Platform Screen Doors and the Interborough Express
By John Pegram[1]
Person-on-the-tracks incidents and driverless train proposals have led to increased consideration of Platform Screen Doors (PSD) in rail systems of all types. This article considers PSD basics, challenges at existing station—especially the potential entrapment problem, and rough cost estimates for installing PSDs in existing stations and in newly built stations on the Interborough Express (IBX) line,
Half Height PSDs on the Disneyland Resort line in Hong Kong.[2]
The purchase and installation of PSDs is quite costly and complex, especially in existing stations, and most especially in the NYC Transit System under its present standards. Although providing PSDs when building new stations could cost less, the costs would still be great as compared with other station costs. My rough estimate of the cost for providing PSDs at stations for Light Rail Vehicles (LRVs) on the IBX line is over 11 Million Dollars per station in 2018 dollars. However, those costs could be justified by a major reduction in labor costs if the IBX line were to use automated, driverless railcars instead of LRVs with operators.
This article is based primarily on a major study by STV Inc. of the potential for use of PSDs at all existing NYC Transit subway stations. The MTA has posted the resulting, 3,920-page report on its website.[3] STV also prepared materials for the MTA’s PSD pilot program request for proposals (RFP), parts of which the MTA has produced to me in response to a Freedom of Information Request.[4]
1. PSD Basics
The most common and generic term for moveable, platform edge barriers is “Platform Screen Door,” abbreviated as “PSD.” I will use that term here.
PSD systems are intended to prevent customers from accidentally falling, jumping, being pushed, or otherwise accessing the tracks illegally. In addition, PSDs can prevent debris and trash from accumulating on the tracks, reducing the risks of track fires.[5]
PSD systems are most often used in one of two forms—a full-height barrier that extends from floor to above the doorways, sometimes to the ceiling, or a “half-height” barrier that extends some distance above the platform slab (typically at least four feet). Both forms primarily consist of a series of motorized sliding doors (MSD) and fixed panels that are affixed on the platform edge, with the platform doors aligning with those on the train cars.[6]
A computerized “berthing system” makes sure the car doors and platform doors are aligned before opening. Some systems, including those specified for the NYC Transit PSD pilot project, also include manually operated, pivot-hinged Emergency Exit Doors (EED), placed between each pair of MSDs to provide emergency egress from a train that is not properly berthed at the platform, and hinged Platform End Doors (PED). The doors and panels are typically steel-framed with laminated, tempered glass windows.[7]
PSD systems are more complicated than other transit station systems. Not only do they have to provide a rugged platform edge barrier with many doors, they also require one or two electric motors for each door, power and communications connections to each motor, and an air-conditioned control room with computer control and communication equipment.[8]
2. PSD Challenges at Existing Stations
STV conducted a System-wide PSD Feasibility Study, addressing the challenges unique to the New York Subway system. Their February 2020 Summary of Conclusions is posted on the MTA website.[9] Based upon an earlier, background PSD study, continued discussions with NYCT and conditions discovered during field surveys, STV identified many challenges to installing PSDs in New York’s century-old subway system. The study found that overall systemwide feasibility of installing PSDs in existing stations would be 27%, or 128 of the 472 stations studied. The principal reasons for infeasibility are summarized in the table below:[10]
Most of these reasons are due to preexisting conditions and could be avoided at new stations, for example, stations on the proposed IBX line. However, one major challenge would exist at both new and existing stations under existing NYC Transit clearance standards, which is avoiding the possibility of entrapment of a rider in the gaps between platform edges and trains. That subject is discussed in the next section.
3. The Entrapment Gap
Entrapment distance refers to the space or gap between the track side of the platform door and the car door. The recommended gap of about 5 inches is based on the size of the smallest human body which could possibly be entering the train – a toddler. Under current NYC Transit clearance standards, the gap would be on the order of 10” to 15.5”, at least double the recommended gap. The reason for the gap under those standards is what the PSD Study called the “conservative criteria” used to establish NYC Transit’s “Limiting Line of Line Equipment” (LLLE), which specifies a railcar clearance slightly outside of the “Dynamic Envelope” (DE) of railcars used on the line.[11] The DE apparently accounts for the swaying from side to side of the railcar at its highest expected speed plus the possibility of one suspension failure.[12]
The PSD Study drawing below of a PSD with a B Division railcar indicates the possibility of an 11-inch gap between car door and platform door along a tangent (straight) track.[13] Gaps could be greater where the track and platform edge are curved.
The PSD Study did not include a drawing showing the gap that might exist with NYC Transit A Division railcars (which could fit through tunnels on the IBX line). However, as indicated in the preliminary drawing below for pilot program PSDs at the Times Square station on the 7 Line, a similar gap problem would exist with A Division railcars under the present NYC Transit clearance standards.[14]
The PSD Study concluded that there are three, alternative solutions to the gap problem in stations on existing NYC Transit subway lines.[15] The first is gap detection, using laser sensors installed above the gap and CCTV cameras to detect obstructions. The second option is to modify the NYC Transit railcar clearance specifications to effectively reduce the gap by moving the PSDs closer to the platform edge. This can be done, for example, by revising the apparent, current assumptions of one suspension failing while the railcar is at high speed, and/or reducing the station entering speed of the cars so that there is less rolling of the car. The third possibility would be to install rubber “bumpers” on the edge of each platform door, so that most of the gap is filled by the flexible rubber edge.
The PSD Study concluded that the rubber bumper “solution” was not viable in under the NYC Transit clearance standards.[16] Constrained by instructions to use those standards, STV recommended gap detection for the PSD pilot program at existing stations, installing entrapment detection and camera assisted visual verification at every door.[17]
There are two major problems with the gap detection approach: high cost and a high likelihood of false detection. Thrown objects, swirling newspapers, birds and lack of cleaning may create false positives and lead to train delays. The STV study estimated there would be a 70% false detection rate for 1 platform over one day, and one to three departure delays per 32-door platform per day.[18]
For new lines, the STV study recommended modifying the vehicle and wayside clearance to geometrically prevent entrapment,[19] as is done in many other systems. For example, they reported that RATP, the Metro operator in Paris, had removed multiple tolerances that were unlikely to occur concurrently.[20] See, for example, the picture below, showing an acceptable gap between a JFK AirTrain door and a PSD.[21] To eliminate gaps caused by curved platform edges, tracks should be tangent (straight) at stations.
4. Costs for PSDs
I have used the estimates of direct costs for installing half-height PSDs at the Times Square shuttle station[22] for analysis and estimating the average cost for providing such PSDs at the 19 proposed stations on the IBX line. The total platform edge length of the Times Square shuttle station is 621 feet,[23] which is very close to that proposed by the MTA for LRT mode station platforms on the IBX line, which is 600 feet.[24] (I expect that, if automated railcars were used on the IBX line, the platform lengths would be similar to the lengths for the LRT mode).
The direct costs estimated by the MTA’s consultants for providing PSDs at the Times Square shuttle station was $9,165,238 in 2018 dollars. The estimated total project cost, after markups, for PSDs at that one station, was $17,291,403.
New stations on the IBX line would not require reconstruction of existing platform edges and finishing the edges, which would be required at the Times Square station, estimated to have a direct cost of $637,752. Also, the Times Square station estimate included a laser scanner system having three scanners per opening, at a total direct cost of $666,611, and a CCTV system with cameras at each opening, having a direct cost of $820,000. All of that would not be necessary if the gap between railcars and platform doors is reduced as discussed in Part 3 above. The sum of those three amounts is $2,124,363, or about 23% of the Times Square station PSD direct cost estimate. The result would be direct costs of $7,040,875 for an IBX station PSD system. In comparison, the projected average direct cost of an entire IBX LRT mode station, with two side platforms, but no PSDs, was only $654,830. Even the projected average direct cost of an IBX CR mode station would be only $3,507,779, although such stations would be 80% longer than for the LRT mode, and have an overhead canopy, additional shelter and a platform snowmelt system, which is not included in the LRT mode station estimate.[25]
After applying the markups, the total projected PSD costs would be about $11,881,476 per IBX station, in 2018 dollars. For the 19 proposed IBX stations, the total PSD costs under this estimate would be about $225,748,055.
It appears likely that the costs for PSDs could be reduced a bit more by elimination of duplication between station construction and PSD installation work estimates, and because multiple stations could have very similar PSDs and installation designs.
PSD costs are huge; however, they could be justified by a major reduction in labor costs if the IBX line were to use automated, driverless railcars. Labor costs are approximately 75% of the operating expenses of NYC Transit.[26] (I will discuss IBX staffing in a future article).
This article expresses the personal views of the author and does not express the views of his employer, or any client or organization. The author has degrees in law and physics, and has taken several engineering courses. After five years of work as an engineer, he has practiced law primarily in the field of patents for over 50 years, dealing with a wide variety of technologies. He is a life-long railfan and user of public transportation in the United States, Europe and Japan.
[1] © John Pegram 2024.
[2] Photo from Platform screen doors Wikipedia.
[3] STV Inc., NYC Transit, System-Wide Platform Screen Door Feasibility Study (PSD Study) (February 2020), available at https://new.mta.info/document/73241. Pertinent excerpts are attached. Citations to pages of those excerpts, as indicated by a PDF reader, are in the form pp. <##/56>.
[4] In June 2023, the MTA Construction & Development Company issued its Request for Proposals (RFP) to perform the work for “Contract C32520, Platform Screen Door Pilot Installation.” The most useful document regarding this RFP that has been produced to me by the MTA is the “Program Requirements and Design Criteria for Platform Screen Doors (PRDC 16), revised as of January 11, 2024. Excerpts from that document are attached to this article. Citations to pages of those excerpts, as indicated by a PDF reader, are in the form <##/33>.
[5] See generally PSD Study excerpts, supra note 3, at pp. <26-27/56> at pp. 1-2 [3-4/3920].
[6] Id. at <2, 6, 18/56>. See also Wikipedia, Technology of the New York City Subway at nn. 40-64.
[7] See, e.g., PSD Study excerpts, supra note 3 at pp. <2, 28, 30-31, 46/56>; RFP PRDC 16 excerpts, supra note 4, at pp. <5-8 & 10-13/33>.
[8] Supra note 3.
[9] See, e.g., PSD Study excerpts, supra note 3 at pp. <1-43/56>.
[10] Id. at pp. <3-4/56>. The PSD Study notes that fixed railing barriers were eliminated early in the PSD Study due to expressed concerns by NYC Transit regarding the potential injury they could cause to customers due to door dragging incidents. Id. at p. <2/56>. However, more recently, NYC Transit installed fixed platform edge barriers between door locations on platforms at several subway stations. See Wikipedia, Technology of the New York City Subway at nn. 61-64; Pix 11, Subway barriers installed at Bryant Park station.
Vertically opening, rope-type barriers (RPSDs) were also assessed and removed from consideration in the PSD Study due to their very limited application world-wide, boarding and alighting issues, and other hazards they posed to customers. Id. at pp. < 2, 32, 34/56>.
[11] See generally MTA, Interborough Express Planning & Environmental Linkages Study (Jan. 2023) (PEL Report), Appendix 1.4, at §§ 1.2.2.3 & 1.2.3, pp. <1 & 3/9>. Pertinent excerpts are attached. Citations to pages of those excerpts, as indicated by a PDF reader, are in the form <#/9>.
The most complete version of the PEL Report that is available to the public, including all appendices, was produced to me by the MTA in response to my Freedom of Information Law (FOIL) request and is available for download here.
[12] See PSD RFP excerpts, supra, note 3, at p. <39/56>.
[13] From PSD RFP excerpts, supra, note 3, Appendix A, p. <38/56>.
[14] Preliminary Drawing PSD-G-103 in C32520-VOL 6 PRELIMINARY SET_23-06.16_Final.pdf (Times Square drawing) (copy attached).
[15] Supra, note 3 at pp. <39-40/56>.
[16] Id. at p. <40/56>.
[17] Id. at p. <41/56>.
[18] Id. at p.<39/56>.
[19] Id. at p. <41/56>.
[20] Id. at p.<39/56>.
[21] Photograph by the author, October 2024.
[22] PSD Study excerpts, supra note 3, Tier 2-3 Report on Feasibility of Platform Edge Barriers for 42nd Street Shuttle Line Stations, pp. <44-56/56>.
[23] Id., at p. <51/56>, item 19, as indicated by “Polyethylene edge strip” length.
[24] PEL Report excerpts, supra note 11, Appendix 1.11, Appendix A of Estimate Summary at p. <9/9>, as indicated by “Tactile warning strip” length.
[25] See PEL Report excerpts, supra note 11, at p. <9/9>.
[26] Federal Transit Administration, National Transit Database, 2023 Annual Agency Profile - MTA New York City Transit, available at https://www.transit.dot.gov/sites/fta.dot.gov/files/ transit_agency_profile_doc/2023/20008.pdf (copy attached).
PSD's aren’t going to happen, There will be major pushback from the unions with not having direct access to the tracks for signal upgrades/track maintenance. Then there's freight rail which shares trackage with the proposed IBX and there is no way that they'll go with it.
I'm not sure if extrapolating from PSD retrofit costs will be that accurate, even though you removed the many parts that won't apply to the IBX. Have you compared to the JFK Airtrain costs (in NYC, but older), or REM (in North America and very recent)? Or any other recent airport people mover with PSDs in the US?