Wednesday, August 25, 2010


Detailed route map of 15-Belmont/NW 23rd, prior to service cut.  Courtesy of TriMet.

By now, Portland readers have probably heard quite a bit about the controversy surrounding TriMet's sudden decision to eliminate the NW Thurman branch of the 15-Belmont/NW 23rd line, due to stated safety concerns.  Residents along the affected stretch of Thurman are understandably upset about no more bus service (for some, the cancellation means a 3/4 mile hike to the nearest bus); and some have suggested that this is little more than a PR move, not a legit safety issue--as busses have been running this route for a long time without any major incidents.

I have serious questions about the lack of notice given to the public--it wasn't cool.  If this was done on advice of TriMet counsel, they should definitively say so. (On the other hand, I wondered over at if this might be a stealth service cut designed to save money; as TriMet appears to be simply routing all 15s on the Montgomery Park branch instead, it appears this is not the case--service hours are not being reduced).  TriMet appears to have gotten the message, and has announced plans to re-open the line starting Monday.  (If it isn't obvious, this post has been in the works for a few days...)

But this post isn't about the Terrible Thurman Truncation of '10.  It's about a possible solution to this problem, and to other problems, such as discussed in this post.

A possible solution:  A "micro-route"

Over in the same thread, I made the following suggestion:

My proposal would be to create a new line, which starts at PGE Park (where one can transfer to MAX or one of several other bus lines), takes 18th/19th to Thurman, and then heads up into the hills to the turnaround. And operate it with the paratransit vehicles. This line would be a short enough line that you don't need a full 40' bus, and the smaller vehicles, I would think, would alleviate the safety concerns. And add service to 18th/19th.
The details of the proposal don't matter much here--the 18th/19th routing and the PGE Park connection are features designed to provide service on a currently un-served street, and a connection to MAX; on further reflection a route down Everett/Glisan to Union Station might work better, simply because that routing provides a place where busses and drivers can take breaks.  (There's no room around PGE Park to park a bus for a spell).

30' busses (courtesy Wikipedia)
By "paratransit vehicles", I mean vehicles smaller than the standard 40' models used for the bulk of TriMet routes.  The agency has about 50 30'  busses, such as those shown to your right, and over 240 minibusses used for the LIFT paratransit service.  (Reportedly, there exists a 2400 series of busses are smaller than 30', though these don't appear to be in service any more).  Use of such rolling stock on the 15--a widely used frequent service route--would be inappropriate, but for a very short route, it's likely that a smaller vehicle will have more than sufficient capacity--and be able to execute the turnaround on Thurman safely and legally.

Hence the micro-route,

What is a micro-route?

A micro-route is a route which is shorter than standard routes.   Micro-routes will often have the following attributes:
  1. Short overall length, with a round trip time of an hour or less (and often a fraction of an hour).  A practical limit on standard bus service is no more than two hours or so from one end of the line to the other, with layovers at either end.  With longer routes, reliability suffers, and you start running interfering with the driver's breaks.  One-hour-each-way routes with a branch or loop (or both) at one end, such as the 15, are also common--the 15 lays over at Gateway TC but not at Montgomery Park.  With a micro-route, on the other hand, the route is run multiple times between breaks.  A bus running a Thurman/Union Station route could probably run the entire route 4-5 times in a two-hour span.
  2. Difficult geography which makes the use of full-size busses impractical.  This can be narrow roads, steep hills, tight turns--or politically-powerful NIMBYs terrified that a 40' bus rolling past their driveway is going to bring about Armageddon.  (Don't laugh).
  3. Connection to a transit center, both to provide a layover point, and due to attribute 1, it is expected that many riders of the micro-route will transfer.  Scratch that--it is expected that most of the micro-route riders will transfer; many micro-routes function as de-facto shuttle services.
  4. A reasonably predictable and continuous load pattern, so that passenger capacity isn't an issue (or if it is, can be dealt with by modest increases in frequency).  This is especially important for minibuses, which may not be configured to permit standing room.
Examples of such systems
Green public light bus (Wikipedia)

Hong Kong is an example of a city where such arrangements are commonplace.  In Hong Kong, in addition to the regular bus service, one encounters minibusses known as public light busses.  These busses come in two colors, green and red.  The red public light busses operate like jitneys or share taxis, and aren't relevant to this post, but the green ones operate fixed schedules just like regular bus service, simply with smaller vehicles.  Hong Kong bus operators also operate some of the "franchised" (full service) lines with two different vehicle sizes, with double-decker busses serving the busier routes, and 40' single-level busses serving he smaller and shorter ones. 

In both cases, these smaller types of busses serve generally short routes, connecting (for example) a transit center or marketplace (these things generally come together) with a housing estate (a common term for a large apartment building or other residential complex) or a small village communtity, both of which are abundant in the rural parts of the region.  The service that they often run is shuttle-like in nature, in that there are two distinct clusters of stops (often a single stop) on either end of the route, without much in between.  The primary franchised bus lines generally serve provide bidirectional service along a linear corridor, just like the vast majority of the bus lines in Portland do.

Where might such service work?

Besides Thurman, where the motivating factor is a geographical limitation, where else might similar types of service work in the Portland area?
  • Office park circulators.  One big issue with suburban office parks is that they are difficult to serve with traditional transit service.  They are often spread out over a large area, have tons of parking lots, and may contain streets that are difficult for large busses to maneuver on--and serving them adequately with mainline bus service often requires deviations which are annoying to the bulk of the riders, and inefficient to the agency.  The Cornell Oaks subdivision mentioned in this post (currently served by the 67 but not for long) is an extreme example of this, but many other examples abound.  Certain runs of the 43-Taylors Ferry, which nominally runs from downtown to Washington Square, venture west of the Square to run around the various industrial parks along SW Nimbus Avenue in south Beaverton, in a circulatory fashion.  This little jaunt has nothing to do with the primary corridor served by the line.  So, given that--why not disconnect it from the 47 and instead run a circulator service from the mall to the various office parks strip malls, and other commercial centers immediately surrounding it?  In addition to the high-tech businesses on Nimbus, it could also serve the strip malls and big-boxes on Cascade Avenue, the Tigard Medical Mall and Lincoln Center on Greenburg, and even venture over to the Tigard Triangle to hit Costco, Freddies, and the like.  (Many of these retail outlets employ low-wage workers who would benefit more from improved transit than would high-tech office workers along Nimbus).  
  • Shuttle service.  Many shuttle service connecting transit corridors with off-corridor destinations already exist, albeit provided by folks other than TriMet.  The shuttle from PCC-Sylvania to Barbur TC is one example--you can take the 44, but the shuttle provides additional frequency.  (You have to be affiliated with PCC to use it, however).  But there are other examples of major destinations with inadequate transit service.  Meridian Park Hospital, for instance, is a major full-service hospital surrounded by the usual assortment of clinics, doctors offices, and other ancillary medical services--but other than two 76s per hour, doesn't get any transit service.  Kaiser Sunnyside Hospital is in a similar situation--although the Green Line terminates a short distance away, the pedestrian exprience in the area is unpleasant, and the nearby bus services (155, 156, and 157) are all astonishingly infrequent.  Most other major Portland hospitals (OHSU, Emanuel, Good Sam, St. Vincent, Adventist) have one or more frequent service lines passing by the front door.
  • Other geographically difficult areasPortland is perhaps fortunate, compared to cities such as San Francisco and Seattle, in that early city planners didn't ignore topography when creating the street grid--Portland has no local equivalent to Lombard Street in SF or Broad Street in Seattle.  As a result, most of the area is reachable by bus, and the areas which aren't are in many cases not sufficiently inhabited to merit bus service.  But there are a few places which come in mind.  Were the Lake Oswego Streetcar project to be built, a line connecting it (in Riverdale) to Lewis and Clark College might be useful; but the most direct route up the hill is bus-unfriendly. 
A few other notes

I probably should add that the suggestions made above, are made as examples as much as anything--whether or not these would be productive lines (or be of sufficient importance to be funded in a limited-funding environment) is questionable.  TriMet has, I'm sure, considered many of these ideas.  Certainly, I wouldn't advocate cancelling any existing services in order to provide these.  But in the future, if gas were to become expensive and transit to become a first option for more resident, many of these routes might make sense. 


  1. How about one up and down Columbia between NE 33rd and Pier Park, and another from Columbia and NE 47th up Cornfoot and Alderwood, hitting all those jobs around PDX.

    I used to work for a Boeing contractor (the big twin hangars there) at Cornfoot and 47th (don't own a car, don't want one), and the closest I could get to work via transit was the 75 bus stop at the corner of Columbia and NE 47th, followed by a 1/2 mile-plus walk up 47th (no sidewalks) in the rain with FedEx triple trailers buzzing me by 18 inches and kicking up all kinds of shit on me along the way.

    And then there were the times construction was going on, and the 75 was detoured right onto Lombard from 42nd, so my walk was well over a mile. And I wasn't the only one. Out of about 25 people on each shift, roughly half of us rode the bus in to work. And this is just one facility amongst dozens and dozens...

  2. I'm skeptical of a micro-route given that similar micro-routes in the West Hills are very poor performers (51, 18, 63). I guarantee it would be one of those bus lines with 2 people onboard and there would certainly be no night or weekend service. It would only be a matter of time before it got the complete ax. It helped a lot in terms of ridership and support that Willamette Heights was getting frequent service as an extension of the major 15 bus because Willamette Heights could never support that level of service if it wasnt on part of a major bus line.

  3. The problem with the 51, 18, and 63 (and numerous other similar routes) isn't that they're short--it's that they are social service routes which cover territory which isn't transit friendly domain--generally middle-to-upper-class residential neighborhoods with few amenities. The exception to this is the 63, serving Washington Park--but it competes directly with the Washington Park Shuttle (which transports passengers through the park between PGE Park and the Zoo MAX station, but without going through nearby residential streets.

    And in each case, I'd use the smallest rolling stock on these routes that I can find.

    As you note though, many short routes might be better handled as extensions or branches to existing major routes--this gives 'em a bit of political cover. (A similar phenomenon occurs in some software companies--the worst thing you can do with quality assurance engineers is to call them quality assurance engineers; it's like painting a big target on their back for the bean-counters).

  4. VTA in Santa Clara County CA tried this. Google search will show the results.

    I think one of the biggest downsides was that since you now have a "different" bus in the fleet, there were new costs on the maintainance end - re-training mechanics, keeping another set of spare parts, etc...

  5. I'm assuming you are referring to VTA's "community busses"--I found quite a few references to them, but not so much to the results (praise, criticism, other analysis)--if you could point me to some, it would be useful. Given that the program started in 2008, I suspect any "results" are inconclusive due to the recession--a fact that I try to keep in mind as I gleefully bash WES.

    That said; it appears that the "community bus" program is intended to be a system of local circulators designed to serve short trips within a neighborhood (complete with a lower fare structure than the regular bus routes). One problem with this sort of trip is that it competes with walking. Another issue with VTA's implementation is that it's hourly, weekday-only service--in other words, not attractive service. (And a final issue is that while San Fran is a great transit community, Silicon Valley is not--the land use patterns found there are going to handicap ANY transit you build.)

    I suspect a better opportunity for micro-routes is in last mile service--permitting someone who is travelling on transit at a distance to get the last mile (or so) to their destination; without having to walk a long distance (often in a pedestrian-hostile environment) or having to bring a bike with them--or to connect with transit in the first place, it it isn't close by. The situation Jay illustrates. Here in Washington County, we've got numerous office parks and such, which are within jogging distance of the MAX line--but which have no (or infrequent) bus service getting you the last little bit; making the entire system not useful for the commutes of those who work there.

    But if you can point me (and other readers) to other resources on what VTA has been doing, and how well it's worked, it would be much appreciated!

  6. Isn't this exactly what TriMet tried back in 1998 with "The Local"?

    Use cutaway buses (i.e. LIFT buses) on very short routes, almost all of them functioned as MAX shuttles and carried an "S" suffix (i.e. 50s Cornell Oaks). There were also a couple of routes that ran out of the Oregon City TC.

    After about a year or so, the M9000 series buses were relegated back to paratransit duty and the 1900s (and presumably the 1600s out east) took over. The 2400s were purchased for these routes but had a nasty tendency to blow up (at least two or three of the vehicles had catastrophic engine fires) resulting in the premature retirement of that fleet. A couple 2400s remained for the Washington Park Shuttle (which now uses 1900s) and Blue Lake Shuttle (which no longer operates).

    The 46 (North Hillsboro) and 53 (Artic/Allen) are relics of "The Local", along with the 154-Willamette, and the various Happy Valley routes. I want to say the 87 was also a "The Local" route.

    The 67 short-line to Cornell Oaks (soon to be discontinued) is a direct result of the discontinuation of the 50s route.

  7. Pictures of the 2400s. Apparently vehicle 2402 was in service last year (nice hybrid paint scheme - old colors but new logos) but I haven't seen it once this year.

  8. Interesting. I didn't pay much attention to TriMet back in 1998, so I'm not aware of the program. I was also wondering about the 2400 series busses; it's unfortunate to hear that they were junk.

    How often did these routes run? And, to ask the same question as asked of the VTA "Community Bus" above, how were they marketed?

    I'm surprised that there isn't any through bus service on I-205 between Tualatin and Oregon City (stopping infrequently, perhaps at Meridian Park, Stafford, Willamette, West Linn, OC TC, Gladstone, Clackamas, and CTC). I seem to recall such a service existing before, but if it did, its long gone. This would be an excellent place to run a BRT-like service, and if offered at reasonable frequencies (i.e. more than once an hour on weekdays only), it could be competitive.

    Far too often, though, one sees the belief that because crappy service does poorly on a given corridor, good service won't do well.

  9. Do we know what the permanent solution is for this intersection? What kind of signalization or signage is being installed?

    If you look at this old 1943 transit map of Portland you see little (Y) and (L) symbols at the end of the bus lines... I have to believe these indicate how they turned around, L being loop and Y being 3 point turn. (Obviously at that time it was still the WH streetcar but look at other bus routes in the system.)

    1943 Portland Traction Company map


Keep it clean, please