31 May 2022

Capital Spending for Better Service

Wouldn't it be great if you could quantify the service benefit of capital improvements, to compare and prioritize them by how much better train service results?  We can, and using our handy Taktulator, we will. This service pattern evaluation tool was formulated around time-based service quality metrics. We use it to explore future improvements to the peninsula rail corridor.

Today's 2022 Timetable: 94 service points -- The current peak schedule with four diesel trains per hour features very generous padding and SF - SJ trip times ranging from 66 minutes (express) to 99 minutes (local). The less-than-100 score indicates that service quality has dropped since 2011 when there were five trains per peak hour. The Taktulator score is calibrated such that the 2011 Caltrain timetable scores exactly 100 points.

Caltrain's 2040 service vision foresees eight trains per peak hour per direction (not counting HSR). Let's start with a service frequency of 8 trains per hour-- except for the sake of exploring and quantifying the value of capital improvements, we'll start from a hypothetical case that will never happen: eight trains per hour of today's diesel service, making all local stops.

Hypothetical diesel all-stops local, 8 tph: Score = 109 service points (+16%) -- The doubling of hourly frequency improves the service score by 16%, despite each train being slower. The extra time riding an all-stops trains is more than offset by the much shorter wait time at the station. For example, maximum wait times in Belmont plummet from one hour to just 7.5 minutes. Unfortunately, this service pattern would take an unrealistic 32 trains to operate, because each train takes 94 minutes to go between SF and SJ. The hypothetical scenario still illustrates the magnitude of the effect of doubling frequency.

Add electrification: Score = 121 service points (+11%) -- Electrification is worth another +12 points relative to diesel, thanks to the shorter trip times that come from the higher acceleration capability of EMUs. Those savings accrue to a full ten minutes between SF and SJ for an all-stops local. Station dwell times are still booked at 45 seconds, a longer duration that reflects the lack of level boarding. Thanks to the faster trip times, the fleet requirement has dropped from 32 trains to 28 trains. Service speed saves money, not just on fleet size but also by increasing the hourly productivity of train crews (in terms of passenger-miles served).

Add Redwood City hub station: Score = 131 service points (+8%) -- If trains cannot pass each other, there is no room in such a frequent timetable for express service. A new four-track station at Redwood City, where express trains can overtake locals on opposite sides of the same station platform (so that passengers may transfer seamlessly between local and express) gives the best of both worlds: frequent service AND express service. For now, we'll assume this station has only two-track approaches, requiring trains to arrive and depart serially. In practice, this means every local must wait more than 5 minutes or the equivalent of two signal headways to let the express catch up before RWC and then pull ahead after RWC. The stopping patterns start to look like Caltrain's 2040 service vision.

Add Redwood City quadruple approach tracks: Score = 138 service points (+5%) -- If quadruple tracks are added approaching Redwood City from the north and south, then local and express trains can make parallel moves into and out of the hub station, removing the requirement for every local to wait there for five wasteful minutes. To unlock this benefit, the quadruple track overtake section needs to extend to one station on either side of RWC, so every local train can make productive use of those five minutes. In the Taktulator, we simulate this by having every local train stop at San Carlos and Atherton, which (despite its closure) stands in for a new Fair Oaks infill station at 5th Avenue. This suggests a hub station is about 1.7x more effective if it forms the center of a three-station quadruple track section. Having fully half your trains save five minutes is a huge service improvement!

Add level boarding: Score = 147 service points (+7%) -- Where electrification saved time in motion, level boarding saves time at rest by shaving 15 seconds of dwell time at each station, as step-free access smooths passenger boarding and alighting. Level boarding gives not only short dwell times but predictable dwell times (for example, wheelchairs don't take longer to board) so we can also tighten up the padding margin in the timetable, cut in this example from 12% to 7%. Interestingly, the end-to-end corridor times fall below a threshold that allows turning a train sooner, reducing fleet requirement from 28 to 24 trains. This isn't necessarily an effect of level boarding itself, and only illustrates that a series of small improvements can result in a discontinuous benefit when a certain threshold is reached.

Add SF Downtown Extension: Score = 250 service points (+70%) -- There are more jobs (over 100,000) located within a half mile of the Transbay Transit Center than there are jobs within a half mile of every other Caltrain station combined. This makes downtown SF a dominant node if added to the system, a fact that is reflected in our census-based weighting of available trips. No other improvement comes close.

Here is how these service improvements stack up against each other, plotted as the logarithm of the ratio of after/before scores, which gives you their relative impact. They can be constructed in a different order than imagined above, but the relative proportion of each improvement should remain approximately similar:

Bar graph of the relative service quality improvement of Caltrain capital projects

Here are some key takeaways:

  1. Grade separation projects do not improve train service. Exceedingly rarely, they do prevent a train delay, something that is not captured in this analysis. On the basis of the time metrics of a typical trip, however, the service improvement of grade separations is ZERO. This should factor strongly into how many billions we are collectively willing to spend on them relative to the other capital improvements discussed here.
  2. The benefits of electrification alone (without other improvements) are mediocre at best. On the basis of our time metrics, service quality is only improved by about 11% relative to an equivalent diesel scenario. Caltrain can't just finish the electrification project and call it good enough.
  3. The Redwood City hub station now in the planning stages is surprisingly beneficial to service quality. While packaged and sold as a grade separation with a bonus of expanding the train station, it is hard to overstate the service quality benefit of the new hub station. Even as planned by Caltrain (with two-track approaches from the north and south) the new station produces nearly as much service improvement as the entire electrification project.
  4. The Redwood City hub station as planned by Caltrain with two-track approaches is operationally ineffective. It can be juiced up to 1.7x more benefit to service quality by making it the center of a four-track overtake facility spanning just three stations: San Carlos, Redwood City and a new Fair Oaks infill station at 5th Ave. The southern portion of this four-track facility already exists today. Together with 4-track approaches, the Redwood City hub improves service quality by a greater proportion than the entire electrification project! That's why it is critical that planning for the Redwood City grade separations allow for four tracks throughout.
  5. Level boarding provides over half the service quality improvement of electrification, and is likely to be a much cheaper capital investment. However, it makes sense to do it after the hub station.
  6. The downtown extension in San Francisco will be a game changer for service quality. The transportation industrial complex knows this and will make us pay dearly for the DTX project. However, the additional billions for the PAX (Pennsylvania Avenue Extension, a city-desired grade separation) add absolutely nothing to service quality, and should never be allowed to be bundled with the DTX project. Every capital dollar should improve service quality.
  7. The Redwood City hub station (with four tracks, not two!) is worth one fourth of the service benefit of the DTX. That means we should (a) not be shy about spending capital dollars to build it and (b) stop selling it as a grade separation, because that isn't the story here-- it should be about a new infill station, seamless transfers, and better service quality system-wide.

As always, the analysis provided here can be quibbled with and improved upon, and you are encouraged to "do your own research" by trying out your own service patterns in the Taktulator.


  1. This is really interesting information - I would have thought that grade separations and level boarding would have a larger impact on service!

    I have a few questions - I hope it's not too much work:

    1. Grade Separations: Do grade separations have anything to do with the maximum speed of the corridor? If I recall correctly, there's a limit based on the type of grade crossing.
    2. Grade Separations: I would think this changes the calculus of pushing for grade separations, re: the benefit being zero. However, wouldn't the decrease in delays and accidents be somewhat cost-effective in the long run? Related to that, what do you think will happen with the grade separations planned for the corridor (e.g. HSR paying for it, cities paying for it...)?
    3. Straightening: Are there some high impact areas where some straightening would increase speed? (If so, what's the relationship to this analysis, e.g. would eliminating one or two curves be as cost effective as electrification?)
    4. Quad Tracking: Are there other stations along the corridor that should be quad-tracked, e.g. Millbrae/the SFO station for similar benefits?

    1. RE: the grade separations having an impact on speeds


      High-Speed Rail at Grade Crossings

      The FRA’s goal for high-speed rail grade crossings is to achieve an acceptable level of grade crossing risk. Regulatory requirements for high-speed grade crossings are:

      110 mph or less: Grade crossings are permitted. States and railroads cooperate to determine the needed warning devices, including passive crossbucks, flashing lights, two quadrant gates (close only 'entering' lanes of road), long gate arms, median barriers, and various combinations. Lights and/or gates are activated by circuits wired to the track (track circuits).

      110-125 mph: FRA permits crossings only if an "impenetrable barrier" blocks highway traffic when train approaches.

      Above 125 mph, no crossings will be permitted.

    2. Grade separations increase reliability but don't improve the theoretical peak service quality. They're mainly for improving connectivity across neighborhoods (e.g. the recent Bay Meadows and upcoming South SF grade separations). Increasing speed limits to 110 tph won't make the local trains run faster because they won't be able to accelerate to that speed between stations, especially if curves exist.

      Curve straightening - The top 10 curves to straighten are here: https://caltrain-hsr.blogspot.com/2009/01/top-10-worst-curves.html. It's extremely cost effective to straighten curves, especially if HSR comes to the peninsula ROW.

      Quad tracking - RWC four-track hub and its midpoint location enables 4 tph local / 4 tph express takt. These service levels can theoretically be doubled if another 4-track station is built midway between SF and RWC, which just happens to be the Millbrae-SFO station. Similarly, for the southern half of the corridor, the existing 4 tracks at Lawrence could be extended to a 4-track station at Sunnyvale. However, quad tracking Sunnyvale is lower priority because there are relatively fewer local stations south of RWC, and from a planning perspective some SF - RWC locals would go to Fremont over the Dumbarton Bridge, leading to fewer capacity constraints in Santa Clara County.

    3. Speed is something you can explore in the Taktulator. The station-to-station times are accurately simulated in a train performance calculator for each different type of rolling stock, so you can change from a KISS-79MPH-4MW (a six-car Stadler EMU limited to 79 mph and accelerating more liesurely) to a more aggressive KISS-100MPH-6MW (the same thing, limited to 100 mph and making full use of its acceleration ability to PG&E's chagrin). Re-jigger the timings to make the local-express connections, and you'll see the score barely budges.

      That's simply because the local trains don't ever reach those speeds in the available distance between stations, and also because expresses have to slow down for curve speed limits that are correctly modeled in the underlying track database.

      Speed doesn't pay for a commuter operation. If HSR wants it they can pay for it, as well as all the other overtake tracks that will be needed to allow Caltrain to continue operating a peak 8tph. It's worth remembering that the ONLY overtake facility needed by Caltrain for as far a future as you can envision is at Redwood City. All the other overtake tracks that have been discussed over the years are solely for HSR's benefit.

      (Also note, the HSR EIR currently undergoing approval without overtake tracks is fundamentally incompatible with 8 tph Caltrain and will actively degrade service quality... keeping Caltrain at a score of 250 and adding HSR on top is going to take beaucoup bucks!)

  2. The 'Add-DTX' link is broken for me (does not include DTX). In terms of scoring capital projects, there should be a cost number included in the denominator.

    1. Thanks, fixed that.

      Good question on the cost. Here's a very rough stab:

      - Running 8tph (doubled frequency) means expanding the fleet to about 30 trains or about 240 cars. There are currently 133 on order, so need another 107. So about $1B if we also toss in a new maintenance & storage facility.

      - Electrification is a known quantity, $2.4B

      - RWC2 hub station and grade separation should cost $0.5B but will likely be double that (see exploding cost post)

      - RWC4 with quad tracking, new infill Fair Oaks and new San Carlos probably closer to $1.2B

      - Level boarding at $10M per station is ~$0.25B, round up to $0.3B

      - DTX is probably at least $6B and possibly far more.

      So the bang for buck order is:
      1) level boarding
      2) doubled frequency
      3) RWC4 hub
      4) RWC2 hub
      5) DTX
      6) Electrification
      The consultant industrial complex will tackle these in opposite order.

    2. I think it's not a coincidence that the items here where the cost is more well known have the worst bang-for-buck. I do agree that level boarding and more trains should be the highest priority, however, and this is the basis of Caltrain's "Enhanced Growth" proposal. which is for 8tph with no overtakes. It looks something like this on the takulator:

      Caltrain Enhanced Growth

      My suspicion is that Caltrain's vague plan for level boarding (if they have ever even thought about it seriously) involves triple-tracking and/or gauntlet tracks for every station to maintain compatability with freight, just like SMART stations in Marin, without needing to ask for any special exceptions to FRA rules. It probably also involves some incredibly expensive new step mechanism for all the EMUs. Seems unlikely that even $1B will be enough,

      As far as a new maintenance and storage facility for all those new EMUs to increase frequency is concerned, the industrial land and closed quarry just north of Capitol Caltrain station seems like an ideal spot, and probably could be accessed by branching off at CP Lick. Long term plans for elevating Diridon envision moving CEMOF anyway.

    3. Using the same assumptions as Clem does (30 seconds dwell and 7% padding) gives the following takulator:

      Caltrain Enhanced Growth

      This has a score of 144, which compares well to the 147 for the equivalent case Clem gives for a 4track overtake at RWC. It makes the bang-for-the-buck of RWC seems a lot lower. However, the stopping pattern is certainly more complicated and it may not use the fleet as efficiently.

    4. There may be some confusion as to which service pattern Caltrain has formally adopted as its 2040 service vision. There were three scenarios on the table during development of the business plan in 2019: baseline growth, moderate growth, and high growth (see charts 36-39). The "enhanced growth" service pattern you refer to above is probably most similar to the baseline growth scenario in this presentation, featuring irregular stopping patterns.

      While the board-adopted service vision didn't spell out a specific service pattern (and why would it over-constrain), the Caltrain business plan website shows the 2040 service vision is the moderate growth scenario with Redwood City standing overtake. The board resolution does call out a "mixture of express and local Caltrain services operated in an evenly spaced, bi-
      directional pattern," which is a description of this service pattern.

      In the taktulator, it scores 241, just a bit short of the 250 in this post if you remove the overtake delay at Redwood City by adding the three-station four-track overtake section.

      As for level boarding, while it is official board policy to pursue this improvement, I don't believe there is any sort of a plan yet formulated at Caltrain. I do think the solution will involve a waiver of GO 26-D (state CPUC, not federal FRA!) and not janky gauntlet tracks. The step mechanism is modular and will likely be easy to retrofit to the EMUs, although it hasn't been engineered it yet.

    5. Clem is.correct that the "Moderate Growth" scenario was adopted as the Caltrain long term "service vision". However, in subsequent meetings they identified the "Enhanced Growth" plan as an optional interim step towards this long-term vision. I think the plan was to use this as a goal for shorter-term funding opportunities as it does not have the large capital cost of the RWC overtake. For details, see the following Caltrain meeting agenda (PDF link)

      Caltrain WPLP Committee Meeting Agenda Feb 2020

      Based on the "bang for the buck" argument it would make sense to pursue something like this in the short term.

      I certainly hope that they pursue a waiver of GO 26-D, I did not realize that was CPUC not FRA, so hopefully that makes it easier to waive. I wonder why SMART did not do this. If the cost of level boarding is really <500M it would certainly make sense to at lease start seriously planning for it now.

    6. By the way, if you include DTX with Enhanced growth you get a takulator score of around 234 using a Caltrain's proposed service pattern and 12% padding.

      Caltrain Enhanced Growth Transbay

      Using Clem's 7% padding assumptions and a few small changes to the service pattern gives a takulator score of 246

      Caltrain Enhanced Growth Transbay 7

  3. Why is a gauntlet track required for level boarding? Couldn't passenger train be made as wide as the width required for freight cars and use a bridge plate (like Brightline) to cover the rest of the gap?

    1. CPUC GENERAL ORDER No. 26-D https://docs.cpuc.ca.gov/PUBLISHED/GENERAL_ORDER/59571.htm

    2. The new EMU fleet was designed with bridge plates (motorized fold-out step mechanisms) on all the high doors, for gap-free unassisted level boarding at high platforms (like Brightline). Complications with wheelchair circulation inside the vehicle as well as general resistance to high platforms among Caltrain staff resulted in this feature being removed before delivery, although the components are supposedly still being delivered for possible future installation.

    3. How do other systems with high-level boarding on bilevel cars (like NJ transit) deal with wheelchair circulation?

    4. First, they board wheelchairs using crew-deployed bridge plates. Once on board, the wheelchair spaces are located at vestibule level, as are the bathrooms, so there is no need for circulation inside the vehicle.

      Not so for Caltrain's EMUs, which have a bathroom on the lower level that must be made accessible. This and the practical considerations of an extended construction period during transition to level boarding platforms, when boarding and alighting might occur at different platform heights, requires wheelchair users to be able to circulate between the mid-level and lower level inside the train, using lifts capable of handling 800 lb. These lifts were de-scoped from the Caltrain design, although the car bodies are designed to receive them.

  4. Hearing a rumor that the PCJPB will hold a special meeting tomorrow, to discuss "governance." If it's not announced by 5pm it's probably not real.

    1. Ok, the meeting was set for Thursday

  5. Grade separations save lives, simple as that. You cannot measure life with a speed graph.

    1. Fast and frequent transit service takes drivers off our roads, which are by far the deadliest mode of transportation. Good train service saves more lives than grade separated car sewers.

    2. "Good train service saves more lives than grade separated car sewers." Good train service is not possible when hitting a car or person brings all service to a halt in both directions for 90+ min. I've spent more than my fair share of time waiting around at SF or PA station for service to restart, or organizing shared rides, taking multiple buses and BART, etc. What happens if/when HSR service starts?

      This is Caltrain, an organization that can't quite figure out how to keep hi-rail trucks and trains off the same track at the same time. Which reminds me, the burned out train and truck carcasses are still parked next to the tracks months later. Even airlines know that part of good service is getting their wreckage out of sight as soon as possible.

    3. Innumeracy has entered the chat

    4. What would it take to make it standard policy to *not* stop all service after an accident, even one with nonzero fatalities, provided the accident train is still railworthy? For the purposes of this question, I consider some degradation of service to be acceptable, e.g. if the policy is for the driver to immediately send a broadcast "level crossing accident at (location), temporary slow order of (speed) there until the permanent way people check that the track is fine" is fine. The fact that the accident train itself didn't derail is evidence that the track can't have been too badly affected. And even if it was, why, that's what the slow order is for, the next train's driver can take a closer look at the end result; and in any case, should a train end up derailing, it would happen at low speed. Likewise, pulling the accident train out of service at the first convenience is acceptable, as long this is done in a way that doesn't block the running line.

    5. You're right. This "chat" ceased to be relevant to much of anything, numerate or not, a decade ago. I pretty much assume that there will be no HSR out of SF for at least another decade, so I see this as a forum to discuss Caltrain service (or lack thereof). Getting stuck on a train or platform for a few hours while they clean up an "incident" at a nearby grade separation does not make the service more popular, even if it "only" happens to regular commuters a few times per year.

      And, of course, there are other things. I was on a train a few weeks ago that stopped and sat at a red signal for a switch for 30 minutes. The crew verified the switch was correctly set (I was standing next to a conductor the whole time), dispatch gave them permission to pass the signal at reduced speed, but apparently the onboard PTC equipment wasn't having it. So we (and a number of other trains) sat there until they managed to reboot everything. Then there is the current chaos of not being quite sure what platform a given train will arrive at, even during commute hours. I was also recently on a NB train that arrived at the SB platform at RWC at the same an SB train arrived at the NB platform, with no prior warning to those on the platforms, big fun!

      None of this amounts to more than a few decimal points of noise in the numeric scheme of things, but it makes it hard for myself and others to have much confidence in the long term safety and reliability of Caltrain. It's fine, as long as we don't absolutely have to be someplace at a specific time.

    6. I agree that a timetable that isn't or can't be adhered to is essentially worthless. I share Marc's experience of irregular and incident-prone service. I also think that after September 11th, we started putting emergency responders on such a pedestal that we now allow them to shut down Caltrain service at the slightest whim, and that practice needs to stop. That being said, crews directly involved in incidents can understandably be shaken, something I once observed firsthand when I was traveling in a cab car that hit and killed someone. I was disturbed from just hearing the horrific sounds of it, never mind the sights the engineer had to see.

      Re: numeracy, in round numbers, Caltrain forecasts ridership growth from 18.4M/year (2019 pre-pandemic) to 33.6M in 2030 with 8 tph but no DTX, and ~42M with the addition of DTX. The average trip length is historically about 23 miles. That means the increase in annual passenger miles between 2019 and 2030 with DTX is (42M - 18.4M)*23 = about 500M passenger miles/year. Meanwhile, California's mileage death rate is 1.06 per 100M miles. So, better service would save about 5 lives every year.

      Deaths in grade crossing accidents (excluding suicides, which would not be avoided by grade separations) do not come anywhere close to 5/year even today, so we can say with good confidence that even 100% grade separation wouldn't come close to saving as many lives as having better train service.

      Notice we can do all this math without having to put a price on life.

    7. I have to disagree about the Caltrain accidents comment. In 2022, BARTs body count is multiples less than Caltrain's (plus the 13 injured in the Caltrain/construction truck crash in March). I think BARTs complete grade separation generates "good will" by being far less deadly, and by not interfering with traffic. I can imagine BART putting track/train access doors at its stations to reduce risk even further. California politicians, government agencies, and government workers in general will spend large sums of money to be on the side of the angels.

    8. Reality Check09 July, 2022 06:10

      @Reedman: the San Bruno crash into 3 on-track work trucks had nothing whatsoever — zero — to with grade crossings. And regarding Caltrain deaths: nearly all of them prove to be intentional. The occasional indeterminate ones are assumed accidental even though statistically, they are likely just unproven suicides. Truly accidental ones are, thankfully, happily, extremely rare.

    9. Only had to wait a couple weeks, and Caltrain hit another person. On July 23.

      Caltrain's five year average is 14 deaths per year. BART's is ....

  6. https://www.theonion.com/china-worried-u-s-outpacing-them-on-poorly-functioning-1849179085 ...low-speed trains