Further Musings on the Transit Mall


So I’ve been musing about the Transit Mall over the weekend, reading all the comments on the prior posts, and have a few questions turning over in my mind, which I’d like to throw out to readers.

But first, I want to thank Jim Karlock. While I generally don’t agree with Jim’s take on transportation, I admire his dedication, including all the meetings he tapes. He has just posted his video of the Downtown Neighborhood Association meeting last week. Here they are, he tells me you need Real Player:

Part 1 – Presentations (43 min)
Part 2 – Q&A (58 min)

Question #1 – Interpreting the APTA Report

In my last post, I suggested that the APTA report was simply a constructive criticism of the design, not a condemnation. A reader has pointed out to me that the scope of the report was limited (i.e., they weren’t asked to review design alternatives):

We are requesting a report that documents your opinions of our planned
operating procedures and any recommendations for improving the movement
between buses and LRVs. We are also interested in any “best practices” that
you can share with us.

The reader has opined that in light of this, the following paragraph from the report is sort of screaming “don’t do this”:

While other city experiences in downtown street operation would suggest that normal
rubber-tired traffic not be mingled with rail vehicles any more than necessary, and
constantly changing lanes by transit vehicles (particularly LRVs) is not
recommended, the following comments are provided in an attempt to maintain as safe
and smooth a mall operation as possible with the policies already adopted for the
Portland Transit Mall.

So is this a red flag, or friendly recommendations to make it work better?

Question #2 – Bottlenecks

I’m trying to understand if the potential system bottlenecks are a function of the Green Line or the Mall alignment.

Currently the Steel Bridge itself is a bottleneck, and the Y junction on the east side of the bridge is a potential bottleneck. Does the addition of another Y junction on the west side further limit capacity, or is capacity already constrained by the bridge or the existing eastern Y?

This matters because if the new Y is the issue, that might argue toward continuing to use Yamhill/Morrison rather than the mall. If the new Y is not the issue, then the question is whether we can get the Green Line across the bridge to any west side destination.

Who’s got answers or opinions?


19 responses to “Further Musings on the Transit Mall”

  1. I’ve never understood why regular vehicle traffic is allowed on the Steel Bridge along with MAX in the same lane. It would be easy to adjust this with one new traffic light on the western aproach to make vehicles only use the outside lane. With the addition of the Green Line and the new Union Station spur this should be the first issue addressed.

    As regards to a bottleneck, MAX trains over the Steel Bridge will always have to travel slow unless the Bridge’s tracks and jointery are redesigned. I see that the addition of the new Union Station spur will help speed up trains progressing over the bridge for the simple fact that the old town stations create a bottleneck already for the Blue, Red and Yellow Lines let alone adding the Green Line.

    I’d say limit regular traffic to one lane each way first, try to upgrade the tracks second for faster train travel time, and in the future perhaps dedicate the Steel Bridge as a transit/ bike/ walking bridge only by adding a 3rd and 4th track to the upper deck.

  2. I was under the impression that cars DO NOT share the MAX lane on the bridge. Buses do, but I didn’t think passengers cars were allowed to cross over?

    Also, the plan is to have the Yellow line use the mall alignment once it is available.

  3. Try driving from downtown over the Steel Bridge and you will find that the both eastern travel lanes are open to regular vehicle traffic (going west only one is open to autos), which is a notion beyond me. Yes, the Yellow line and the new Green line will no longer travel to the congested oldtown stations. I am for this mall alignment, and would love to see the potential southern lines (to Milwakee-Oregon City and down Barbur Blvd. to Tigard and further)come about someday. The region needs a rail transit plan, and a permanant means of financing it over time.

  4. On the Steel Bridge, private vehicles are excluded from the MAX track in the eastbound direction, but not the westbound. Trains run more slowly than buses or cars due to the tracks.
    There must be a video or some kind of simulation for the operation of the Ys at each end of the Steel Bridge; being able to feed the Bridge from two different lines would seem to be an advantage, not the other way around.
    No question the weave is not the best from an operational viewpoint, but we have had MAX running on city streets for 20 years without much trouble, so in order to please the downtown merchants (and to get a piece of the needed funds) you take a chance.
    Morrison/Yamhill seems to work…though the trains are slooow. Buses are the wild card; I expect there will be lots of training and high senority operators will avoid the Mall more than ever.

  5. Chris,
    The wye at the west end of the Steel Bridge can seriously disrupt the Blue and Red line trains during the peak hours and routing the Green and Yellow Lines onto the Transit Mall will not increase the capacity of the light rail system.

    Think of the current light rail alignment as a through road and the rail branches off that road as signalized intersections. When the Yellow Line was connected to the through road at the Rose Quarter, the signal caused some disruption of through traffic. When the second signalized intersection is added at the west end of the Steel Bridge with more traffic than the first one, the through traffic flow will be further compromised.

    It is these intersections, not the ability of Morrison and Yamhill to handle the increased traffic or the station stops that will limit the systems capacity.

    By the way, the southbound and northbound Mall trains will have to cross each other’s path at Union Station and then again at the south end when the Milwaukie extension is built because the Mall runs counter-flow like roads in England.

  6. I mis-spoke…private vehicles are excluded from MAX tracks in the WESTbound direction out of the Rose Quarter; both lanes are open EASTbound.

  7. Jim,

    Using your signalized intersection analogy, if the intersections on both sides of the bridge are balanced, then the overall capacity of the corridor is not reduced or increased if one of them is removed (because the other is still the capacity limit).

    So could you say a little more about why the western junction removes capacity that the the eastern junction has not already constrained?

  8. Chris –

    There are two competing “bottleneck” issues at play here, and is difficult to know which is most important without running simulations and seeing actual numbers.

    But, to answer your question, “So could you say a little more about why the western junction removes capacity that the the eastern junction has not already constrained?”, there are indeed new situations created by the new Y that will arise.

    Consider a situation that can arise today:

    1. A westbound Blue-line train is about to leave Rose Quarter.
    2. A downtown-bound Yellow-line train leaves Rose Quarter a few seconds sooner, so the Blue train must wait.

    An _eastbound_ Blue line train coming over the bridge is unaffected by this problem, because it can get through the Y without being interfered with by westbound traffic.

    (An _northbound_ Yellow line train coming over the bridge will have to wait for both westbound trains to clear before it can turn north through the existing Y.)

    Now, add the Y and the Green line and the exact same starting conditions as above, but NOW, a our _eastbound_ Blue-line risks interference from eastbound Green-line trains that want to use the new Y to move into the eastbound track. So, not only do you have the conflict posed by conditions #1 and #2, you have a new conflict condition on the other side of the bridge.

    Now, extend this out into kind of a worst-case scenario:

    1. A westbound Blue-line train is about to leave Rose Quarter.
    2. A downtown-bound Yellow-line train leaves Rose Quarter a few seconds sooner, so the Blue train must wait.
    3. An northbound Yellow line train is on the bridge. It cannot move until the two opposing trains have cleared.
    4. An eastbound Red line train is waiting at 1st and Davis for the bridge to clear. So, it is waiting for that Yellow line train to get off the bridge.
    5. An eastbound Green line train is also waiting to get on the bridge at the new Y. But now, because the two westbound trains (from conditions #1 and #2) are finally coming across the bridge, it must wait for them to clearly the new Y.
    6. During this time, the yellow that was waiting on the bridge now gets to move into the Rose Quarter station, and the Red that was waiting at Davis now moves onto the bridge.
    7. Now that the red is on the bridge, and the two westbound trains have cleared the Y, the eastbound Green is still waiting because now the Red occupies the bridge.

    In a perfect world, schedules could be adjusted so that the worst case never met up at the bridge at the same time. But of course this is the real world, and a simple delay of a minute or two (passenger holds the door, jaywalker stops the train and then the train misses the light, emergency vehicle stopped on the tracks, etc..) can lead to cascading conflicts and schedule disruptions.

    Now, that being said, the conflicts that the Y’s cause us are mitigated by the fact that once you get through the Y, there is somewhere else for trains to be. So, if you have a train that is taking a long time to exit 1st and Davis heading west, as long as the other trains coming downtown over the bridge are yellow or green, they will have somewhere to go and won’t get backed up.

    But, how do we guarantee that there won’t be a Blue stuck on the bridge waiting for 1st and Davis to clear up? Because the signal system will be coordinated to not dispatch a train over the bridge until there is room for it on the other side. This will help prevent true gridlock disasters, but it means trains will have to wait longer before entering the bridge. That is why my above examples only show one train on the bridge in a given direction at any time.

    Consider also the peak-hour headways… I believe I have these right based on currently-published information:

    1. Blue Line – 10 Minutes (6 per hour)
    2. Red Line – 15 Minutes (4 per hour)
    3. Yellow Line – 10 Minutes (6 per hour)
    4. Green Line – 15 Minutes (4 per hour)

    That’s 20 trains per hour per direction needing to cross the bridge. (40 per hour both directions). A train will be entering the bridge, on average, every 90 seconds. A cascading failure of just a few iterations can mess up the entire schedule.

    Two iterations of failure means that a train cannot leave Convention Center because there will be one in the way at Rose Quarter. Four iterations of failure will have another train backed up behind that at 7th Ave, and so on.

    And here is where things get really bad: There isn’t much leeway in the system to work our way back to normal once a backup occurs. We can’t move trains across the bridge any faster to make up for lost time, so the best we can hope for is that delays do not continue to cascade once they begin.

    Hwy 217 used be mostly traffic lights at intersections, and even the transition from I-5 to 217 used to involve a traffic light or two. Over the years these bottlenecks were removed by going three-dimensional and creating grade-separated interchanges (overpasses, off-ramps, etc.) Yet we continue to build light rail with compromise alignments, compromised river crossings, and compromised interchanges. Now we are paying the price for not accommodating future capacity in past interchange designs.

    We did this with the Red Line, too, by creating a single-track U-turn into Gateway at the absolute minimum radius. This made construction cheaper and required less interference with I-205, but as a result Red trains absolutely crawl into and out of Gateway. To redo this at a future date would require shutting down the red line.

    I think this is where rail operations nuts like me get the most heated up… It is very difficult to redo things once a system is up and running. Bottlenecks should be avoided wherever possible so that future expenses are reduced. We must make better accommodation for future growth in the design of our extensions.

    Finally, an aside on the topic of eastbound autos sharing a lane with LRT on the bridge… I have never seen a problem with this because the trains move so slowly over the bridge that they do not get stuck behind cars.

    – Bob R.

  9. Pretty safe to assume Kari is not a right wing anti-transit gadfly but he asked,

    “”””””” February 10, 2006 12:51 PM
    “”””””””Kari Chisholm Says:

    “”””I’m still having trouble figuring out what’s broken about the current bus mall. Why are fixing something that ain’t broke?”””””

    I didn’t see any answer so I’ll take a shot at it.

    More Light Rail at all costs.

    Just like high density.

    The fact is the Transit Mall doesn’t currently have light rail and the Mall is a way to advance light rail, period. It also sets up additional LRT expansions which will also be perpetrated without voter approval.

    The illegitimacy of our LRT is growing as comments on this thread reveal.

    The impending LRT gridlock is chaos in the making.

  10. Steve says: “More Light Rail at all costs.”

    Now, what we have is some more light rail at moderate costs. If the situation really were “light rail at all costs”, we wouldn’t be having this conversation, because we would have a new river crossing, grade-separated rail interchanges, and no auto lane on the transit mall.

    Nearly all of the compromises in this proposal are to accommodate cost considerations or to make things work in the context of adding a through auto lane.

    I’m perfectly comfortable with debating the cost and utility of light rail for a given application, but extremist views such that rail is being pushed on us “at all costs” are not worth debating.

    – Bob R.

    PS… You still have not provided documentation to back up your earlier claim that downtown employment has gone down. I’ve asked you three times now.

  11. re current operations at the eastside wye…it is my observation that Yellow Line trains cross through at approximately the same time. As the City Center train pulls out of Rose Quarter/Interstate, the Expo Center train is coming off the bridge.
    re Lightrail…the residents of the City of Portland, where most of the coming projects are, have never voted down lightrail. South/North, which included a $400M property tax, passed in Portland by a 60-40 margin.

  12. Lenny wrote: “it is my observation that Yellow Line trains cross through at approximately the same time.”

    Yes, under ideal conditions, two trains can make use of the crossing at one time. But this requires (in your example) that a northbound Yellow Line train happens to be coming over the bridge at exactly the same time as a downtown-bound Yellow line train has closed its doors at Rose Quarter and is ready to proceed.

    There are multiple combinations of trains that can make it across the bridge simultaneously. I’m trying to point out that there are also multiple combinations that can lead to small delays, and that adding another Y adds to those possible delay combinations, and that these can cascade into further delays.

    Conflicts which lead to delay already occur today. They are not a major problem because there is room in the schedule for trains to make up time (at least outbound trains). But we will be adding, at minimum, 4 more trains per hour across the bridge when the green line opens. Peak hour headways will no longer have room to make up for delays.

    – Bob R.

  13. Bob said

    “””””””””I’m perfectly comfortable with debating the cost and utility of light rail for a given application, but extremist views such that rail is being pushed on us “at all costs” are not worth debating.”””””””””

    It’s not an extremist view it’s reality. From every angle.

    It isn’t valid to suggest that because we don’t already have more of it is evidence to the contrary.

    The push for light rail has cost billions, has cost other planning, has cost needed road expansions and improvements, has cost endless development dollars at Cascade Station, Gresham Station the Beaverton Round , etc., without the promised public benefit and has spawned a culture of Urban Renewal TIF borrowing and devastated basic services budgets.

    It continues with SoWa and the Tram and at Clackamas Town center where $23 million in UR borrowing will be used for the Shopping Mall expansion.

    I understand your advocacy for more light rail but you do so without tracking any of it’s detriment.
    Thereby never fully weighing it’s net worth.

    “”””””””””””PS… You still have not provided documentation to back up your earlier claim that downtown employment has gone down. I’ve asked you three times now.””””””

    As you know full well I handed that off to others who posted in repsonse to you.
    If you didn’t like what they said fine.

    Lenny “”””””where most of the coming projects are, have never voted down lightrail. South/North, which included a $400M property tax, passed in Portland by a 60-40 margin.””””””””

    That’s real slick. Take a sub vote, turn it into approval and then make the whole region pay for something they did not approve.
    Nice mental gymnasitics but thoroughly disingenuous.
    I can’t imagine why there was not another vote to see if the 60% of the Portland vote wanted to also pay for it.
    The only reason there continues to be an avoidance of public votes is the results would be non approval.
    Instead we have the plundering of basis services budgets like that which funded the Interstae MAX and Airport MAX both.
    It’s too bad light rail advocates seem not to have any compassion for those basic services.
    But as I said, It’s more “rail at all costs”.

  14. By any broad and comprehensive look
    Light rail is not succeeding.
    According to the Metro report on regional transportation alternatives published around 1997 their LRT expansion plan, operating plus capital costs, weighed in at $6.5 billion. This did not then even include the I-205 South segment (which prior published Metro studies showed to have ridiculously low ridership potential). If we updated their $6.5 billion estimate to include all the extensions they now have in mind — most of the “dozen or so” advocated for over a decade by the Willamette Week publisher — the figure would be approaching $10 billion. This wacko ’60s dream is becoming reality — at huge long term public expense.

    That same report showed that the $6.5 billion for LRT would be 1/40th as cost-effective in reducing congestion as about $9 billion in “Preferred” highway improvements. (Oh, that $6.5 billion would raise the transit share of all trips from 3.9% to 4.2% according to Metro except that the share of “pure” transit trips in which one wouldn’t need an auto to access transit would actually decrease as walk-to-bus trips would be replaced by drive-to-MAX trips such that auto trips would actually increase in market share in one includes trips to MAX). I wrote a whole report just on that Metro report showing that their own forecasts proved the $$ for light rail was a sheer waste of money from every perspective.

  15. Steve writes: “I understand your advocacy for more light rail but you do so without tracking any of it’s detriment. Thereby never fully weighing it’s net worth.”

    I do so without racking any of it’s detriment, huh? To state as such is to disregard most of the posts I have made on this forum, and clearly you haven’t been any of the public meetings where I’ve delivered comments. I critical of a number of aspects of light rail, including where I think it should and should not be expanded, and I offer up constructive criticism to appropriate agencies at nearly any chance I can get. And things don’t always go my way, and sometimes I am frustrated by aspects of the process, but I don’t run around making absolutist blanket statements like “light rail at all costs” or “without tracking any of its detriment”.

    Steve says: “As you know full well I handed that off to others who posted in repsonse [sic] to you. If you didn’t like what they said fine.”

    No, I don’t know that “full well”. First, you never stated that you had handed that off to others. I’m glad you have others who you can give the responsibility of responding when you are directly asked to provide facts for a statement you made. And no, a “didn’t like” what “they” said because NO ONE has posted any evidence which shows a decline in downtown employment, which was your original claim.

    If you don’t like the fact that I am upset that you make wild claims and don’t present any facts, well, “that’s fine.”

    – Bob R.

  16. Bob, I appreciate the comprehensive response on the Y junctions. One more detail:

    On the eastside we’ll have a Y with 3 lines on one side and one line on the other, on the westide it’s 2 and 2. Is the 2 and 2 ‘more balanced’ and therefor less of a constraint?

  17. Chris wrote: “On the eastside we’ll have a Y with 3 lines on one side and one line on the other, on the westide it’s 2 and 2. Is the 2 and 2 ‘more balanced’ and therefor less of a constraint?”

    I think it’s the opposite… I may have to break down and buy some simulator software to continue this discussion! :-)

    Think of it this way… Every time a junction is used (much like a traffic light), traffic in at least one direction is stopped. Thus, in terms of throughput, the less often you have to use the junction, the better.

    Imagine if you had a junction much like Rose Quarter, but with 19 trains coming from I-84 and only one train coming from Interstate avenue. You’re hardly ever going to need to switch over to Interstate trains and the I-84 trains can largely be unconcerned with delays. Much like a traffic light with sensors that stays green for a primary arterial most of the time, and only switches over when it senses that someone is waiting on the side street.

    Now imagine those same 20 trains being split up 50/50, 10 on Interstate and 10 on I-84. The odds of one being interrupted by use of the junction have just gone way up. 50/50 may in fact be the worst risk of interference. Just like a traffic light at an intersection of two arterials, at peak hours no matter when the light changes, somebody is going to have to wait.

    Also, track switches take a finite amount of time to activate, and the trackway must be clear of trains before the switch can operate. Thus, every time the switch changes state, a little bit of time is used up where no trains can occupy the junction. A 50/50 share of traffic maximizes the odds that the switch will have to change state.

    I can write a program that generates a table of switch conditions for any possible combination of queued up trains, but it would be a really big table and might not be correct… If you have friends in operations planning at TriMet, this would be a good time to ask them for the results of internal simulations. :-) (Not just a statement of “we are confident that this will work”, but the actual numbers…)

    – Bob R.

  18. PS…

    I don’t know if the bridge can support it structurally, but you can slightly reduce number of switching operations involved by adding a third track across the bridge. Not by a lot, but it would help.

    – Bob R.

Leave a Reply to Steve Schopp Cancel reply

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