Platform Edge Doors

In New York, a well-publicized homicide by pushing the victim onto the subway tracks created a conversation about platform edge doors, or PEDs; A Train of Thought even mentions this New York context, with photos from Singapore.

In Paris, the ongoing automation of the system involves installing PEDs. This is for a combination of safety and precision. For safety, unattended trains do not have drivers who would notice if a passenger fell onto the track. For precision, the same technology that lets trains run with a high level of automation, which includes driverless operations but not just, can also let the train arrive with meter-scale precision so that PEDs are viable. This means that we have a ready comparison for how much PEDs should cost.

The cost of M4 PEDs is 106 M€ for 29 stations, or 3.7M€ per station. The platforms are 90 meters long; New York’s are mostly twice as long, but some (on the 1-6) are only 70% longer. So, pro-rated to Parisian length, this should be around $10 million per station with two platform faces. Based on Vanshnook’s track map, there are 204 pairs of platform faces on the IRT, 187 on the IND (including the entire Culver Line), and 165 on the BMT (including Second Avenue Subway). So this should be about $5.5 billion, systemwide.

Here is what the MTA thinks it should cost. It projects $55 million per station – but the study is notable in looking for excuses not to do it. Instead of talking about PEDs, it talks about how they are infeasible, categorizing stations by what the excuse is. At the largest group, it is accessibility; PEDs improve accessibility, but such a big station project voids the grandfather clause in the Americans with Disabilities Act that permits New York to keep its system inaccessible (Berlin, of similar age, is approaching 100% accessible), and therefore the MTA does not do major station upgrades until it can extort ADA funding for them.

Then there is the excuse of pre-cast platforms. These are supposed to be structurally incapable of hosting PEDs; in reality, PEDs are present on a variety of platforms, including legacy ones that are similar to those of New York, for example in Paris. (Singapore was the first full-size heavy rail system to have PEDs – in fact it has full-height platform screen doors, or PSDs, at the underground stations – but there are later retrofits in Singapore, Paris, Shanghai, and other cities.)

The trains in New York do not have consistent door placement. The study surprisingly does not mention that as a major impediment, only a minor one – but at any rate, there are vertical doors for such situations.

So there is a solution to subway falls and suicides; it improves accessibility because of accidental falls, and full-height PSDs also reduce air cooling costs at stations. Unfortunately, for a combination of extreme construction costs and an agency that doesn’t really want to build things with its $50 billion capital plans, it will not happen while the agency and its political leaders remain as they are.


  1. Gregory Homatas

    In a new subway system, you can air condition/air temper the passenger area from the design stage whereas with an older system like NYC, which has many older stations and some new stations such as the Second Ave Subway where air tempering or air conditioning was built into the station from the beginning (the second ave line so far doesn’t have platform doors either) how would you build platform doors with adequate HVAC for the passenger crowds that NYC experiences? NYC has a wide variety of station configurations such as subway, open cut, elevated, etc to be accommodated. I am not saying it’s a bad idea, just wondering about the practicality.

    • Samuelito

      This is not a complete answer, but for what it’s worth, some of the smaller BMT stations – for example, many along the J line – have small waiting rooms. You wait inside, fully enclosed with push doors (and heated during the winter), until the train comes, then either the countdown clock or an old-fashioned beeper that triggers an alarm when a train is approaching will let passengers know that it’s time to get onto the platform. 111 St on the 7 line has this also.

      Also, I believe the 4/5/6 platforms at Grand Central are legit air conditioned, while platforms at Union Square use ceiling fans.

    • Daniel Bowen

      As the Singapore photo above shows, doors don’t necessarily have to be floor to ceiling.

      Singapore has retrofitted them into elevated stations. Full platform roof cover probably helps.

    • Sassy

      You can build half height platform doors that allow more airflow, such as is common in Japan. In addition, the vertical rope style of gates used to accommodate a wider variety of door placements (which NYC might need anyways), provide a ton of ventilation. In addition, half height and vertical rope style doors tend to be lighter, and can be placed on platforms that may not be able to support the weight of a full height door.

  2. Phake Nick

    Vertical platfirm door is probably not a good option, as it require much extra space under the platform to fit such system in, and also have chance of tripping those who try to rush into the train last-seconds
    Japan have done a lot of development for platforms with different train door alignments, like extra wide platform doors to accommodate different types of trains.
    In Hong Kong, there have long been discussion of installing platform doors onto the East Rail Line, which is domestic section of century-old KCR Railway (although most platforms nowadays are from when the line double-tracked and electrified hence a bit younger), but the project have been delayed quite a bit of time as they’re introducing a shorter train to accommodate with station and gradient restriction of planned line extension, which also tied in with signal system upgrade to maintain capacity amid shortened trains and supposed to allow for more precise control of train, and these three interlocked works mean platform doors can only be installed after all the works on them are completed to avoid the need of dealing with trains with different lengths and different door alignments, and the result is roughly two decades of waiting, and still counting.

    • Eric2

      I think vertical screen doors are top down so there is no tripping hazard (I suppose there is a minimal crushing hazard, but they can be designed to not do a passenger serious harm). And since they would not deploy until the train door is already closed, there is no reason for people to be running onto the train at this point.

    • Andrew in Ezo

      At JR West Takatsuki Station you can see both the extra wide (start of video) and rope type platform doors(from 2:39), as the trains come in both 3 door (special rapid service) and 4 door (stopping service) variations, as well as the occasional vestibuled ltd. express types (doors at both ends):

        • Mark N.

          That really is a lovely, lightweight solution. It would be interesting to see how its costs compare to the other solid-screen systems. I would certainly expect a fair amount of savings alone from the far lower amounts of glass/metal and the less powerful motors required. It also appears to be better adaptable to varying door spacings. I guess it doesn’t provide quite the degree of protection as the solid screens, and it certainly wouldn’t provide any air-quality advantages, but the cost/benefit aspect appears very high.

          • Borners

            Well as far as I know nobody else other than JR West uses it. And notice the above video is for Takatsuki station which is on the same line. I suspect I didn’t work as hoped. Most doors in Japan are full shebang. I’d love to see a study comparing them.

          • Phake Nick

            A Japanese site mentioned some demerits of such system (together with some possible mitigation measures)
            They includes:The
            1. Are such rope really safe enough from preventing people from falling? They do not provide the same level of visual security, and the gap beneath them are also wide enough that babies can easily pass through. Although the article also mentioned it can function against the main objective of platform floor being preventing accidental falling. But there are also cases where people being pushed over the edge of platform that I am not sure whether this is being tested for?
            2. The special design mean there are tons of sensors in the equipment to detect whether there’re anyone beneath the ropes or between the ropes and the trains. Which translate to higher installation and maintenance fee although the system’s lightweightness mean cost on platform reconstruction work can be reduced. The article mentioned that improvement have been made such that fewer sensors are needed and cost can be saved, but it also resulted in higher error detection rate when passengers or staffs staying or walking next to the ropes especially on narrower platforms, preventing the ropes from closing normally and disrupting train operation
            3. Despite the ropes can prevent passengers from falling down the platform, it cannot protect passengers from fall on or lean on the ropes. And to prevent passengers in such situation from contacting the train and causing danger, additional space between the rope and the train is necessary which would be more space than regular platform door, and this could make some part of platforms a bit too narrow depending on platform design
            4. The rising rope and supporting structure could block trains’ side display
            5. It cannot prevent passengers from walking into the space between train cars, then fall and die from it.
            6. Unlike platform door which clearly indicate the door position, such design does not provide such indication and require additional judgement by users, especially those visually impaired, to determine the door position, and it is actually become more difficult for them because with the system there would be alert sound when they try to pop their stick to seek out door’s position which would scare them.
            7. The ropes would also open when a shorter train stop at the platform and the part of the platform have no train space, thus enabling users to simply fall from the platform through the lifted ropes.

            (Relevant images and videos embedded in the linked webpage)

          • Tom the first and best

            The platform ropes also have effectively as little capacity to block particulate pollution from rail tunnels, meaning they are an open air station only solution.

  3. Joe Wong

    Its a good idea, if you only have one standard subway car size, and doors spaced accordingly. However, the IRT cars like the R62/62A’s have three 50″ doors on each side, and the newer R142/R142A/R188’s have three 54″ wider doors which on the A or cabs cars are opposite each other, while the B or non-cab cars are staggered. On the BMT/IND cars you have the R46’s which have 50″ doors opposite each other and have more room on the number 1 ends of the cars where the cabs are, while the number 2 ends have less room. On the R68/R68A’s the doors and both ends are evenly spaced, while the number 1 end has the full width cabs while the number 2 ends have the half cabs. Now for the R143’s, R160’s these have four 50″ wide doors and are opposite each other, while the non-cab cars the spacing of the doors are opposite each other, unlike the IRT’s R142/R142A’/R188’s which their B cars have staggered doors. Its a tough choice with so many car sizes.

    • Alex Cat3

      They could at least plan for them in the future, picking one final door design to use, using it for all future rolling stock orders, and installing the doors on lines once the rolling stock has been completely replaced. With this method it would take 50 years to get it on all lines but 50 years is better than never.

      • df1982

        It’s also an argument for de-itnerlining the system and having dedicated rolling stock for each line, which Alon has argued for in the past.

        What are the experiences of platform screen doors with curved platforms? Are they able to provide ADA access? Do they enable automated running? Do they require gap fillers which then introduce another level of mechanical complexity (and potential breakdown, Union Square is notorious in this regard). In other contexts curved platforms have been argued as the sticking point for conversion to automation. Or can you go down the Vancouver route of driverless trains and no PSDs, but with motion sensors on platforms?

        • Eric2

          I don’t think motion sensors do the job – trains physically can’t stop fast enough to avoid people who are shoved into their path.

        • Alon Levy

          Gap fillers on the trains (not platforms) are a solved problem and the FRA was looking at mandating them on mainline trains (I don’t know what’s come of the 2020 plans).

          Accessibility levels in Singapore are very high; I don’t think there are curved platforms, but the PSDs are perfectly compatible with wheelchair access.

          • Jan

            > Gap fillers on the trains (not platforms) are a solved problem

            Are they, really? The ones I’ve encountered so far still take a few seconds to deploy and retract. On the most intensively used urban railways, every second of dwell time matters in order to achieve the desired throughput, so even a few seconds may already be too much during the peak.

            On other, not quite as intensively used regional trains you might think that the additional time for the gap fillers can somehow be fudged into the timetable, but in actuality the effects are noticeable even there. Since a few years e.g. the national infrastructure operator in Germany has mandated higher minimum dwell times for trains utilising gap fillers. This means that whenever there’s a timetable recast in a place where the current timetable predates the widespread introduction of trains with gap fillers, there’s a good chance that journey time will have to be lengthened, with all the associated knock-on effects that may cause (connections no longer work, on single-track lines the crossing stations might now be in the wrong place, etc.)

          • Alon Levy

            In Zurich the S-Bahn has gap fillers and the dwell times stick to the 30 second standard. I don’t know why the dwell times are longer here; it’s not that they’re busy, because the S-Bahn in Zurich is plenty busy as well.

          • Max Wyss

            The forced switch to low-floor entry has caused more issues for the S-Bahn Zürich than the retractable steps, because, compared to the first generation trains, the passenger flows are less distributed (upper and middle level vs. lower level), and the doors are only double-wide. But the schedules do allow decent operation. I also thing to believe that retracting the steps is faster than expanding, which shortens the delay when the doors are closed a little bit. I think they also worked on the locking door to departure process, so that some steps can be done while the retractable step retracts.

          • Jan

            > I think they also worked on the locking door to departure process, so that some steps can be done while the retractable step retracts.

            Meanwhile in Germany, things *used* to work like that on e.g. the previous generations of the Karlsruhe trams (i.e. the folding step on the high floor trams and the gap filler on the first generation of low floor trams would to some limited extent move in parallel with the opening/closing of the actual door leaves), but for the latest generations of vehicles the railway/tram inspectorate (respectively possibly their subcontractor who had the actual responsibility for certifying that the doors were safe for operation) insisted on strictly sequential operation for reasons of safety paranoia… (at least at that time a few years ago, when those trams were being introduced into operation, the applicable European Standard for rail vehicle doors definitively still would have permitted the old model of parallel operation as long as the doors aren’t opened more than 40 cm while the step/gap filler isn’t fully in place. It might in fact even do so today, although I don’t have access to the current version of the standard.)

          • Max Wyss

            Note that when the retractable steps start retracting, the doors are already closed and locked. When opening, the retractable step moves out and is in position before the doors start opening. Because they are only double-wide, they can open relatively quickly.

            The triple-wide doors of the DPZ cars are mechanically linked with the fold down step, and they will close in synch. (the DB variant of those cars has the fold down steps removed, in order to deal with 76 cm platform heights).

  4. Tom the first and best

    Platform Screen Doors also dramatically reduce the level of particulate pollution in underground stations (the enclosed nature of the tunnels prevents break, wheel and rail wear particles from dissipating). With partial coverage platform edge doors providing some reduction.

  5. thenerdysaxophonist

    Off-topic question: Would installing Platform Edge Doors be more expensive for S-bahn- or RER-type systems, since the platforms are also shared with intercity trains?

    • Alon Levy

      I don’t know. I only have Paris Métro costs – I couldn’t even find Singaporean retrofit costs. Tokyo retrofit the Yamanote Line with PEDs, but this is the most metro-like of all commuter lines I know of.

  6. Borners

    Sounds like cost disease trauma brain. I suspect London’s similar reluctance to embrace automation/retrofitted PED’s is the fear this will take away from expanding coverage. Though I can’t think doing for the Jubilee Line would be that expensive, its Eastern half is already PED and it only has 4 non PED underground stations. And its stuck on 30tph too.

    • Matthew Hutton

      London has the very small tube trains though which potentially make things more difficult.

      • Borners

        That’s the usual excuse, which is why I focused on the Jubilee. But if they could do it on Paris Metro Line 1 they can do it for those 4 remaining underground stations on the Jubilee. And it doesn’t have the ventilation problems of the Bakerloo/Central lines.

        • Matthew Hutton

          Bond Street and Green Park on the jubilee line are certainly the most plausible, Baker Street, St. John’s Wood and Swiss Cottage are older of course.

          • Borners

            I’d still wager the extra capacity of getting to 40tph would out cancel losing the Yellow line to a PED in terms of congestion.

    • Rational Plan.

      The problem has at the moment is an extreme lack of money, Until a different government comes in, it will be fighting to maintain service. There was a plan to install them on the Piccadilly line. But with constrained funds, we are only seeing new trains and upgrading frequency from 23 trains an hour to 27 an hour. Stage 2 would see that go up to 34/36 an hour. And did call for platform screen doors. But money is short and other lines needs new trains. And if commuting does not rebound. I doubt they need the extra capacity anytime soon.

  7. James S

    “PEDs improve accessibility, but such a big station project voids the grandfather clause in the Americans with Disabilities Act that permits New York to keep its system inaccessible ”

    Cuomos $250 million enhanced station initiative project spat in the face of this

  8. Charlie

    Why is it that PEDs are so rare in North America? Almost non-existent except for people movers and Toronto’s Union-Pearson Express

    • James S

      Theyre actually very common. By my count, 20+ systems have it. Aka, JFk Airtrain… Theyre just not run by old-fashioned transit agencies

  9. Herbert

    I’ll never understand why a decade after automaton, Nuremberg U-Bahn still doesn’t have platform screen doors – even tho there’s literally “here’s where the doors will be” painted on the platform at the U2/U3 stop at Nuremberg Hauptbahnhof…

    Like… How much could it cost?

    • Alon Levy

      Yeah, I don’t know. Vancouver has no PEDs, and it’s fine, but people do complain about accidents and say they should add PEDs.

  10. Paul Wennberg

    Full height also reduce noise. They also allow trains to enter stations safely at higher speeds.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.