How can we use technology to help with subway station crowding?

Published on: July 12th, 2016 Written by:

CoverPhoto_IMG_5302
Like most New Yorkers, I’ve noticed that subway stations are getting very crowded these days. As a UX designer, I wanted to take a closer look at this problem, and to strategize ways to address it.

To solve a problem like this, I wanted insights from a whole lot of actual commuters. I opened an online survey and promoted it via social media. The link to this survey was retweeted by the popular Second Avenue Sagas blog. The survey stayed open for a week and received 144 respondents.

A disclaimer I want to make is that some of these numbers might be skewed, as many of the respondents came to the survey via twitter and likely care about transit related issues. The survey was only conducted online. A survey that reached a wider audience might yield somewhat different results (or the results might be more or less the same).

Chart_MoreorLessCrowded
One of the broadest questions I was was if the respondents felt that subway stations were getting more crowded. An overwhelming 96.5% stated that this is indeed true.

Chart_DoYouhaveASpecificRoute
86.1% of those who responded to the survey reported that they rely on a specific subway route to get you to and from work. This might seem like an obvious question to ask, but it establishes that the average commuter routinely adheres to a specific route to get them to and from work — which is important to note.

When this routine is broken, some respondents indicate strong emotional reactions. 14 stated they would bitch and moan, freak out, or just plain cry. This is obviously pretty sad, and hints at a general depth of misery in commuting. Only 8 respondents indicated Uber or a Taxi would be their go-to alternate route, and some would only take it to the next closest running subway line. Likewise, Citibike was only mentioned as an option by 2 respondents. One reported their alternate commute as walking 2 miles. This person specified the time it took including time waiting for traffic lights, so they have clearly made this trip enough times to know how long it would take on foot. Another single respondent indicated driving as their ‘Plan B’ commute option. Only 5 respondents indicated working from home was an option. In an era of mobile devices, laptops and video chats, this number also struck me as low.

Chart_BeforeYouLeaveDoYouCheck
The next finding from the survey surprised me: Only 56.9% of respondents check for service disruptions online before they leave home or work. Of those who did check before leaving, the MTA website was most commonly cited as a source for information (62.3%) — this was followed by apps and other sources, such as social media. A very small amount of respondents (7%) checked for service disruptions via traditional media outlets (NY1 or news radio stations, for example).

Chart_IfYouCheckDisruptionSources

The good news here for the MTA is that their website and twitter account are clearly valued by commuters who do check online. As someone who often checks online before leaving, I can say that the usefulness and accuracy of this information often great. The bad news is that only 56.9% of commuters are actually checking these sources before commuting. A really simple potential fix for this would be to increase public awareness. It would cost the MTA next to nothing to create advertising posters to be placed in subway cars and additional announcements promoting these invaluable sources of information for commuters. If more commuters used these free services, they’d be able to make better informed choices. (Yes, sometimes train delays only start when you’re on your way to the train or already on it. Solving that problem is obviously harder, if not impossible).


Crowding Conditions

On the topic of station crowding, I asked two questions: “If you can see a platform is crowded before swiping, do you still pay to get on?” and “If you knew a station was crowded before you left home or work, what would you do?”.

Chart_IfYouCanSeePlatform
was surprised to find that in situations where the commuter could see that the station was crowded, 77.8% of respondents indicated they would still pay to go wait on the platform. Back before the internet, a crowded subway station was often viewed as an indicator that trains were delayed, and that it might be worth taking an alternate route instead of paying to get on and wait (and wait and wait with no announcements or smart phones).

I would hypothesize three reasons for this:
1) commuters have come to expect stations to be more crowded today than in the past.
2) After walking, cycling or taking a bus to the station, commuters are already invested in taking a route.
3) The particular station they’ve arrived at is likely their best or only route to work or home.

Things get more interesting in response to the second question. If respondents had a way of knowing that a subway station was crowded before they left home or work, 41% indicate they would take an alternate route. 36.1% indicated that they would still wait it out at the station, while 17.4% state they would leave home or work later to avoid the crowd.

Chart_IfYouKnewBefore_Crowding

When you look at these two questions as a set, well over half the people who would have originally waited at the station are now taking alternate routes or delaying their commute. By finding a way to inform the public of station crowding, we might be able to reduce the crowding via stemming the flow of additional commuters arriving at a crowded station.

Another key insight here is the sweet spot for informing the public. If commuters know of service changes and crowding before they leave, they are significantly more likely to consider or take alternate routes. If commuters only know of a problem or crowding when they arrive at the station, it’s too late — as most will simply elect to wait it out at the station.

Potential solutions, from the silly to the serious
So how do we fix this? I had some fun sketching out bad ideas on how to let the public know if a subway station is crowded. Pardon my rusty art skills, but let’s take a look:

Periscope
Station Periscopes: see if an underground station is crowded before going down the stairs.

Drones
Super Subway Selfie Sticks and Drones: for finding out if an elevated platform is crowded.

As silly as these are, they are bad ideas in that a commuter would still need to be just outside of the subway station to find out about crowding, which as we’ve seen is simply too late.

A better solution would be to utilize existing MTA technology to provide these insights. There are at least two potential methods I can think of.

Cameras
Cameras
Many subway stations are now covered in surveillance cameras. If commuters could view these cameras from an app or website, they’d be able to see station crowding in real time. I suspect this idea would meet political resistance – being that the are cameras were installed for the intended purpose of surveillance. However, the MTA does provide cameras for traffic conditions on its bridge and tunnel crossings. Likewise, the city also provides real time traffic camera access via the NYC TV Drive cable TV channel, so it is not entirely without precedent. If the security cameras are a no-go, why not an separate system with limited views, say, 2 or 3 per platform? It would cost money to build, but it would provide real time information to commuters. Some companies already utilize this technology for their customers. Anyone can see how long the line is at Shack Shake via this method.

Algorithmic
CrowdingMap
Another method might be mathematical. Leveraging existing Metrocard payment data, it should be fairly simple to predict which subway stations are the most crowded at specific times. Some of this data has been published, but there’s no public facing, real time live data available at the moment. One measure would obviously be the number of people swiping to enter each station each hour, while another would be to look at commuting patterns to establish an estimate of crowding at transfer stations. These measures could be synced up with the existing service alerts to offer predictions on unusual crowding conditions. This would obviously be less direct than the video feed option, but it would offer insights that are not currently available to the public.

No better source than the MTA
Speaking of the availability of transit information, the MTA itself is the best source of this information, though when they provide this data to third parties, the results can be even stronger. I took a detailed look at 11 free iPhone apps which provide subway and bus scheduling information and service alerts. The two MTA-developed apps I reviewed were decent, but still didn’t contain nearly as much information as some privately developed apps (Citimapper and Transit topping the list of free apps, though paid apps such as Kickmap, Next Stop and iTrans also seem to do well). All of these apps rely on the MTA’s data concerning train schedules, real time train locations (for routes with countdown clocks) and GPS data for buses. The MTA seems to do a great job of providing this data to the public. Station crowding data would just be one more stream of information. The key thing to note here is that none of the apps on the market today account for subway station crowding. This lack of information is due to the data not being accessible from the MTA, likely because no one seems to be asking for it yet.

Conclusion
As the population of NYC continues to increase, ridership numbers on the subways will also continue to climb. Subway station crowding is an issue that is not going to disappear, and is only going to get more dangerous over time. Fixing this situation need not only be the responsibility of the MTA. Large companies could affect change by promoting more work-from-home time and varied work hours. Meanwhile, the MTA should more actively promote the existing service change information on their website and social media, and devise a way to inform commuters of crowding conditions before they leave for the train station. This information will allow for better informed commuter decisions and expectations, which would reduce some of the stress associated with commuting.

2 responses to “How can we use technology to help with subway station crowding?”

  1. Mitch45 says:

    The MTA is a victim of its own success. The agency essentially rebuilt the stations, banished graffiti and brought in comfortable air conditioned cars, while the NYPD reduced crime to negligible levels. The masses flocked back in and now there’s a problem with overcrowding. The MTA did too good of a job.

  2. Alex says:

    If you become the mayor of this zoo, then ima leave a flaming bag of doggie do on city halls door step.

Leave a Reply

Your email address will not be published. Required fields are marked *

  • About The Author

    Joseph Anastasio

    Design & History nerd, open space & infrastructure advocate. 
  • Recent Comments

  • Check out our Twitter feed

  • Social

  • Instagram Feed

  • Featured Press