October Performance

Lots of interesting tidbits in TriMet’s October Performance Dashboard.

Ridership Changes

Ridership statistics

MAX ridership has gone down, no big surprise since there’s no more free rail zone so people won’t necessarily wait for a train to get from one point downtown to another if a bus shows up first (or if they just walk…) What is surprising is that despite service cuts in September, there are more people riding buses than this time last year when there was comparatively more service. WES is doing marginally better, but at $11.69 cost per boarding ride and an essentially flat graph, it’s nothing spectacular.

On Time Performance

I’d modified my last post with this graph (at the time I published that post, only performance through September was available, but October’s data made it even more interesting). In an Oregonian article, TriMet spokesperson Roberta Altstadt attributes the cause to external issues (e.g. cars crashing in the ROW, and as I said on Twitter, it’s a nice change of pace that the most recent drunk dumbass to go off-road didn’t do it over MAX ROW) and the inexperience of new operators. That new operator inexperience is not even just an issue of familiarity with the alignment – new operators are more likely to have rule violations that delay their trains (e.g. tripping a signal), and are generally not as capable as most seasoned operators are at quickly and accurately describing to Control any sort of mechanical issue they might have which leads to a delay in fixing it. Yes, these are problems that time and experience will help (until they hit the complacency mark around 6-12 months when rule violations often spike again, but that’s another issue…)

So while it’s true that inexperienced operators are part of the reason for the downward slope in MAX on time performance, that doesn’t really get at the root cause of WHY there are so many new operators out right now. TriMet went for about a year and a half without having new rail operator classes, and lately they’ve been run almost back to back due to operator shortage. It’s the same at bus – a long hiring freeze on bus drivers and now oh dear, there aren’t enough operators. So a hiring rush ends up where there’s a lot of inexperience on the roads and rails at once. This could have been avoided if more focus was given to actual operation and new people had been added at a steady, constant rate instead of in rapid succession after a long period of none at all… and I think that’s one of the contributing factors to this:


Bus collisions per 100,000 miles

Ok, if we’re serious about this whole “Safety is  a value, not just a priority” and that’s not just a catch phrase to try and look good, then this warrants a long and hard look. Bus collision rates have been consistently higher (with 2 exceptions) than last year for the past year. After the Sandi Day incident, TriMet head of training Allen Morgan developed an annual bus operator recertification training program, which theoretically would reduce the number of bus accidents. Well it’s a wonderful idea, but it doesn’t work. Or maybe it does work but it’s the initial operating training that isn’t adequately preparing new drivers. Or maybe there’s just too many new drivers at once due to the hiring freeze. Whichever it is, this trend of increasing collisions needs an immediate response, and not just a safety committee that’s all talk and no action.

And so it doesn’t look like I’m just picking on bus and leaving rail alone: Even though it’s not one of the graphs presented in the dashboard, let’s take a look at rail rule violations as well. If there’s a similar trend, then we need to stop seeing how many times we can fit the word “safety” into a speech and actually do something to improve safety.


Tax Revenue

I’m no financial analyst, so maybe I’m looking at this all wrong, and if I am, feel free to correct me. But after that whole panic attack about TriMet having somewhere between $12-17 million budget shortfall… it looks like the actual revenue is almost $16 million over the budget for FY 2012, and about $3.5 million over for FY 2013. Even taking into consideration the passenger revenue chart which shows the budget being slightly higher than the actual (about $2.3 million for FY 2012 and $1.1 million for FY 2013), it doesn’t look like we were anywhere near being short – if so I would’ve guessed that the actual would be under the budget, or perhaps taking the recent cuts into consideration, about even. Instead it appears to be well over. What’s the story?

8 responses to “October Performance

  1. Concerning the last chart, the FY2012 budget was for last fiscal year ending in June, so it doesn’t really factor into the “shortfall” equation. However, the new budget reflects the changes they made to make up for the shortfall, but I think the fact that revenue is higher is why, for example, when you call TransitTracker or go online you don’t see or hear advertisements like they said they were going to add. Appears they didn’t need the $300K after all, so they just haven’t done it. Why that money wasn’t cycled back into service instead is anyone’s guess. But I think the chart does show that the budget is more accurate than it would have been had they not made the changes. Maybe for the spring sign-up they’ll cycle the money back into service.

  2. So the fact that I have been always 15 minutes later getting everywhere these past two months isn’t just in my head. However, I have noticed that slowly but surely service has been getting more on time recently.

  3. How are ridership numbers calculated/estimated? Especially considering if they are reporting a decrease in MAX ridership partially attributed to people not using it for quick downtown trips. For example, how would someone riding from Goose Hallow to Library be counted as a trip?

    • I know that there are passenger counters at the train doors, and TriMet has provided ridership by offs/ons at each stop (see PortlandAfoot for a listing of stops and ridership for 2010). I’m not sure exactly how those work though, and how much of those are calculated mathematically versus being physically counted by the door counters. There had been a directive put in place after the Green Line opened that the Type 3s and 4s were being used to calculate ridership numbers, so Control was to avoid coupling 3s together, and Type 4s and consists with 3s should be distributed on all of the lines, so I’m guessing based on that some of the ridership numbers have to be inferred (e.g. in a Type 3/Type 1 consist, I would assume the ridership numbers would be multiplied by about 2, using the passenger counts from the Type 3 and assuming roughly equal passengers in both cars).

      • It would be really silly (and just like management) to take numbers of passengers on a type 3/type 1 consist as being equally distributed on both cars except maybe during crush load situations. The type 1’s typically have about 1/2 to 3/4 the number of passengers of the low-floor they are coupled to.

  4. Automatic Passenger Counting equipment, manufactured by INIT – Innovations in Transportation GMBh (Germany) are installed on car 222, all type 3s, type 4s and will be installed on the type 5s.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s