Sunday, October 20, 2013

The Lost Bus

I remember the story of a guy who received paperwork from the government because, apparently, he had died. He had to work really hard to convince the government that yes, in fact, he was still alive. Well, this is a similar situation. There is a bus that the system thinks has never been used when, in fact, it has been running for four days now.

The bus is 3159. It is one of the new Gillig buses running out of Merlo Garage (the newest, actually). I have been watching the new buses on the Rose City Transit SystemMapper as they get their CAD/AVL systems installed at Powell Garage, get transferred to Center Garage for some final preparation (probably farebox installation), and then are moved to Merlo Garage. This bus got its CAD system installed, then was moved to Center Garage and seemed to sit there for some time.

Thursday morning, I checked the bus feed to find that all buses up to 3158 were in service, 3159 was still at Center for some reason, and 3160 was still lacking a CAD system (thereby placing its coordinates at 0,0, which is in the Atlantic Ocean just south of the African nation of Ghana). At about 4:23pm that day, I was walking down the street talking to my mother on the phone when bus 3159 turned left from SW Harrison onto SW 6th Ave as a line 56 bus. I was shocked to see that it had gone from Center to Merlo and out on the roads that quickly. But when I looked at the SystemMapper, it was still broadcasting as being at Center (buses continue to broadcast their true current locations even when the time stamp isn't up to date), and there was no 56/69 on the map where it should have been. Of course, I wrote it off that I must have misread the vehicle number. Maybe I had seen 3149, and confused the two.

Later that night, I was at Pioneer Square North when I looked down the road and, behold, 3159 was coming toward me. It still was registering on the app as being at Center. I have seen it every day since, either on a 54/56 or 57, and it still showed itself as being in the north lot of Center Garage. I wanted to go find the bus to snap some pictures and ask the operator if anything was weird on his/her side, but it's kind of hard to do that when I can't track the bus to go find it.

Tonight, I was coming home from PSU when I happened to check the SystemMapper Interactive to see which buses were sitting at Beaverton TC as I passed by. It showed one, 2940. When I pulled in on the MAX and saw a 3100 sitting in the 52 bay, I knew immediately that it was my bus. Although it left before I could get to it, I caught up with it at Willow Creek TC. The bus driver was nice enough to let me get on board and take the picture of the vehicle ID number (as it was night and I couldn't get a good exterior shot).

So, all this to say, here is proof that this bus has absolutely no idea where it is:

Notice that on the picture on the left, my watch says 9:49:11 on 10/20 (today). The right screenshot is time stamped as 9:48:48 on 10/20. My watch is running a little fast, so in reality I snapped the left picture about 5-6 seconds before I took the screenshot on the right. Regardless, you can see clearly that this bus has no idea where it is.

Now, I asked the driver if she had noticed anything strange on the CAD system. She told me that she had not noticed anything, and that everything seemed to be running smoothly. However, the app data being sent on the new Vehicles feed (which the SystemMapper uses) isn't showing it. As of yet, the only times when there are glitches in that feed are when (1) the bus dies somewhere where it has to be completely turned off, even the CAD system (i.e. when 3008 got crashed into head-on at Clackamas Town Center), or (2) when the CAD system isn't sending smooth location data (I have only seen this happen with one bus on one day; even if the bus is having issues signing in as a block or a line, it still sends accurate GPS data).

The point is: don't go looking for bus 3159. You will have no idea where to start.

UPDATE: On October 21, I was contacted by an insider at TriMet who informed me that the bus was not showing up on internal systems either. I was usually able to find the one Merlo block that didn't show up on any of the trackers or maps and decided it was probably 3159 running on it. On October 24, I was informed by the same insider that the bus was now registering on the system, and I confirmed this as I found the bus sitting in Merlo Garage. It has since moved within the garage, telling me that the system is active.

You can find the current location of the bus here.

One more to prove: this bus is running!


  1. What's interesting Patrick is that you actually know this is true!
    Incredible that you possess such detailed knowledge of the bus fleet

  2. Hi Patrick,

    A correction on this piece for you. The system wasn't at issue here, it didn’t have any position information because the radio wasn’t working. The operator should have known the radio wasn't working correctly. This is likely a training issue.

    If you are interested in learning more about the nuts of bolts of the new CAD-AVL system from the engineers who work on it everyday, please let me know. Glad to get you plugged in.

    Diane Goodwin


Please keep it clean! Thank you.