My bus stop
New technology can have a profound impact on an old industry–in this case public transit. No single innovation has improved bus transit service like GPS. It has transformed bus operations and the customer experience. Here’s how.
Reassuring the bus rider
Taking a bus trip requires some trust on the part of the customer—trust that the bus will show up when the schedule says it will. Before the advent of GPS, there was always that shadow of doubt. What if the bus is late? Did I already miss it because it ran early? What if it broke down and isn’t coming at all? This sort of uncertainty is a powerful deterrent to riding.
GPS—or AVL (Automatic Vehicle Location) as it’s known in the transit industry—changed all that. Unlike regular GPS in a car that just gives the location, AVL tracks the bus location and compares it to the schedule, a crucial additional step and much harder to do. For this story we’ll just call it GPS.
Now the customer can track the bus and get a constantly updated, estimated time of arrival. It’s available on any computer, touch tone phone or mobile device. Real time displays are being installed at the busiest stops. The mystery and uncertainty is gone, replaced by peace of mind. Physical discomfort is minimized as well. On a cold or rainy day, if the bus is running late, you don’t have to wait out in the weather. Stay indoors until a couple of minutes before it arrives.
If you’re a regular bus rider, you probably knew all that. If all GPS did was reduce customer uncertainty, it would be a big deal. But there are many more benefits, all behind the scenes, that have greatly improved how buses operate.
Behind the scenes benefits
Every bus now has a display that gives the driver continuous feedback on his/her on-time performance. It’s that little screen next to the farebox. If you stand behind the driver you can watch it and see how the bus is doing compared to the schedule. This is a labor saver for the driver, who previously had to manually compare a run card to his watch, and had to synchronize the watch daily with a master clock at the garage. That same information is simultaneously available to the Transit Control Center, which supervises bus drivers while they’re on the street.
Having both the driver and supervisor getting a continuous on-time status has eliminated or sharply curtailed the small minority of drivers who had intentionally run off schedule, bad practices that had gone on for over a hundred years. There were never enough on-street supervisors to prevent it.
Some drivers would run early, a major sin, because it reduced the number of passengers they had to carry. The following bus would get overloaded, fall behind schedule and soon there would be two buses running together and a big hole in the service. Now the computer immediately flags anyone who does it. The result—virtually no early buses.
Or a driver might intentionally run later than his scheduled follower, sticking the follower with the load. These practices tended to be worse on lines operated out of two garages, since the offending driver was less likely to meet his victim at the garage.
Before GPS, some drivers would arrive late at their terminal but still take their full layover time between trips and leave late. Their justification was that this is their break time. It isn’t, and never has been. It’s “recovery time”, used to get back on schedule after a late trip. Thanks to GPS, that practice has been drastically curtailed. Also virtually eliminated is the practice called “shortlining”, failing to complete a full trip, usually the last one of the day.
Besides its day-to-day use, the GPS data is archived, which allows it to be analyzed. This has all sorts of benefits. Before GPS, scheduled running time was based on a very limited sample of manual ride checks, load checks at selected points along the line, and complaints from drivers and customers. Now the on-time performance of every trip, every day at every time point is known. A complete data set means a much more accurate schedule can be written.
Like anyone else, bus drivers run the gamut from slow to fast. Because of that archived data, the slow ones who habitually run late can be identified and retrained.
Put all these things together and on-time performance (defined as zero minutes early to 5 minutes late) improves. In fact GPS has improved it from about 80 percent to something approaching 90 percent.
Another good thing about GPS—it has completely changed the investigation of customer complaints. In the past it was impossible to determine if a complaint about the bus being early, late, or even off route was valid. Maybe the customer’s watch was a couple of minutes off. However, with GPS, the customer service representative can replay the bus GPS and much of the time can get to the truth while still on the phone. The customer gets the correct answer. If it’s the driver’s fault, he hears about it. If the customer was mistaken, it ends there and the driver doesn’t have to defend himself against a wrong accusation.
One last GPS benefit is currently being tested, automated street calling. It’s Metro Transit policy for the driver to call out streets that have a traffic signal or a stop sign. Most do, some don’t, so it’s been a supervisory challenge for years. Now the GPS determines the location and an automated voice calls out the street. It can add information, such as landmarks and connecting routes. GPS street calling is being tested on Nicollet Mall routes, and will be implemented systemwide.
I learned about seven new things about bus drivers in this article. Thank you! I wonder if the drivers themselves like the GPS tech? It seems to have taken all the naughtiness out of the job.
Do they publish the raw data? I’m sure it could be requested, but it’d be nice if it could just be downloaded from the website like Nice Ride does with its annual data set release.
The data is all in-house. During snow emergencies they publicize a system on-time percentage to give riders an idea of the delays.
If you *really* want it, you can start gathering it now from the NexTrip API: http://svc.metrotransit.org/ With hundreds of buses per day, and data on each every 30 seconds, I can see how that’s going to get big pretty fast though!
Who knew bus drivers had such underhanded tricks? (I’m sure it’s not ALL drivers.)
Could this data be used by traffic planning engineers to identify traffic patterns more cheaply/quickly?
Please note that in the story I said it was a small minority of bus drivers who misbehaved, but it’s less now.
I don’t think the data would be useful for general traffic planning. Because of stops on local routes, bus speeds don’t resemble auto speeds. On freeways with express service, the highway is already instrumented to captured speed data.
Good article!!! I understand and appreciate the benefits that the “GPS” CAN have, however I am convinced Metro Transit does not use this data properly. In the age of Google Maps, people expect this technology to work accurately and improve over time. I understand Metro Transit isn’t going to have Google level computer scientists writing code to analyze this data, but in this day and age, people expect this stuff to work, and if it doesn’t it should not be rolled out.
It is extremely common to stand at a bus stop on Marquette during rush hour and watch the board sit at “9 minutes” for over 10 minutes. Then suddenly it will drop to 4 minutes, then suddenly the bus arrives and departs while it still says 4 minutes. (these are numbers I see frequently)
There a route I take (out of 46th &35W station) where the driver arrives 2 minutes ahead of schedule EVERY day (and has been arriving 2 minutes early for years). Every time she arrives (and even as she pulls out), the board says 2 minutes. Additionally, on this same route on my way home, I don’t think I’ve ever seen the driver’s screen telling them they are at better than “-5” (minutes behind schedule). Most days its far worse than -5. This should be an indication the schedule needs to be updated to accommodate rush hour.
Long story short, this is a great article about the things that COULD happen if this technology and data was used properly. Based on almost 2 years of trying to use NexTrip and the results never getting any better, I can say with certainty that Metro Transit does not use it properly, and thusly the only thing I can do to ensure I make my bus is ignore NexTrip estimates.
As I understand it, the buses are polled by the computer once a minute, so actual arrivals can vary by a minute or two because the computer is making a prediction of arrival time. Regarding the early bus at 46th & 35W, freeway expresses are expected to be on time in their pickup area, but are allowed to arrive early after the express portion of the trip. That recognizes the volatility of freeway traffic. It wouldn’t make sense to get to the outer edge of downtown early and then sit. I don’t know if there is a requirement to not leave an intermediate freeway stop like 46th Street early. Falls into a grey area.
At least you’ve got a number that sort-of changes. On the light rail platforms, there’s nothing but a two-minute cricket.
Great to hear about how GPS is being used to improve bus service. I’m sure it’s also being used to spot regularly occurring hang-ups in the system. How feasible would it be to develop an app that allows riders to track buses? Is Metro Transit unwilling to this info public, or would the cost simply be too high?
Hopefully someday MetroTransit can apply this fancy new GPS technology to their trains!
Nextrip, and the third party apps are great, but I’d love it if I could actually see the current bus location on google maps too. A problem I’ve noticed with nextrip is that if the bus is really late, it will eventually disappear altogether from the real time schedule leaving you wondering if it’s still coming. It happens a lot on wintery days when the buses get backed up on the mall, you wait 30 minutes in the cold, then your bus vanishes, drives me crazy.