Among the transit cognoscenti, the promise of countdown clocks on the B Division has always seemed frustratingly out of reach. While the A division — the numbered lines — received countdown clocks as part of a signal upgrade a few years ago, the upgrades for the lettered lines are still decades away. Meanwhile, every few months, the MTA would promise some sort of interim solution in “three to five years.” Well, three to five years may finally have arrived as the MTA and Governor Cuomo announced a Bluetooth-based, eight-station 90-day pilot program for B Division countdown clocks.
The pilot — along the BMT Broadway line (N, Q, and R trains in Manhattan) — will feature to-the-minute countdown clocks with similar information as the ones on the A Division share but a different design. The information will be delivered from data transmitted by Bluetooth receivers on the trains to those on the platform and then fed into the digital displays. The countdown timers won’t be based off of data received while trains are in between stations so precise train location will still be an unknown, but the data should be reliable enough that passengers won’t know the difference.
Here’s how the press release described the technology:
The new clocks rely on technology that is straightforward, cost effective to deploy, and does not require large infrastructure. The system uses the existing wireless network in the stations and cloud computing, and involves four Bluetooth receivers placed in each station, two at each end of the platform. These receivers communicate with four Bluetooth devices that have been installed in the first and last cars of each train set running on the line. As the train enters and leaves a station, the system uses its arrival and departure time to estimate the time at which the train will reach the next stop in the line, and display the arrival times on the two LCD display screens that have been installed at each station.
The new displays, as you can see from the photo above, will feature the countdown timers but can also show PSAs and other contents (such as ads) concurrently, solving a major design flaw inherent in the current two-line displays. Now, when the MTA wants to issue a not-so-important message from the NYPD, it can do so on the portion of the screen that doesn’t include the countdown information.
“These actions,” Cuomo said in his press release today, “are the latest steps toward rebuilding and transforming the MTA into a unified, state-of-the-art transportation network that will meet the needs of current and future generations of New Yorkers. With this new and updated technology, we’ll help ensure riders have the information they need to get where they need to go.”
As part of the 90-day test, the system will be in use at the N/Q/R stations at 23rd Street; 28th Street; 34th Street; 42nd Street; 49th Street; 57th Street; 5th Avenue/59th Street; and Lexington Avenue/59th Street. During the evaluation period, the MTA says it will “identify and correct any issues with the new system. The goal is to evaluate the accuracy of location data, performance of Transit Wireless infrastructure, performance of the LCD displays, physical and network security of Bluetooth devices, security of data being transmitted, and internal access and use of data being generated.”
The governor says these clocks will ultimately be installed at all 269 stops along the lettered lines, but it’s not clear on what timeline these could be rolled out or at what cost. It is markedly cheaper than the CBTC upgrades, but unlike the CBTC upgrades, Bluetooth-based countdown clocks don’t increase service. They are a customer satisfaction measure through and through, one that both is welcomed and shouldn’t have taken so long to realize. But with Cuomo’s push to roll out Transit Wireless at all underground stations by the end of the year, this style of countdown clock became feasible. It is not yet clear how these could be deployed in stations that are above ground.
Still, the MTA appears committed to this way forward, and although I don’t always agree with the Governor’s transit priorities, he deserves praise for finally getting the MTA to move forward with technology projects that had been stalled for years. “Governor Cuomo challenged the MTA to develop an aggressive approach to putting countdown clocks on the lettered lines, and our technology team’s response has been phenomenal,” MTA Chairman and CEO Thomas F. Prendergast said in a statement. “In very short order they developed an easy to deploy, cost-effective system that we think will play a central role in bringing this essential service to more and more of our customers. We look forward to learning from this test, as well as to developing a roll out plan based on our findings.”
It seems as though the main problem with the clocks being inaccurate would be if a train gets stopped/stuck between stations after it’s already signaled its departure from the stop just up the line. If the route is fairly straight — like the N/Q/R between Times Square and 23rd Street — that shouldn’t be an annoyance with the clocks, since passengers on the platforms can see the train’s lights in the tunnel.
It might be a bit more aggravating, say, between 57th-7th and Fifth Ave. if the train’s stopped for some reason, while the countdown clock continues to show it a minute away and the people on the platform can’t see any indication of a train in the tunnel because there’s a curve between the stops. But that’s not much different from the same thing happening with the countdown clocks in the A division right now.
That should be easy to handle with software only, the system is already set up to calculate an estimated arrival time based on the departure time from the previous station(s). So, if the train doesn’t arrive at the next waypoint in that time, the display can switch to read “Delayed” until it is “seen” again.
Not a capacity upgrade, but definitely an improvement that can assist with some of the frustration associated with the blind transfer. If I know the 8th Ave train won’t be at W 4th for 6 minutes, I’ll continue on the F to Herald Square and walk west.
Would also like to see integration with BusTime for last mile routing questions.
Amazing news! This project has taken way too long to complete and I’m glad we’re finally moving forward with this. Countdown clocks at station are a huge customer service improvement for riders. This should be at all stations. Great news.
Every time something like this rolls out (like BusTime), I’m so conflicted. On one hand, I’m excited this is finally happening, and kudos to the MTA for figuring out an interim step. On the other hand, this technology (either Bluetooth or RFID) has been available for years and years, and has been suggested even by people on this and other boards for years.
In fact, this isn’t *terribly* different from what’s already done in some 8th Ave stations on CPW, where a train trips a sensor which sends off an announcement that a train is 1/2 stations away.
The biggest difference is that this system identifies the train’s route, which is a huge improvement. It also displays it on a screen so you can make a decision rather than waiting for an annoying recorded voice.
Glad to see more than two lines of information on the screen, but you could get a second column by getting rid of that I heart NY logo.
I think the logo is replaced by ads. I see the weather info on the bottom line, but if that’s where the ads are going to go and the ny logo is remaining, then hey, that’s good (this way the ad isn’t in our face). But, yes, getting rid of the logo would open up the screen to more useful information. What might go there? Tourist info for such places? Bus transfer info? One thing I hope is happening is that all this info (and ads) are SILENT.
The thing is before there was no way for the information to be communicated throughout the system. But with WiFi being installed at every station, this provides the opportunity to piggyback that infrastructure.
Yesterday, I was at the 36th Street D,N,R station in Brooklyn and noticed that the big information panels had time estimates for when the next train would arrive. I don’t know if that had anything to do with this program or was just estimated things according the usually useless printed schedule.
In any event, this is certainly good news. I wonder also if there are plans to change the green LED countdown clock signs on the A division to the really nice LCD panels we see above.
A couple of questions:
1. Most of the ‘subway” system outside of Manhattan, runs outside. (J/Z/B/Q/F/D/N in Brooklyn and Queens, the IRT lines in the Bronx etc.).
Why can’t the MTA simply use on all those lines (about 40% of the whole system) what it’s already using for BusTime, or a simple signal, that would sync with Waze?
2. The LIRR stations have all had countdown clocks for years – what technology have they used, and why was it incompatible with the subway system?
3. Ben, do you remember the link that you put up some six months ago for an in-depth review of the delays for clocks for the B-Division?
This is an illustration of why we need to have a PR to dispel myths promoted by the MTA (two big ones: that existing capacity doesn’t allow for more trains; and that CBTC will increase capacity – both are not true).
1. The MTA can put GPS on all the trains and track their locations, but 60 percent of the system runs underground, and so the countdown system would essentially be useless there. I would imagine that any countdown clock system would have to be uniform, not piecemeal.
2. The LIRR’s countdown system isn’t really a countdown system. It displays the arrival times of trains based on the schedule. Lirr/MetroNorth trains run much less frequently than the subway and so it’s easy to tell riders when the next train is coming.
This is great and much needed improvement. Even though it won’t technically improve the service, it will in fact reduce riders’ travel times. By having information about arrival time, riders can optimize their routes on the fly.
I don’t believe the outdoor stations will be difficult to integrate into this scheme as they can use the cellular data network. The system would work similarly to the one currently being tested, except the Bluetooth receivers at the station would use the cellular data network to communicate with the countdown clocks at the next station. The system could even be made to be more accurate by installing cell-network enabled Bluetooth receivers at different intervals between stations, although running power to these devices in between stations may be too expensive.
The major downside to using the cellular data network is reliability, as any interruption in cell service will tank the system outdoors. However, I don’t see this being a deal breaker as this isn’t a mission-critical system such as the signal system. If there’s a cell outage and the countdown clocks are down for an hour the trains will still run, people will still get to work, and life will go on.
The blue HelpPoints are installed outdoors even though there is no TransitWireless infrastructure on els. The MTA plans to install 4.9GHZ staff-only wifi at all outdoor stations anyway for NYPD/roaming NYCT employees with job issued tablets or phones, so Div B bluetooth countdowns and HelpPoints will run on a non-public 4.9 GHZ wifi network I think.
I’d like to see actual tech specs. Some edumacated guesses: BT receivers at stations, at end of each platform, times how many platforms, local and express? Lets pretend its 200 platforms (it’s not), so that’s 400 individual receivers which are either battery powered or wired into the station infrastructure. Then we need BT transmitters on each end of each train. Heh. How many is that? 1000 cars that need equipping? 3523 cars on the B division according to the internets but not all lead I don’t think, so lets use 1000 (may be fewer). Again, either battery or wired into the car’s systems. I have no doubt they the station equipment will squirt their data back to the mothership via the station’s Wi-Fi, so that’s a relief. Regardless, there are still literally *thousands* of transmitters and receivers that need to be maintained. Yikes. So much for not requiring “large infrastructure”.
Fun idea, and for an interim maybe it’s OK, but it really needs to be a part of the railroad’s signal system
Only cabs need the BT transmitters. MTA might be cheap enough that the conductor “cabs” might be forever missing the equipment (so 1 more nail in the coffin against ever seeing a 4/5 car NTT train). I believe conductor cabs are already missing some “redundant” equipment like rescue coupler adapters.
Will this arrival data be relayed to the OnTheGo kiosks (that currently just show scheduled arrival times on the B Division)? For that matter, does anyone even use those kiosks?
Although I know that they’re wildly inaccurate, I occasionally use the On the Go kiosk to help decide if I should take an R at Union Sq to downtown Brooklyn if there are delays on the 4, 5.
Will the information be publicly available outside the station as with BusTime? (i.e. through apps like iTrans)? This would make it helpful for deciding which station to head to (for example I have to choose between Nevins and Dekalb when catching trains into Manhattan)
I’m very curious to know how they implemented this.
Has anyone seen what the hardware looks like? I can’t find anything new-looking that’s installed at all of these stations. Where is is located?
I’m also curious to know what hardware they’re using. Raspberry Pi?
These look like the same type of LCD displays with a much uglier UI than Chicago’s version. Hopefully a designer will pretty it up a bit.
CTA did a really nice job with the interface: http://www.transitchicago.com/.....pdate.aspx
That said, even having this information is a big leap forward for the MTA.
Not sure they spent enough years overthinking this. Bluetooth? Yeah, this is going to work GREAT. Have these guys ever actually used Bluetooth before? I’m sitting here not 6 feet away from my mouse and keyboard and they constantly connect and disconnect if I don’t turn Bluetooth off.
My far simpler and more reliable plan that could have been implemented for pennies 5 years ago:
(1) Put (reasonably) large QR code stickers on the sides of the train cars. 2 per side per car ensures there will be at least one sticker to scan successfully for any given train. Since we only need to encode a short number, we can ramp up the error correction, so even partially covered or even destroyed stickers can scan successfully.
(2) Slap together a halfway decent camera + bright light + arduino or raspberry pi that does nothing but scan the video feed for a QR code and then shoot the decoded value off to a central system.
Call it a day.
That sounds like the old KarTrak system used on freight trains. It wasn’t as successful as you’d think; the industry has moved to passive RFID identifiers, which are much less susceptible to grime.