02 September 2023

Level Boarding: Still Not Getting It

The good news: Caltrain has initiated a small study effort to develop a level boarding roadmap, as part of its portfolio of capital projects.

The bad news: in the summary of this study, Caltrain shows no sign of grasping the purpose of level boarding. We might need to display it on a freeway billboard, like this:

Caltrain does a very nice job of explaining the benefits of electrification. Faster acceleration leads to shorter trip times, a strong message that they hammer often. On level boarding, however, the messaging is muddled. It's something-something about steps? Easier and more inclusive access?

NO!!!

It's about shorter trip times, just like electrification. While electrification saves time in motion, level boarding saves time at rest. The savings are big: cutting station dwell time from 45 seconds (typical for Caltrain today) to 30 seconds (typical for BART, which has level boarding) is worth almost as much "acceleration" as electrification. Electrifying without level boarding is half-baked, and Caltrain should be spending a lot more on level boarding than distractions like BEMUs. Please return to your core mission to quickly and efficiently get people where they are going.

About these numbers: the diesel trip time is for an all-stops local with 45 second station dwells and 15% padding. The EMU trip time has the same dwell and padding assumptions. The level boarding time assumes that station dwells drop to 30 seconds, and padding is cut down to 10%. The lower padding is appropriate because level boarding not only makes dwells shorter, but it makes dwells much more consistent and predictable, as discussed previously. If two wheelchair users need to board, it takes the same 30 seconds, not five minutes of staff assistance. Here are the detailed stop-by-stop stats and string diagrams if you want to tinker with assumptions.

41 comments:

  1. I notice that Caltrain are proposing running at 79mph when HSR want the ROW to give them the ability to run at 110mph. Would running at 110mph give Caltrain a shorter run time? Won't both running at different speeds cause scheduling problems/

    ReplyDelete
    Replies
    1. Yes, speed differentials would cause major capacity problems. In a 110 mph two-track blended corridor that spans 50 miles, there would have to be severe compromises for both operators, which is why you don't see this sort of operation practiced anywhere else in the world. Caltrain couldn't run any local trains as simulated here (see business plan technical materials) and would have to skip a lot of stops to pick up the average speed.

      If you skipped a lot of stops (similar to today's Baby Bullet), had a 110 mph corridor + level boarding, and really pushed the throttle in a Caltrain EMU, then you could do SF-SJ in 50 minutes flat. HSR would have to slow down to match that.

      But why do this to Caltrain and give up on frequent, high-quality regional rail service? Fundamentally, HSR's blended plan is operationally incompatible with dense & frequent Caltrain service patterns. I think it's exceedingly unlikely to ever happen as currently planned and environmentally "cleared."

      Delete
    2. I remember riding the Eurostar before the English built their high speed link to the Chunnel. I was on a train back from Brussels that had been slightly delayed in Europe that caused it to miss its slot on the network into London. It was start and stop all the way into London. I fear that someday that will be the ride from San Jose into SF on high speed.

      Delete
    3. We stayed a few years ago near Ivrea, Italy, which was served by 4-6X a day electric loco-hauled trains Aosta-Turin, and hourly (I forget if diesel or electric?) FLIRTs Ivrea-Turin. On our way to Milan or more distant adventures, we'd burble down from Ivrea to Chivasso at moderate speeds on a leafy single-track line. A 4-10 minute pause at this small town's 6-platform junction station to switch directions, wait for our slot, then we'd take off for Milan at full throttle on the old Milan-Turin main line, keeping pace with the Freccarossia high speed trains after the HSR line merged in. That last leg was an impressive boogie (I don't recall any intermediate stops).

      Delete
    4. Under the Swiss Triad of "Clockface Schedule", "Schedule-Rolling Stock-Infrastructure", and "Run as fast as necessary not as fast as possible" the goals should logically be a one hour SF-SJ express (56 min technical time) and a 90 minute SF-SJ local (84 min technical time). Depending on the frequency of each (2 or 4 tph) you would end up with knots where services meet in both directions (for transfers to other modes) or overtake (for transfers to/from express) at Millbrae/SFO, Redwood City, and Mountain View. Together with 4th&King and SJ this is conveniently five of the seven busiest stations. Such as plan would both optimize rolling stock usage (the 9 am departure from SF arrives in SJ at 9:56 and becomes the 10 am departure northbound) and passenger ease-of-use (trains would depart SF, Millbrae, RWC, Mountain View, and SJ, both north and southbound where applicable, at :00,:15,:30, and :45 after the hour for a 4 tph express, whenever trains run; at Palo Alto you might see northbound express always leaving at :10,:25,:40,:55, southbound at :05,:20,:35,:50).

      HSR in this plan, if scheduled to leave just before an express, would, for a SJ-SF time of 36 min, only need to pass a single express (at 4tph) and a single local (at 2tph), both nominally at RWC where the express would be "overtaking" the local at a timed cross platform transfer. Accommodating HSR by arranging for a single high speed bypass ("schedule-rolling stock-INFRASTRUCTURE") shouldn't be a major undertaking and shouldn't cause a 'major capacity problem'.

      My figures and rough and illustrative, different plans/assumptions (more than 6tph Caltrain or 2 tph HSR, etc.) would of course eventually cause conflicts as track space fills up.

      Delete
    5. Hi Onux, have a trawl through Clem's past articles tagged "timetables". (This blog's article tagging is incomplete, but many of the greatest hits are there.)

      We've been pointlessly and depressingly banging on about this obvious "Schedule-Rolling Stock-Infrastructure" stuff for 20 years.

      Delete
    6. @Anonymous,

      I’ve read every post on this blog for the last 10 years (I think, maybe more) hence my (hopefully mild) criticism of both Clem’s original post and his comment that you can’t have HSR on the same corridor. know Clem is well aware of S-RS-T yet his initial post seems to fall into the “fast as possible” mindset. No one should ever want to run any rail service with a 78 min trip time, it doesn’t fit into any sort of takt/pulse/knot system. This is independent of the question of blending HSR. Same for a 50 min express.

      Similarly, it is not reasonable to suggest that 6-8tph prevents multiple speed classes, certainly not if you plan the infrastructure for regular timed overtakes based on a takt. The only way this would seem to be a problem is if Caltrain is running ONLY all-stop locals, but I can’t see how that is the only viable service schedule.

      I can channel my inner Richard Mlynarik and point out that just because HSR can be fit into the Caltrain takt from SJ-SF without impact doesn’t mean it will, the “finest professionals” on both sides could choose instead a plan that hurts Caltrain or HSR or both.

      I will grant that blended ops does impose a capacity contraint, at some point frequent enough trains require continuous overtake which would mean 4-tracking the corridor.

      None of this takes from the fact that Clem is completely right about level boarding being the highest priority after electrification, and maybe should have been higher before it. However “faster, faster” isn’t a solid argument for it if one is following the principle of “fast as necessary.”

      Delete
    7. Hi, I don't exactly see how to make sense of much of what you wrote.

      The good news is that you can show your work using the world-famous barnacle-encrusted Taktulator, instead of typing words whose meanings people have to try to guess.

      Delete
    8. I don't see how this can work. The HSR is going to end up in Oakland or just end at San Jose. "Blended" operation without three tracks throughout and lots of high-speed cross-overs to the local tracks will be a joke.

      Delete
  2. While Caltrain’s new trains are tested at up to 121 mph and capable of 110 mph in revenue service, raising top cruising speeds from 79 mph to 110 mph will not shorten run times for most trains (which make all or most stops — some as close as a mile apart) anywhere near as much as reliably shortened station dwell times afforded by level boarding.

    That said, Caltrain may someday choose to raise its maximum authorized speed (MAS) in some areas — maybe even before HSR gets here — so that its trains with widely-separated station stops (eg: express and Gilroy trains) may benefit.

    ReplyDelete
  3. The most ridiculous aspect of this plan is the planning/installation of mini-high platforms that will be going on through at least 2025. Why get paid once to do platforms correctly, if instead you can get paid to do them over, and over, and over again....

    ReplyDelete
  4. There is certainly seems to be no urgency at Caltrain to work on this Level Boarding study. The latest capital program update says that there has been "minimal progress due staff availability" although they did introduce the project to their intern. It does sound like staff views this as a vague future program rather than something that could be prioritized and implemented in the shorter term.

    I noticed that the $31M estimated cost of retrofitting the EMU fleet for level boarding has been added to the DTX budget based on FTA requirements. I hope this doesn't mean that Caltrain intends to wait until DTX opens before implementing level boarding anywhere.

    ReplyDelete
    Replies
    1. Dear intern, if you read this blog, please get in touch (contact info in "About This Blog") and I will buy you as many coffees as you want.

      Delete
    2. I do wonder if the automatic step at the lower doors can be used as a gap filler. Based on images I've seen of the mechanism it looks like it can lock into place before it drops down to form a step. That would perhaps negate the cost of retrofits (which are probably related to compatibility with the upper doors). While the upper door would certainly be better from an HSR compatibility standpoint I get the feeling Caltrain will end up making the platforms level with the lower doors.

      Delete
    3. Damn that post sounds like it was written right before a pandemic, which it was...

      Reece Martin pointed out that Metrolinx (Toronto) has engineering standards in place to future-proof the construction of new station platforms, which if built at 8 inch height have design details to make future platform raising as simple and cheap as possible. It's not much, but it runs rings around Caltrain. They're going to rebuild Broadway Burlingame at 8 inches with ZERO future proofing, I guaran-f'in-tee.

      Delete
    4. I recall during a presentation the City of Burlingame gave that someone on Caltrain staff said the station would be designed to accommodate level boarding in the future for what that's worth.

      Delete
    5. "I recall during a presentation the City of Burlingame gave that someone on Caltrain staff said ..."

      Q: How can you tell when somebody from Caltrain is lying?

      A: Their lips are moving.

      Delete
    6. Why would they lie about that? I can't fathom any reason why they'd lie about something that, frankly, the audience at that city council meeting could not care less about. Could be this person was wrong, but that's just being misinformed, not something with malicious intent.

      Delete
    7. I agree, never ascribe to malice that which is adequately explained by incompetence.

      Delete
  5. InfrastructureWeak06 September, 2023 15:17

    Meanwhile, under the 22nd St ADA project section, Caltrain reports that they "Will continue to investigate procedure for requesting a waiver" of level boarding requirements triggered by platform modification there, rather than providing level boarding as part of that as-of-yet-unfunded project.

    ReplyDelete
    Replies
    1. Hey at least they got beyond the "introduce the project to the intern" step of the process.

      Delete
  6. BART’s scheduling guru John FitzGibbon seems to get it. After listening to this interview with him about how & why he developed BART’s new “clock-face” schedule, I’m hopeful Caltrain could somehow hire-borrow him (or ape his scheduling philosophy and sensibilities) to design their post electrification schedule(s).

    “On September 11, BART will roll out a reimagined service plan that will eliminate 30-minute wait times on nights and weekends. That means no BART rider will have to wait more than 20 minutes for a scheduled train no matter the hour or day of the week. This new schedule is responsive to post-pandemic commute patterns and opens new opportunities for BART to expand its ridership. To do a deep dive into the new schedule Jay Sathe sat down with BART’s Manager of Scheduling and Planning John FitzGibbon (below) on the latest edition of our podcast series, 'Hidden Tracks: Stories from BART':”

    Podcast & transcript: BART’s new, reimagined schedule
    https://www.bart.gov/news/articles/2023/news20230905

    ReplyDelete
    Replies
    1. 100% agree. John FitzGibbon also has some really cool software to help him with the constraints. I highly recommend reading his earlier article where he describes that newly purchased software.

      Rather than hiring John, it sounds like Caltrain's RoW and Schedule should be added into his software package to ensure Millbrae transfers are covered from the start.

      Here's his earlier interview: https://www.bart.gov/news/articles/2022/news20220908-0

      Delete
    2. The BART software in question appears to be a tool named HASTUS. During the Caltrain business plan effort, that team was using sma's Viriato which is also used to plan the Swiss railway network.

      Delete
    3. Somehow nobody seems to have noticed that BARTs new schedule generally screws over Caltrain passengers who transfer at Millbrae. Now ALL Millbrae trains will stop at SFO on the way to/from SF, adding 7mins to the trip.

      It also reduces Millbrae service from 4tph to 3tph during daytime which makes matching the 4tph/2tph caltrain schedule more difficult. Caltrain are adjusting their schedule on Sept 25 (2 weeks after BARTs change!) to make it a bit better but it will still suck. See BART-Caltrain Transfer times for details. Check out all those 1m/2m transfer times for SB connections! Get ready for some cardio running up/down those stairs! (or get there 20 mins early)

      With the combination of these 2 changes it will now often be faster for daytime SF-bound Caltrain passengers to get off at San Bruno and walk the ~10min to the San Bruno BART station which sees increased 9tph service under the new plan. Of course many Caltrains do not stop at San Bruno....

      Delete
    4. People do seem to have noticed that SFO has 180% of the ridership of the insanely stupidly configured Millbrae station. Unfortunately for us and unfortunately for the planet and unfortunately for BART operations and service planning (who are and for decades have been really good at dealing with the shitty fucked-up insanity thrown at them by the corrupt cretins behind BART extension and major capital spending) Millbrae isn't on the way to anything, so it goes to the back of the queue.

      Millbrae, unlike Colma, San Bruno and South San Francisco BART stations, isn't a total</i ridership basket case, but it's absolutely nothing worth going out of the way for, and of course should never have been built. It also needs to be blown up entirely, because it's good for nothing and nobody, fucks over riders, and fucks up Caltrain.

      Also, anybody who thinks Caltrain can offer reliable convenient scheduled transfers to or from anything is somebody who's a sucker and got sucked in by some misleading map or theoretical timetable or something.

      Meanwhile, Steve "$4 billion Bay Bridge cost overrun" Heminger, former Quentin Kopp go-to boy, MTC "politican liason" (if you think this sounds Soviet, you're not alone) fixer in the mid-1990s when BART to Millbrae was being rammed through at maximal expense for maximal contractor profit, and later MTC Executive DIrector For Life, now sits on the Caltrain and SFMTA boards. Truly a gift that has kept on giving and giving for decades and decades.

      Delete
  7. Lima (Peru) wants to buy Caltrain’s old fleet for a new 25-mile Lima-Chosica service on their existing Ferroccaril Central Andino (FCCA) RR corridor expected to carry 500k daily riders!

    Details on PDF page 4: https://www.caltrain.com/media/31492/download

    ReplyDelete
    Replies
    1. Someone should alert Chicago's Metra. Let's start a bidding war.

      Delete
  8. BART Tweets that its last-ever revenue service run of its legacy fleet departs Millbrae for Richmond tomorrow (Sunday, Sep 10) night at 7:39 pm: twitter.com/SFBART/status/1700689276823093551

    ReplyDelete
  9. It's not "only" about shorter trip times, it's about
    * Predictable trip times; and
    * Required for One Person Operation, making more service possible (shorter trip times, which at the margins allow fewer trains and fewer crews to provide more revenue service per hour, plays into this as well, more marginally but still importantly.)

    Clem of course (of course!) wrote a nice article about the necessity of predictable dwell times way back in 2016. Nothing's changed in the Caltrain bunker since then, of course, just as nothing has changed in 20 years.

    Predictable trip times make
    * Different and better service patterns possible,
    most particularly the only service plan that anybody should consider (though it's increasingly hard to throw in the towel and think about uniform all-stops all-day all-hours 20 minute headways.)

    Any sort of timed transfer or overtake on a rinky-dink amateur unreliable unpredictable operation like Caltrain can only be achieved at the expense of massive schedule padding, further lengthening trip times, requiring trains and crews to sit doing nothing while the clock ticks, wasting riders' time, and discouraging ridership. Caltrain has no problem with this, of course. Of course. Idle trains and idle crews mean more earmarked capital costs and more subsidies, and earmarks and subsidies are what they live for.

    ReplyDelete
    Replies
    1. At least any overtakes are by faster Caltrain runs, with fewer stops. Why would anyone expect, actually expect, high-speed trains ever to appear on the Peninsula or between San Jose and Gilroy? Why? Not even in a demo tour like Amtrak in 1993 is the safe way to bet, not even as a bid especially in the Bay Area and between Palmdale and LA for new funds or bonds. The project is going so poorly that no completion can be counted on, and they can't or won't even tow high-speed rolling stock at slow speeds before then to spark interest at visits, as Amtrak did long ago. Notice that the high-speed rail folks have never done a demo tour with towed Siemens Velaro or other trains or whatever else, in the state, anywhere, along the route segments, stopping for weekend visits at stations, and I bet if anything those in charge are increasingly less likely to think of doing one. Caltrain blended? Not even for a demo (with Caltrain pulling the high-speed train, even) the way things are happening now.

      Delete
    2. I'm not so sure anyone between L.A. and Anaheim these days may even be interested, and it's misleading to take the trains all the way to San Diego on that route, even though improving that beats the new high-speed route intended as a later part of the actual project.

      Delete
  10. Level boarding is not only about wheelchairs, parents with prams, travellers with heavy luggage, all these benefit from level boarding…

    ReplyDelete
  11. Caltrain unveils its proposed electrified service plan for Fall of 2024: caltrain.com/media/31624/download

    Notably, it proposes to run only EMUs between SF and SJ, with cross-platform transfers to/from diesel locomotive hauled Gilroy shuttle trains at San José’s Diridon station.

    ReplyDelete
    Replies
    1. I didn't spend any time trying to massage the departure times and train spacing, but I threw the schedule into taktulator [1] and got a score of 116 with a peak fleet utilization of 19 trainsets. That sounds somewhat borderline but it does technically fit into their train inventory.

      Otherwise this looks reasonable. Based on the fixed transfer times between Caltrain and BART throughout the day the schedule seems to be more evenly pulsed which is a nice improvement. The possibility of 100% electric power is also nice to see. The time savings of that magnitude on the local are somewhat unexpected so we'll have to see how that plays out in the real world.

      [1] http://mly.users.sonic.net/Caltrain-Timetabling/201105-takt/takt.php?width=1000&height=1200&period=60&start=7&direction=&title=Caltrain+2024+Schedule+%28Draft%29&t1t=KISS-79MPH-4MW&t1=.900.45..45.45.45.45..45.45.45.45.45.45.45..45.45.45.45.45.45.45.45..900.&t1d=&t1h=60&t1l=Local+A&t1s=-23&t1n=-11&t1c=0000FF&t1p=7&t1f=L&t2t=KISS-79MPH-4MW&t2=.900.45..45.45.45.45..45.45.45.45.45.45.45..45.45.45.45.45.45.45.45..45.900&t2d=&t2h=60&t2l=Local+B&t2s=-6&t2n=0&t2c=0000FF&t2p=7&t2f=L&t3t=KISS-79MPH-4MW&t3=.900.45...45..45...45..45...45...45...45.45....900.&t3d=&t3h=60&t3l=Express+A&t3s=3&t3n=-2.5&t3c=FF0808&t3p=7&t3f=L&t4t=KISS-79MPH-4MW&t4=.900.45...45..45...45..45...45..45.45.45.45.45.45.45.45..45.900&t4d=&t4h=30&t4l=Express+B&t4s=10&t4n=5&t4c=CC3D27&t4p=10&t4f=L&t5t=NONE&t5=900...........................900&t5d=&t5h=30&t5l=&t5s=10&t5n=5&t5c=000000&t5p=10&t5f=&t6t=NONE&t6=900...........................900&t6d=&t6h=30&t6l=&t6s=10&t6n=5&t6c=000000&t6p=10&t6f=&return=basic

      Delete
    2. Here you go, Nick: Caltrain 2024 Schedule (Guess)

      Notes:

      * I fudged turnback times to 14 minutes (vs 15 minutes) to get this to come out as a 12 train/crew timetable vs 13 trains (with 15 minute turnbacks) because that's the sort of pointless stuff I do.

      * To model Caltrain's advertised end-to-end run times I experimentally determined that 8% timetable padding (realistic for Switzerland; not remotely realistic for pre-Neanderthal Caltrain "operations") with 25s local train dwells (excessive for alternate-universe Caltrain with level boarding; super optmistic for shit-reality we-luv-freight-trainz Caltrain) and 30s limited-stop train dwells (ditto) model what their sketchy proposal.

      * The pointless second limited-stop service pattern that stops Stopping Mission Bay, 22nd Street, So. San Francisco, Millbrae, San Mateo, Hillsdale, Redwood City, Palo Alto, Mountain View, Sunnyvale, San Jose but pointlessly skips Menlo Park, California Avenue, San Antonio, Lawrence, Santa Clara in order to arrive at San José Cahill Street (where's that? Nobody cares!) eight minutes earlier is ... utterly pointless. Three service patterns aren't better than two.

      * BART's gone to a 20 minute Takt. But sure, Caltrain, you go on, I guess. 60 minute takt FTW! You do you. I'm sure everybody's all about matching up with the ten suckers a day or whatever who get sucked into riding sucktastic Caltrain. Be best!

      Delete
    3. Here is the whatever, prior to the pandemic, without the reverse hype and hatred. May it recover, then grow from there.

      https://www.caltrain.com/about-caltrain/statistics-reports/ridership

      Delete
    4. Hey @Anonymous, unfortunately the latest data on that Caltrain “Ridership” web page is from 2019!

      The most current ridership data & pre-pandemic comparisons available (from last month!) begin on PDF page 18 of the Caltrain CAC September meeting agenda package.

      See (and calculate) the percentages for yourself … but the TL;DR is that ridership recovery has long been stubbornly stuck/plateaued at under 30% of pre-pandemic for around a year or so now.

      Delete
    5. The emphasis on pre-pandemic earlier was to make clear it is the initial level, goal, "target," wish, desperate hope, or otherwise for recovery before growing from there. What may be impeding recovery, much less growth, well, we can all guess a few things. And even at low levels I'm not so sure transfers are in the neighborhood of ten daily.

      Delete
    6. And yes, Reality Check, recovery with the numbers on the meeting agenda package is literally a generational undertaking with some of the numbers. I hope any changes for the better that come result in more recovery than we're seeing. Return to work would help but we can't count on that, nor changing some offices in downtown San Francisco to homes (not easily) to generate new growth in the reverse direction than traditionally preponderant.

      Delete