quickly back into the station. I still wonder about the matter of breaks on such frequent services as the 6. Guess I’ll have to observe sometime.
You’re right, of course. However, the basic idea of the crew getting off their train and then (after a break, perhaps) boarding another one remains.
I live at a subway terminal. I dont ride every day, but atleast once a month 1 of the 2 train crews never shows up. Dispatcher rings the “time to leave” bell, and train doesnt move. Eventually the next train’s crew is standing on the platform, and the 1 crew member leaves the train and the next pair get on. So the train “never ran”. Other times the TO or TC slowly walk down the stairs on to the platform 5 or 10 minutes late not giving a fuck sipping coffee. The train then skips 3-5 stations to “catch up”. These would be termination events at any other metro in the world.
Leaving the terminal on time is bad in NYC.
Um, yes, this. I wish this were more of a story. The J train from Broad Street frequently simply never makes scheduled runs. The next scheduled run happens on time as if the previous one didn’t exist.
You failed to mention reopening closed subway entrances which the MTA is supposedly studying but is dragging their feet on. That would even out crowding along the platforms and on the trains where everyone rides the same car in order to be near the only exit.
The reasons for closing many of these entrances in the 1970s is no longer valid. Safety in the subways and on the streets is greatly improved and station agents would no longer be necessary at these entrances which were the reasons for closing them in the first place. The capital expenses would be minor for the benefits that would result with five minutes being cut from many trios do to decreased walks to access stations. The IND was all designed to have an entrance at each end rather than only one over station which is why the stations are further apart.
This is really a no-brainer. All that is needed is a study to determine which entrances are most in need of reopening and setting a time schedule given available funding and resulting benefits. Yet the MTA is looking for every excuse not to go ahead. This is why people wonder if the MTA has any regard for its passengers or is only interested in its budget and in appeasing real estate and banking interests.
I’ve written quite extensively about closed entrances (March, November, January 2015), but I think you’re conflating station passenger flow with train capacity. Reopening closed entrances is a rider-friendly initiative that the MTA should pursue (but is hesitant to follow due, in part, to the cost of ADA compliance). It won’t help with the fact that trains can’t fit more people, and if anything, would likely lead to more crowding in the system as more people find transit more accessible.
Try getting out of the R at Whitehall. The lone narrow N. side stairway gets so jammed up that the line backs into the cars I shit you not.
The MTA added a second stairway on the south end of Whitehall Street platform leading up to the SI Ferry terminal, but of course it’s at the narrowest end of the platform behind the original staircase, where the platform has stalagmites and it smells like a gallon of piss is dumped there every ten minutes, so of course few people use it. Visitors from other cities must shake their heads and the decrepit state of NYC subway stations, and Whitehall Street is far from the worst of the lot.
Oh but I guess you’re still right. That’s not a station exit issue. Though they should still add another stairway to the S/B R at Prospect Expwy.
Even if you separate the two issues–train overcrowding and insufficient entrances at some station–both are exacerbated by record ridership. Not all trains are crush loaded. But there are many stations, including ones on lines that aren’t at capacity, that face serious issues of insufficient entrance/exit capacity, which leads to both delays getting in and out of the stations and overcrowding on the portion of the platform near the entrance/exit. While it is true that adding stairs at stations along the L train won’t help with the fact that the trains are at capacity, it will help in getting in and out of the stations, as would the transfer redesigns along 14th Street that RPA recommended.
I can think of a dozen stations that need more entrances and exits, some of which have closed stairs that could be reopened. For the MTA to dismiss adding station entrances because of the potential ADA costs is both myopic and untrue. Smith-Ninth and Fourth Avenue were both rebuilt as part of the Gowanus viaduct reconstruction. Neither was made ADA compliant and Fourth Avenue even reopened the entrance on the east side of Fourth and 10th Street. It can and should be considered to decrease commute times by improving passenger flow.
If they’ve gotten away with top-to-bottom overhauls of dozens of stations without adding ADA access, why would opening an extra exit or two suddenly trigger these rules?
(Law in question: https://www.access-board.gov/the-board/laws/americans-with-disabilities-act-intro, section 227 Alterations of Existing Facilities)
Good question–I heard that the FTA changed its interpretation of the law, but I don’t know exactly how or why.
Ben,
Please explain to me how Avenues H, J, M, U and Neck Road were all thoroughly demolished and then reconstructed without being made ADA compliant. I fail to understand how opening a station entrance that was preciously closed requires the station to be made ADA compliant when a 100 percent destruction of an entire station and subsequent rebuilding of it does not fall under ADA regulations.
When they talk about people falling off the platform, they are echoing a prediction I made at the end of this post.
https://larrylittlefield.wordpress.com/2016/01/02/the-nyc-subway-squeezing-the-serfs/
“As New York’s Ghost of Christmas Future, I can see it with perfect clarity. Signal or train car troubles that delay service on a very crowded line. A platform filling with impatient subway riders. Trains finally coming that are too full to board. More crowding on the platform, with people now packed to the edge and unable to leave the station unless they are directly adjacent to an exit (yes I have seen something like this happen recently). Some tomfoolery by a group of teens, a little pushing, a panic – and a few dozen people tumble to the tracks and are killed and maimed by an onrushing train.”
“May these shadows yet be altered? Probably not.”
And then all the people who made this inevitable will posture.
“How about providing more accountability for unexplained gaps in service?”
The IRT had lots of hidey holes for trains. Check out the track maps.
http://www.nycsubway.org/wiki/.....Track_Maps
For the number 6 there is one between 68th and 59th. Another one south of Union Square facing north. Another one south of Brooklyn Bridge.
Back in the day, empty trains would sit there, and if a gap in service arrived, they would fill it. It costs money. We’re keeping a second person on the train — the conductor — on low traffic and CBTC lines, and cross-subsidizing featherbedding on the LIRR, instead.
I was thinking about the pocket tracks the other day when I was experiencing a big gap in service. Are these obsolete now, or were they extended to accommodate ten car trains?
Many (though not all) of them are still perfectly functional. While some must be kept clear to allow malfunctioning trains to be stashed away until the end of rush hour, there are still places for gap trains to be stored where they are not.
I believe gap trains are still used during big events (Yankees games, parades, etc.) but not rush hours. I wonder what it would cost to do so, and if it’s being considered. It would seem like a worthwhile way to protect service.
We’re keeping a second person on the train — the conductor — on low traffic and CBTC lines…
You’re looking in the wrong place regarding manpower required to operate an NYCT train. There are 4 additional maintenance workers keeping each train operational for each revenue train hour.
However, when I compared the level of staffing at Car Equipment, Maintenance of Way, Rapid Transit Operations and Stations over the years I found big decreases at CED and MOW — at the same time that car reliability and track maintenance were increasing. No such productivity gains at RTO and stations despite the investment in the master towers, the RCC, CBTC, Metrocard Vending machines.
Might have changed the past few years.
Focusing only on the impossibility of doing anything on the Lex is missing part of the point. Riders will take alternatives that they perceive as more convenient, more comfortable, faster, or at least 1 of those 3 without being significantly worse in the other 2. For instance, most will not take the R via Lower Manhattan to have a more comfortable ride vs. the N or Q…but many would consider taking the 2/3 instead of the 4/5 from Brooklyn, if their destination was below Chambers and the 2/3 was less crowded.
The Lex might not be fixable in the short term, but the A/C is also crowded during the peak. So is the G (which is gonna be a huge problem needing fixed before any L shutdown). So is the F.
We don’t have to do nothing just because the biggest problem can’t be solved. Add 1-2 trains to the C during the peak (I believe there’s room for 1 more train during the peak thru Cranberry Tunnel without reducing A service). Bring back the W sooner than SAS opening, which fixes the problems with the N/Q. Can one 6 train be stored on the South Ferry Inner Loop during daytime hours, ready to run light up to Brooklyn Bridge should a hole in service arise?
It’s amazing when you go to other cities on a lesser plane than New York, and they are able to provide significantly better train service. Montreal and Toronto can provide 25-30tph during the rush and 15-20tph off-peak. London and Paris are even better. New York insists they can do no better than 6tph off-peak and maybe 8-10tph peak on all but the busiest lines.
“It’s amazing when you go to other cities on a lesser plane than New York, and they are able to provide significantly better train service. Montreal and Toronto can provide 25-30tph during the rush and 15-20tph off-peak. London and Paris are even better. New York insists they can do no better than 6tph off-peak and maybe 8-10tph peak on all but the busiest lines.”
Is that because the cover less of their costs? Or is it because of national health are and retirement schemes that leave public employees, including transit employees, in the same boat as everyone else?
It’s because we’re on the lesser plane, and too proud to admit it.
Bingo. NYC loves to tout itself as the “Capital of the World” and in many ways it totally leads the pack, but public transit is not one of them. I’ve ridden subways in Philly, Chicago, Los Angeles, DC, Paris, heck even Cleveland and the trains almost always come as scheduled and magically don’t smell like piss. The state of cleanliness of a PATH train platform compared to the Sixth Ave IND right next door is startling.
Yep even Chicago has better off-peak and overnight headways on some lines.
Is there truly nothing to be done about the Lex?
Isn’t there spare capacity on the 6? I know it’s the local, but we all secretly know that the local is in fact not very much slower (even if it should be), and if you weren’t going beyond Brooklyn Bridge anyway, it might be a decent trade off even if you needed to transfer once. I’ll grant you that transferring a second time at Brooklyn Bridge to go beyond there could be asking too much. Some kind of skip stop service could also be implemented on the local tracks to cut off little more time at the expense of some confusion/irritation.
The reversing at Brooklyn Bridge is a loop, so unless there are some time-consuming work-rule procedures to perform there, I would think that there need not be much more delay than at a regular thru stop. If turning at Pelham Bay is already maxed out, perhaps some trains could be short turned at Parkchester and/or run into the Westchester Ave yard to be turned.
So, if it were possible to run more 6 trains, then some amount of 4/5 line traffic might be divertable. It’s more or less as likely as the expectation that people will get off the Lex at 125th, then walk a fair bit to transfer to the SAS. If there is room to beef up the 6, it is if nothing else, one hell of a lot cheaper than all the $Billion on the SAS.
Neither the 4/5 or 6 are run at international-best-practice levels. The issue is dwell time: in fact, the 28tph that enter Manhattan on the 6 during the AM rush thin out to 26tph by Grand Central, simply because of the dwell time and resulting congestion.
the 28tph that enter Manhattan on the 6 during the AM rush thin out to 26tph by Grand Central, simply because of the dwell time and resulting congestion.
You have it backwards. Increased dwell time is an indication of an insufficient number of trains.
Regarding the 6 in Manhattan: the current schedule has 21 #6 and #6X trains leaving 125th St between 7:30 and 8:30.
Another consideration is headway uniformity. Consider a 24 tph service level. The headway is 2:30 between trains. This could also be achieved by alternating headways at 3:00 and 2:00. The problem is that passengers are assumed to enter stations uniformly. There will be 50% more passengers on platforms at the 3 minute headway than at the 2 minute headway. These extra people translate into approximately 50% more dwell time.
Here’s a list of the scheduled arrival times for the Lex Local at 125th St
07:32:00,07:36:00,07:39:30,07:43:00,07:45:00,07:48:00,07:51:30,
07:54:30,07:57:30,08:00:00,08:02:30,08:05:00,08:07:30,08:10:30,
08:13:00,08:16:00,08:19:00,08:22:00,08:24:30,08:26:30,08:29:00
This is the headway from the train to the previous one.
00:03:30,00:04:00,00:03:30,00:03:30,00:02:00,00:03:00,00:03:30,
00:03:00,00:03:00,00:02:30,00:02:30,00:02:30,00:02:30,00:03:00,
00:02:30,00:03:00,00:03:00,00:03:00,00:02:30,00:02:00,00:02:30
Following trains can proceed only as fast as their slowest leader will permit. These figures show that despite operating 21 tph, they are more likely to operate closer to 15 tph because of the non-uniform scheduled headways.
The problem with “Increased dwell time is an indication of an insufficient number of trains” is that there is a minimum dwell time below which it is nearly impossible to go. If the 6 was shifted to 30tph today, perhaps it would be able to sustain such a service level for some period of time. However, as crowding increased, the same issues with long dwells that exist today would resurface. As currently operated, the subway system is unable to keep dwell times consistently low enough to run world-class service.
Furthermore, uneven headways obviously need to be remedied, since there’s no operational reason for them that can’t easily be fixed. However, saying that 50% more people on the platform means 50% more dwell time isn’t totally accurate – the relationship is not linear, and not everyone fits onto the first train anyway.
“saying that 50% more people on the platform means 50% more dwell time isn’t totally accurate”
The number of people waiting on a queue is described by a Poisson distribution. The average number of people waiting is proportional to the time interval that people are entering the queue. The variance is equal to the average.
This means that non-uniform headways make for longer and less predictable loads at stations and in trains. This results in longer and less predictable dwell times.
“There will be 50% more passengers on platforms at the 3 minute headway than at the 2 minute headway. These extra people translate into approximately 50% more dwell time. ”
Worse, the 3-minute train will travel more slowly, so the trains will bunch.
It’s absolutely ridiculous that a supposedly over-capacity line (with no interlining) can only run 21 trains per hour, when comparable lines in other cities worldwide can run 30 to 40 trains per hour.
The 4/5/6 are really where the MTA should put the open gangway cars. Too bad they’re only starting the open gangways on the R211, which is going to the B division (Lettered lines).
“If turning at Pelham Bay is already maxed out, perhaps some trains could be short turned at Parkchester…”
That is in fact exactly what happens. Most 6 local trains turn at Parkchester, while the expresses go to Pelham.
There are numerous other short-turns and terminal swaps to balance turning capacity–2/5 trains to New Lots, E trains to 179 St, F trains to Kings Hwy, L trains to Myrtle-Wyckoff, etc.
Given that we consider 30tph to be capacity, while other places run 40 with block signals, we shouldn’t be complaining about how overcrowded the subways are. There are many things that can be done to relieve capacity crunches without building new subways at prices that would seem like the stations are made of gold…
WHat grates me about these articles in the mainstream press is the lack of any discussion as to the reasons for these issues and what can be done about them, and the simply repeat the few words thrown at them by the MTA. Even the one example given about London didn’t discuss what London is doing about it in building their subway, and discussing how an option like this is possible with the regional commuter railroads like or the earlier attempt by Governor Rockefeller.
Historically when the Subways were built, the press was extremely active in the discussions and in the push to create them. So I wonder where are they when we need them?
WHat grates me about these articles in the mainstream press is the lack of any discussion as to the reasons for these issues and what can be done about them, and the simply repeat the few words thrown at them by the MTA. Even the one example given about London didn’t discuss what London is doing about it in building their subway, and discussing how an option like this is possible with the regional commuter railroads like this recent one by Alon Levy or the earlier attempt by Governor Rockefeller.
Historically when the Subways were built, the press was extremely active in the discussions and in the push to create them. So I wonder where are they when we need them?
http://www.thetransportpolitic.....ty-part-i/
My personal solution to subway overcrowding is to avoid the subway as much as possible (and I also do my best to avoid taxi/uber). I know it’s not realistic for everyone, but I walk and bike to work and to almost everywhere else I go. I happen to live reasonably close to work and also Citibike (despite their own problems with reliability) has been a gamechanger for many destinations that are too far to walk, especially those that aren’t well connected by the subway.
Somehow I must just have bad luck with the subway though — I end up taking it a few times per week and over half the rides I end up standing there on the platform waiting well over 10 minutes for the next train to come. Which of course discourages me even more from using the subway.
Good for you. I swear, there’s always that one a-hole on every article who thinks they’re so “ahead of the curve.”
I’m just spitballin’ here, but wouldn’t open gangways allow for longer trains? For example, you could have a 12-car train where only the middle 10 cars platform, and the doors on the first and last car would remain closed. The open gangways would allow people to enter via the platform and spread out through the whole train. You can even still have conductor’s cabs in the middle, making the train from a pair of six-car OG trainsets.
That is an excellent idea, and the IRT had a variation of that in the 1910’s where 6-car locals would only open the first door of the last car and last door of the first car. However, with today’s inattentive passengers and holdable doors, can you imagine the dwell time at every stop? There’s always going to be that genius in the back of the twelfth car that will realize it’s his turn to exit just as the train is pulling in, and he’ll have to fight his way through a carload of people while another “good samaritan” holds the doors to let him off.
Furthermore, the signal system at stations is not configured for longer trains. There are signals directly before and after most station platforms. The MTA did a pilot with an 11-car F train in 2009 (?) and that was one of the issues identified.
While I realize that keeping older cars around longer that planned costs money, why can’t they delay the retirement of whatever is to be retired next and that way have more rolling stock? If they keep 100 old cars for extra 2-3 years(or more) and run about 50 of them they could use the rest for spare parts resulting in mostly labor costs with little capital expense(it ain’t perfect as I am sure some parts are going to run short and they will need new spares after taking all from the old cars).
Or have they already gotten rid of all old cars and now won’t be able to use even this scheme? They might have gotten rid of the old retired cars so due to lack of space to store them.
Furthermore, it is obvious that they need more yard space. I have never heard of anyone talking about new yards. Are we basically bound to the current number of cars/trains because of the lack of yards? Are there actually any reasonable locations left within the city where they could put yards in the next 10-20 years?
They’ve already done that with the R32 trains, which should have been retired years ago. They’ve performed admirably but now break down constantly and really can’t hold on much longer than they’ve already been tasked with.
The funny thing about yard space is that trains on the road don’t need yard space. If more trains were run, at all times, there’d be less need for yard space. So you could add more trains by simply running more trains overnight. Obviously there are labor costs involved with that, which is a whole separate (solvable) issue.
There are also maintenance issues. If more trains are running all night when do you maintain them?
Other solutions:
More doors
Larger doors
No door recycling (closing means closing, period)
Less seats
The problem with the door cycling is that people stay holding the doors, selfishly holding up the whole train until they are allowed to enter. In other cities like London, when the doors begin to close, everyone dutifully steps back and waits for the next train.
However, this is intertwined with a frequency discussion because passengers in London can be confident the next train is coming in 2 minutes, because #1 it always does, and #2 the sign says so. Not so in most of New York.
What about setting up a peak/off-peak fare structure? If it works to help reduce rush-hour crowding on the commuter rails, why wouldn’t it work on the subway? Yes, it would make unlimited ride MetroCards more complicated (you’d have to have two kinds, one good 24/7, and a cheaper one that’s off-peak only). People wouldn’t like that. But people don’t like crowding either! I think it’s the lesser of two evils.
While the split fare structure works on the railroads, doing that for the bus & subways isn’t going to work due to most folks’ having jobs that require them to travel at what are called the peak times. Hence, why they are the peak times. I don’t see the MTA going that route. Not everyone can tweak their start times so that they travel on an off-peak schedule.
Not everyone can tweak their start times, true. But you only need *some* people to tweak their start times and you’d reduce crowding.
The issues you describe for why the MTA can’t implement peak fares are just as true for the commuter railroads where you say a split fare structure “works well”
i’m not sure its an idea solution, but combine that with Move NY, and lots of investment in expansion and it might eventually be worth it as ridership continues to grow.
What would you think peak is?
7:30-9:30?
Many cities implement peak fares in the AM only, as the PM peak is more spread out.
Because you’d penalize people who need to get to work at 7 or 8 in the morning because their bosses say they have to be there at that time. What’s the incentive for business to stagger their hours, especially in this service-based economy that Bush/Obama have left us with? Some poor deli worker from the Bronx has to pay higher subway fares so a banker on the Upper East Side can cruise into his office at 11 AM or whenever he wants to because he can make his own hours? I don’t think so.
Fares keep going up anyway, so, it’s a question of whether it makes more sense to raise fares in a way that can reduce rush hour crowding, or whether to just raise fares on everyone.
Singapore has/had a very successful program where you pay no fare if you exit the system before 7:45AM. NYC could use that model…
http://www.citylab.com/cityfix.....ss/374909/
Wait, one train every 8 minutes is the most they can run on the Q?
Why so few?
How about, get a bike? If 10% of subway riders switched to bicycle, we’d be back to 2010 levels. And NYC has a lot of great but still under-used bike lane capacity.
I addressed that in today’s post. The short of it is that you’re not going to get 10% of subway riders to switch to biking.
I’d offer two suggestions. First, reduce the severity of disruptions. Unions are adamant about 2 person train operation. Other metros prove one person or automated control is possible & often faster. Rather than fight them we ought to update their abilities. The middle individual should become First Aid rated. Thus, when a person becomes sick, this person should be empowered to jump off and treat the individual while the train moves on. Surely, this would be an acceptable way to keep these workers while actually making them an objective value add.
Second, we need to consider demand pricing. It’s critical to realize that the system isn’t unable to move 5mil people, just in the distribution when people currently travel. Obviously no one seeks the crush periods & some don’t have infinite flexibility. But, we need to have some method of encouraging some smoothing of demand. There needs to be a discounted period both before and after rush hours to allow people to schedule lives appropriately. If we can get 10% of people to either come in early & partake in local gyms, breakfast meetings or other activities or do so back home and come in slightly later, it will vastly improve overall performance. Moreover, because pricing is flexible we can decide whether we want it to be revenue neutral or not. So Off-peak trips could be a $1 and peak trips $4, thereby peak people pay to encourage off peak trips and no revenue is lost. Or just keep peak the current fare and reduce for off-peak. That would reduce revenue but if we are asking ourselves so what we would have to spend to get a 10% increase in peak capacity, I think it’s the cheapest method available. Presuming some come into the city early or stay later, the city will make up for it in other economic activity.
What makes you think people who already have the option to come in early or late don’t already do? They can avoid the crush.
Are the MTA’s ridership numbers total trips or total daily users? I cannot find anything online to suggest one way or the other. I know that RITA quotes transit numbers in trips but my hunch is that this is a passenger number — not that this matters for the point either way.
I believe they count the number of Metrocard swipes. Therefore, a person going to work and then coming back home would count twice.