UW Station, 8 LRVs, 0 passengers (during Connect 2020) / photo by wings777

Link Light Rail has become little more than an exercise in keeping a transit line running during the pandemic. Although it is one of the safest transit options in town due to social distancing, because of dramatically low ridership and the ability to choose different cars, it has nevertheless become nearly useless for getting its passengers somewhere in a decent amount of time. Frequency matters.

An obvious part of the problem is that even while Link ridership has dropped 90%, SeaTac Airport ridership plummeted 95%. The climate activist in me says that is a good problem to have, just not for the reasons I hoped.

Currently, Link runs every 20 minutes each direction during the day on weekdays and every 30 minutes on weekends. If you want to know when to be there to catch it, you have to have a smart phone, have looked up online or printed out or memorized the schedule, or have printed out Alex’s table. The schedules showing at the stations are what would have been if not for COVID-19.

Sound Transit has published its preliminary planned schedule (or at least weekday frequencies) for the service change that takes effect and runs from September 19, 2020 to March 19, 2021. Link is slated to run every 8 minutes during peak, every 15 minutes mid-day, and every 30 minutes in the evening. Planned weekend infrequency has not yet been published.

Link has been running full-length 4-car trains since the beginning of the Connect2020 construction slow-down in January. The epidemic was declared just in time to avoid a temporary restoration of 10-minute off-peak and 6-minute peak headway. For those of us looking forward to the old normal, it has been a long wait.

Now, King County Metro, which provides the operators for Link, has laid off 200 part-time operators. This comes as somewhat of a surprise since the Link schedule reductions were in part due to lack of operator availability. This seems as good a time as any to rethink the principles driving Link’s schedule, since there are ample operators available now, and we want to keep them employed.

So, let’s start with some assumptions:

  • There are enough operators available to staff whatever Link schedule Sound Transit wants to use (as evidenced by the layoff).
  • Capacity on the trains is well beyond what is necessary for passenger safety (which is nonetheless threatened by the presence of a few riders not wearing their masks properly ¯\_(ツ)_/¯).
  • Sound Transit wants to minimize the cost of operating Link during the pandemic.
  • The hourly cost of an operator is more than the hourly cost of running a light rail vehicle (LRV), but less than the cost of running two LRVs.

To keep the math simple, I will be using the viewpoint of a passenger at a station.

On weekdays, three operators and 12 LRVs (three 4-car trains) pass by in each direction most hours. What would it look like if Sound Transit were to instead run 2-car trains, and run them more often? By adding an operator per hour per direction (from the station view, that is), headway could be reduced to 15 minutes, and only 8 LRVs would be necessary. That would be a substantial savings for ST and a noticeable service improvement.

But consider how far ST could go, while still saving money, with 1-car consists. ST could run 6 consists per hour per direction (from the station view), restoring the old normal of 10-minute off-peak headway. For every operator added, two LRVs would be removed from the current fleet schedule. That should save ST a good chunk of change, even before accounting for improved ticket and pass sales.

On weekends, 2-car trains would allow restoration of 20-minute headway, with some savings. One-car consists would allow 15-minute headway, with even more savings.

If peak ridership returns (which is more dependent on downtown employees going back to on-site work and somewhat less dependent on UW’s planned hybrid teaching plan for the autumn quarter), that will drive the need to have 4-car trains. That presumes there will be a massive abandonment of work-from-home long before the pandemic is over. However, warnings of a seasonal resurgence of the novel coronavirus in October have been coming from various epidemiologists and agencies. In the meantime, we (as a country and a species) have not gotten our collective act together to tamp down the current resurgence before the seasonal one we know is likely to come.

I would be surprised if there is enough of a peak ridership resurgence to justify the planned peak capacity any time before next March. The low peak ridership would leave ST with flexibility to use shorter trains and improve all-day frequency.

Using fewer LRVs extends the life of the fleet. Using more operators leaves Transit in a better position to bounce back when the ridership comes back. I predict that when that happens, Metro will go back into its permanent chase to catch its tail on staffing all the service that it wants to provide. ST would be doing Metro, riders, and itself a favor by investing now in keeping operators working and saving money daily on operational costs and on long-term fleet replacement.

Regardless, if more capacity is needed to make riding Link safe, so be it. From my most recent rides on Link, it really has less to do with spaciousness (of which there is plenty) and more to do with riders not pulling up their masks, all the way over their mouth and nose.

90 Replies to “A way to restore Link frequency and save money”

  1. ST has reported cost per hour on a per car basis to make the comparison with bus operating costs look better. It’s a gross distortion because adding an additional car is a relatively small incremental cost (if you already own the equipment). The big cost is system operation and maintenance. That’s more of a fixed than incremental cost. The only major cost saving that can be had is elimination of Sunday or weekend service. The big incremental cost is labor. Part time operators are laid off first because they have the least seniority. So now each operator is not just the cost of hourly wage but benefits and pension funding.

    1. What should transit drivers be paid? Whatever befits about a million years’ evolution of the only control system capable of dealing with the multitude variables of sensory input and reflex necessary to prevent every plane flight and transit run from becoming one more multi-casualty strain on our already overworked hospitals.

      Seems we’ve very little pushback over paying billions with bonuses to people whose skill is manipulating controls that are a hundred percent imaginary symbols.

      Who’ve come to attach no importance whatever to the fact that we didn’t evolve from financial institutions, but from creatures whose second to second lives depended on what actions they took in response to a flashing barrage of what they could hear, feel, and smell. Since ability to work in the dark was really key, eyesight was probably an add-on.

      Intelligence called “Artificial” is the last thing a computer program was told by the stupidest, drink-and-drug-impaired, exhausted, vindictive, genetically-damaged, ill-informed ex-monkey to make its keyboard “click” last. Two recent mass-casualty Boeing products pretty much make the case.

      If we value the life of anybody we care about, we need to resign ourselves to sharing our six-figure incomes with creatures who can handle, meaning mainly “by feel” a lot of really critical function in three dimensions. I-5/SR101 conjunction at full-speed rush hour plus trucks, probably more dimensions than that.

      And around machinery, smell’s an especially great heritage lifesaver, the more urgent the louder you’re ordered not to pay any attention to it.

      Mark Dublin

  2. This is an excellent idea. Some points of comparison and food for thought from BART land:

    1) Running shorter trains at baseline frequency during midday was standard operating procedure pre-covid, though the agency’s stated goal was to reduce wear and tear on the aging first-generation fleet.

    2) Riders and advocates have called for a similar approach to be used on weekend for years, but the agency has maintained that either demand was too low, or more recently. that long headways are needed to allow for single-tracking during maitenance windows. (It sounds like this will be less of an issue for Sound Transit going forward.)

    3) Conventional wisdom amongst riders and advocates is that the real reason for not increasing weekend frequencies with shorter trains is that the extra labour costs would actually outweigh the cost savings of reduced fleet mileage and electricity consumption. For both BART and Link, the only way to be sure is to find the actual unit costs for an operator-hour and car-hour, then do the math. In theory, this should be possible to do from public budget info.

    4) Lastly, while Translink is different in many ways from both BART and Sound Transit, it’s quite impressive that SkyTrain has apparently gotten back up to something like normal service. Could the Bay Area and Puget Sound afford to preserve more bus operations if core rail transit was automated?

    1. Great points.

      Another issue is security and fair enforcement. To a certain extent this scales, though. Just because you run twice as often doesn’t mean you need to have twice as much security.

      1. Agree – security and fare enforcement should scale with ridership and system size, not frequency.

    2. Ellen, reason Sky Train can be automated, is because it has one thing Link doesn’t and for cars, should be illegal even to think about: Its whole trackway is a horizontal express-elevator shaft.

      Zero vehicle cross-traffic even possible, though have seen driving that classifies as a good try, usually giant black pickup trucks. And human intruders spotted and swiftly removed, with instant notice to train-control system-wide.

      Until every foot of track south of International District Station is either under-cut, like every major MLK crossing should’ve always been, or elevated, Link. Just. Can’t. Automate. end of story.

      Great picture!

      Mark Dublin

      1. Strongly disagree. We may not have the technology to have driver-less buses operate in general traffic, but we currently have the tech to run driver-less shuttles in moderately controlled environments like a college or corporate campus, which means we can certainly navigate at-grade crossings right now in 2020. A robot train will be much safer on MLK than any human driver.

        I don’t mind the Siemens vehicles given they were procured based on specs nearly a decade old, but if we aren’t looking at driverless tech for the series 3 vehicles, then either our leadership is incompetent or the unions won.

      2. Mark, you’re right to note that the at-grade right-of-way in Rainier Valley would be a challenge. As AJ suggests, one solution could be to make future light rail vehicles smarter, using similar technology to driverless buses in partly-controlled environments. Full grade separation would further increase speed and reliability, with fewer technical hurdles. Either way, for Link and BART alike, automation would be a long-term discussion, but one worth starting now.

      3. Yeah, I’m not making an argument against grade separation in MLK. There are good reasons to invest in under/over passes, both for safety and for speed & reliability. I just feel strongly that grade separation is unrelated to automation.

      4. Grade separation should have been in the minimum criteria. The purpose of high-capacity transit is to optimize people’s mobility. That means running at full speed so that they can accomplish more things in the day rather than wasting unnecessary time traveling. That also enables trains to make more runs in a day, which increases frequency practically for free. And it eliminates a large class of collisions because people can’t just casually walk or drive onto the tracks from the street; they’d have to climb up to them or go down to them. Those collisions cause loss of life, property damage, insurance costs, massive delays as service is interrupted for hours, and impact on the economy as people can’t get to work or shopping or cultural/family activities. Running trains at full speed on grade-separated lines should be as basic as having station entrances and next-arrival signs.

      5. Grade separation should have been in the minimum criteria.

        That ignores the financial reality. There is only so much money. Put it this way — what would you give up to bury the line in Rainier Valley? The line to the airport? A couple of stations? How about frequency?

        You are writing about spending more money on a capitol project, when the agency is so short on money that they are running the trains every 20 or 30 minutes. If we had just a little more money we would spend it on increasing frequency of the main line — running it every six minutes, all day, every day.

        Even from a capital standpoint, burying the line in Rainier Valley is way down on the list. On the existing line, you would start with a station at First Hill. Then move the Mount Baker Station to the triangle (where the transit center is) and move the UW station into the triangle. Add the Graham Street Station, as well as another station in the U-District (around Campus Parkway). That is all before we start getting into new lines. What is more important, running the trains underground in Rainier Valley so that Tukwila riders can get to work a few seconds faster, or running a line from the UW to Ballard, shaving 15 minutes off of that trip? Its not even close — I doubt that burying the line would ever pay off — we made the right choice by running it on the surface.

        Oh, and it is worth mentioning that burying the line means that the riders in Rainier Valley would we worse off. Even from the far end (at Rainier Beach) you would spend more time getting to the platform than you save with the faster train.

        The only argument for burying the train is that it would improve headways (you could run the trains every three minutes). But with an agency that clearly doesn’t care about frequency, it is hard to see why that would matter. Besides, that went away as an issue once they decided to pair the two lines (another good choice). Both East Link and South Link will run every six minutes, while the core can run every three.

      6. If grade separation had been in the original requirements, the entire panning process would have been different, the ballot measure would have been different, and the debates among boardmembers and local governments would have been different. It wouldn’t have been “Let’s build from the U-District to Rainier Beach because we can’t afford SeaTac”, it would have been “Where should we start with a grade-separated line?” And if it had been more grade-separation minded because it was more passener-minded, it might have been more urban-minded too, and would have hired real transit experts and listened to their advice about where high-capacity transit is the most effective.

        ST chose light rail because “it can run on surface streets as well as elevated and underground — it can do everything”. At the time ST envisioned a much more surface-oriented network. The existing light rails at the time were over 90% surface: MAX, San Jose, San Diego. That made them low cost but it also made them slow. ST intended to build something like that. It would have more tunnels than them because of the hills and waterways and the DSTT already existed, but it would have been surface from Mt Baker to SeaTac and presumably to Federal Way. The Tacoma dream might have died when they realized what the travel time of a 35 mph surface line to Tacoma would have been, and they’d figure out something else for Tacoma.

        If surface alignments were off the table, light rail would not have been as attractive, and that would have lead to more consideration of heavy rail and driverless technologies. Surface running precluded driverless techologies, and that in turn limited frequency because of the high cost of a driver for each run. Vancouver’s Skytrain runs every five minutes because it can: it’s cheap to run trains twelve times an hour when there are no drivers. The level crossings on MLK also limit frequency: it can’t run at more ten per hour at peak because it would screw up the signal timings of surrounding streets and create more of a barrier to crossing the street.

        “running the trains underground in Rainier Valley so that Tukwila riders can get to work a few seconds faster,”

        That’s a false choice. Grade separation in Rainier Valley would also benefit Rainier to UW trips, Rainier to Capitol Hill trips, and Rainier to Northgate trips, among others.

        And it doesn’t have to be in an expensive tunnel. It could be elevated or sunk into an open trench.

        Going back to how much surface alignment there is, ST originally envisioned a lot more. Rainier Valley was the first one through the design process, so the surface assumption was strongest then. ST would deviate from it only for physical barriers like hills, waterways, and freeways. Rainier Valley was flat and didn’t have physical barriers, so ST said grade separation was unjustified. But when Tukwila’s turn came, the city of Tukwila objected to surface on Intl Blvd because it had just beautified the street and didn’t want it torn up again, and it didn’t want Link going through a corner of Southcenter’s property. So Link became elevated south of Rainier Beach even though it was more expensive than the representative alignment. (It had to be all elevated to cross several highways without steep inclines.) Later, Roosevelt argued for a tunnel so the underground segment was extended for them, and then ST realized it was cheaper to remain underground till 95th rather than weaving around I-5. When ST2 came around, the presumption was that everything would be grade separated because the neighborhoods didn’t want surface and taxpayers were willing to pay for separation. And it was 100% grade separated at one point. Then Bellevue wanted its downtown tunnel and asked ST to economize elsewhere in East King for it. That led to surface segments in the Spring District and Redmond. Other than that it’s fully grade separated. While it’s technically on the surface in north Seattle and Kent/Federal Way, it’s in the I-5 right of way so there are no level crossings so that’s not a problem. ST3 is 100% grade separated at this point as far as I know. (Except the Ballard Link bridge, which will probably open for ships.) The point is, nothing in ST2 or ST3 was designed like Rainier Valley.

        You could ask the same question then: should the Ballard line be surface in Interbay and SLU so that it could go further north for the same price? Should there be surface segments in Fife and Everett? Should the Issaquah line run on the surface like the 271 does? Should the West Seattle line run on the surface on Fauntleroy Way?

      7. You make good points about surface light rail, Mike! It’s also important to note that the top speed of light rail is lower to sync better with surface operations.

        The ST3 rail projects beyond Tacoma were presented as more “Metro” technology. Had the discussions opened the technology options, we may have had a different outcome. I don’t think most of the advocates and elected officials fully fathom the disconnect between metros and light rail lines.

      8. At grade segments in Rainier Valley can be automated as follows:
        1) Install train gates where there are conflicts with left turns.
        2) Program trains to stop when there is an obstruction detected at an intersection. This may require somewhat slower speeds through the valley, but may be made up for with greater reliability.
        3) Give Darwin Awards to drivers who try to beat the train.
        We have had tech to do this for decades.

      9. That’s a false choice. Grade separation in Rainier Valley would also benefit Rainier to UW trips, Rainier to Capitol Hill trips, and Rainier to Northgate trips, among others.

        No it wouldn’t — that’s my point. If there is a tunnel or elevated line, someone in Rainier Valley spends more time getting to the platform than they save on the train. Even with a cut and cover line, riders would lose time from Rainier Beach, let alone Othello.

        If grade separation had been in the original requirements, the entire panning process would have been different, the ballot measure would have been different, and the debates among boardmembers and local governments would have been different. It wouldn’t have been “Let’s build from the U-District to Rainier Beach because we can’t afford SeaTac”, it would have been “Where should we start with a grade-separated line?”

        You are ignoring the reasoning that has drove Sound Transit from the very beginning. They knew that UW to downtown was the most cost effective starting line. Some argued for it on that basis. But others wanted something more symbolic, and longer. They wanted to send a signal to everyone that they could build something big. It has always been about symbolism and length.

        There is no reason to believe that Sound Transit would have made better decisions just because the line was grade separated. If anything, it would be the opposite. Mount Baker station is awful in part because they wanted to save a little money. Having even less money might have meant fewer stations, or worse stations. As it is, the underground parts have just as many flaws. The UW station is in the wrong place (it should be in the triangle). There is no station between that one and the one at 45th. We didn’t bother with a First Hill station. Station placement and station omission is by far the biggest weakness within our system, and yet most of that is underground.

        ST has shown that you can make big mistakes with a grade separated line, just as you can with one that runs on the surface. Most of the mistakes are because they are focused on quantity over quality (more miles of track, but fewer stations or poor stations). Committing to 100% grade separation would probably have made the situation worse, not better.

        The only thing we might have gotten out of it is automated trains. But even then, it is no guarantee. And even if we had them, there is no reason to assume that the trains would run more often. The biggest determinant for frequent all-day service is ridership per mile. Shorter, more urban lines are more frequent that commuter rail systems, even when you own the line. ST never thought ridership per mile was important, and that contributes to the really bad frequency.

        If they had focused on this important metric, then they would have started with UW to downtown. They they would have extended to Northgate and Rainier Valley. A line from Northgate to Rainier Beach would have much higher ridership per mile than the current line, which means running it more often would have much greater fare return. Even if it extended to SeaTac, you could use the turnbacks at Rainier Beach, which would allow the trains to run every ten minutes in the core, and every twenty to the airport. Even just running UW to downtown every ten minutes would make a huge difference — but they can’t, because the only turnbacks are in distant Tukwila and SeaTac.

      10. I’m with Ross on this. If we had a fully grade separated system, it would have been much worse. Given financial constraints, we would have had even more freeway alignments. I’m deeply skeptical we would have had driver-less technology because Sound Move was predicated on joint operations in the bus tunnel.

        And Ross’s point on station access is very important. In the RV, this is obvious, but it will also be true all over East Link. East Link will have none of the headway limitations of the original RV segment, but by using LRT technology rather than heavy rail, we are still able to:
        1. Save a ton of capital money running at grade through Bel-Red while getting much better TOD than a freeway alignment (the likely outcome of heavy rail on the same budget).
        2. Save capital & O&M money by having riders simply cross the tracks to access the other side of a station, rather than needing to put in elevators & an over crossing. Look at East Main or SE Redmond: the station is 100% grade separated, but we are still saving ST money AND riders’ time by virtue of LRT technology facilitating a very basic station design.
        3. Improve station access by having stations at grade and avoiding the need for a time wasting change in elevation, particularly with 130th.

        #3 is a repeat of the same argument in Ballard about an elevated alignment being faster for riders vs a tunnel. An at-grade station (even 100% grade separated) is the best for station access, better than elevated or tunnel.

        We can continue to reap the financial benefits of LRT over heavy rail if we don’t conflate exclusive ROW with grade separation. ST2 had much grade separation because it was building interurban sections, but with ST3 (& beyond) we should move back to more at-grade running as the spine is completed and future extensions are more intraurban, focusing on movement withing subareas rather than between. There are several future alignments where Link should be able to run at-grade, such as Paine Field along Airport road, Kirkland along the ERC, Central Issaquah (if the terminus station is outside the freeway envelope), and likely the Eastgate station with an at-grade crossing of 142nd.
        Looking ahead to “ST4”, Ballard north of Market and south Tacoma through the Mall are good candidates for RV-esque alignment, where slower speeds and close stop spacing make good sense at the end of a line.

        The only argument in favor of heavy rail is top speed, but we’ve already optimized away from speed with our alignment choices (RV for the south side, Bel-Red for the east side, Paine Field for the north side), so there are very few segments where the speed limit is relevant. IMO, Mercer Island to Judkins is about the only segments where riders would have materially benefited from a higher max speed. Everywhere else is limited by stop spacing or curvy alignments.

    3. TransLink has been able to go back to semi-normal schedules because Canada has handled the pandemic better, thus economic activity has returned faster.

      While SkyTrain is cheap to add more trains due to the automation, overall the fixed costs are fairly high. Thus, they don’t even start operating until about 7 am on Sundays (2014 timetable).

  3. Thanks, Brent. If the math works out on passenger loads per car, I think you’ve got the solution. But about willfully-sloppy mask-misuse, let’s understand one thing. To me, it’s criminal assault with an uncountable number of possible victims.

    But during my own term of service driving, while Thank God it never happened to me, if memory serves I remember three incidents where a passenger at a zone ended up under the duals of a departing bus. Running and yelling and banging on the side as the coach pulled away.

    Much less danger with trains than with buses. But “dynamics” the same. Individually and together, a platform-load of intending passengers isn’t standing still. Their own internal reflexes tell them they’re already in motion to board. With the ones in back pushing the whole crowd toward the side of the vehicle.

    Your own excellent pictorial, slogan and all, was dated a month ago. Are you now seeing general cooperation becoming more and more just “the right thing to do?” Public Service People-Does the “Seat-Hog” have a furry partner doing masks? Or a child pointing to her correct mask and saying “Thank You?”

    For plain enforcement, which was never more warranted, I really doubt either , King County deputies, Fare Enforcement, or especially SPD just now want the job. Wrong training, wrong description. This really is duty for a no-nonsense nurse. Preferably in pairs.

    I apologize for gender bias, but from life experience and some really memorable observations aboard transit, uniformed women carry the right “feel” for the specific kind of authority required.

    Employment-wise and budgetarily, mask duty aboard transit definitely counts as active duty service in a public health emergency gone Global. And nothing to lose by asking for volunteers.

    Mark Dublin

  4. It is ironic that the one part of our transit system that has the potential to be the safest is also the part that has seen the biggest reduction in service. Sound Transit security staff can require masks at the stations and on the trains. They have the ability to level fines, as well as kick people off the trains (and out of the stations). In contrast, Metro bus drivers can’t do that, nor should they.

    I think your plan would minimize the extra cost of running more often. It is possible that fare revenue alone would make up the difference. But my guess is it wouldn’t, and ST doesn’t think it would either. Financial considerations are a big part of what is driving this.

    But it isn’t the only factor. Some of it is a philosophy. They are actively discouraging transit use, on the outdated belief that it is a major contributor to the spread of the pandemic. There is plenty of evidence to show the opposite. The most important issue is wearing a mask, whether you are on a bus, a train, or anywhere inside. People are riding transit. (Not to the airport, but a very tiny number of people ever rode transit to the airport). To be clear, ridership is down. Metro ridership (which is being hurt by the very poor ST service as well as the pandemic) is down to about 150,000 (from about 400,000 last year). While this is a major reduction, that is still about twice as many riders than Link carried at its peak. The point is, people are taking transit — there is no good reason (other than money) for Link to provide such poor service.

    Just run a train every ten minutes. That isn’t as good as 6, but it is sure a lot better than 20 (or 30).

    1. Agree. Not much of a transit spine right now.

      Drivers have cameras, easy to stop train till masks are on.

      1. Psf, stopped trains stop other trains. From Angle Lake to UW Station. Leaving passengers trapped aboard every one of them, including the majority who not only have their masks on, but wear them right.

        And if an altercation breaks out, over masks or whatever else is bothering anybody, as STB recently reported happening on the 372, which also resulted in passengers confined amid trouble- I think the record for police-instigated train delays is 30 minutes but these records are always challenged by real sportsmen.

        What the driver can do is radio ahead for transit security, giving train number, coach number, and perpetrator description. So police wait for the train, not the other way around. Bad service is not a tool of law enforcement, period.

        Mark Dublin

      2. The spineless board won’t run a transit spine. Even though they’ve been dead-set on extending the spine to Everett and Tacoma.

  5. The fact that Sound Transit’s service change page doesn’t even bother to mention weekends is just more example of the Sound Transit leadership believing that the weekday service is what really matters and service on the weekend is an afterthought.

    However, when I actually observe the ridership on the ground, even with 30-minute frequency, the weekend trains are starting to get fuller. Last Saturday, for instance, I rode Link one stop in the Rainier Valley and, at least the car I was on, was full enough that people spaced 3 feet apart filled the entire car. (Even at 30-minute frequency, I still believe Link to be a superior option to the #7 bus, which is much slower, still requires up to 15 minutes of waiting, and has much less capacity, creating a risk of being passed up).

    It is probably ok, at least in the short term, to do 30-minute frequency after 9 PM, with almost all nightlife closed due to the pandemic. But, during the middle of the day, on weekends, 30 minutes is no longer sufficient.

    By contrast, I agree that 8-minute peak frequency is probably not necessary, while most downtown office jobs continue to be work-from-home.

    I think the sweet spot, for now, would be something like 12 minutes peak, 15 minutes midday and early evening (including weekends), 20-30 minutes evenings after 9 PM, with the same schedule 7 days/week, except for a small number of additional peak runs.

    I also don’t think Sound Transit has bothered to consider the impact that Link frequency has on numerous bus routes, after multiple restructures to funnel riders onto Link. I have already read various comment posts suggesting that the restructures be rolled back to workaround Link’s poor frequency, which, if implemented, would greatly reduce frequency on the buses. Sound Transit has the money to run Link at reasonable frequency. It’s time for them to stop make excuses and do it.

    1. Down some $600 million in revenue any increased frequency means dipping into the already depleted capital construction budget.

      1. Yeah, I’m sure that is the primary motivator. But ST is highly dependent on fare revenue. The frequency is so low, that more frequency will result in a significant increase in fares. Thus it is quite possible, with the mechanical costs being the same (the same number of trains going along the track) that the increase in fares more than makeup for the extra money going to drivers.

      2. In the best of times operations are always still subsidized. You can’t lose money on every rider and make it up in volume. And the basic premise that you can double your labor and not have a significant increase in cost is fundamentally flawed. There’s just not that many choice riders to fill the seats. Demand is way off with people working from home and entertainment venues (bars, restaurants, sports, theaters, etc.) open at survival level if that. Gas is cheap and there’s really no stuck in traffic reason to not just drive.

        Ironically, the service that probably would have operated at full capacity would have been Trailhead Direct. The best bang for the buck might be to run this service with 40′ coachesl

      3. We need ST’s actual budget and values decisions, not just speculations about how much it would cost and what we’d be sacrificing for 15-minute service. If it would really cut significantly into the capital budget, then maybe that’s worthwhile. Are we sacrificing frequency now for the sake of frequency in twenty years? What’s the point in that? Some of those who voted for Link and are paying taxes for it will be dead in twenty years. A rapid transit system should have the quality of a rapid transit system; otherwise what’s the point in having it?

      4. Demand is way off, but it isn’t nonexistent. Metro daily ridership went from 400,000 to 150,000. For Link, that would mean ridership going from 80,000 to 30,000. My guess is Link ridership is much worse, because frequency is much worse. That is because transit ridership is very dependent on frequency (https://pedestrianobservations.com/2019/02/25/frequency-ridership-spirals/). There is plenty of anecdotal evidence here that lots of people have given up on Link. I don’t have the numbers for Link, but my guess is daily ridership is well below 30,000.

        So let’s say that doubling frequency would increase daily ridership from 20,000 to 30,000. That is 10,000 additional fares. ST gets an average of $2 per rider. So that is $20,000 dollars in fare revenue per day. Some of those riders would be switching from bus-bus to bus-train. So lets be conservative, and say only $10,000 a day in extra fare revenue.

        Would it cost 10 grand a day to hire more drivers? I have no idea. As Mike wrote, ST doesn’t show their costs. But I really doubt that drivers are the biggest cost. You have train maintenance and track maintenance. You have security guards on the trains and at the stations. You have people cleaning the trains after they are used. It all adds up. That is why operations are subsidized — not because drivers are making bank.

        I have no idea if running the trains would pay for itself. But even if it doesn’t, it would probably fall short only by a small amount — a few hundred dollars. Better transit — which is their mission — would be worth it.

      5. The June performance numbers suggest that Link has about 20K on an average weekday in June. It’s probably closer to 25-30K now.


        When ST had ridership like that in 2011 or so, there were trains arriving in each direction every 10 minutes. Granted the operation stopped at a Westlake, but it suggests that Link now should be operating similarly to 2011. Keep in mind that more frequent trains will attract more riders, that average weekday number would surely rise another 5-10K more riders on a weekday.

        Operating at half of 2011 today because of “funding” seems questionable especially since property taxes haven’t been lowered and car tabs are still being collected at the same rate for now. Even if farebox and sales tax revenue are significantly down, the operating pot seems like it should enable more frequent Link service than every 20 minutes.

      6. Keep in mind that more frequent trains will attract more riders, that average weekday number would surely rise another 5-10K more riders on a weekday.

        Yeah, that is exactly my point. The fare revenue would make up for much of the cost of extra drivers, if not all of it.

    2. The fact that the stations don’t have temporary schedule signs is a major failure in rideability. I read STB and know where the real schedule is and can write the next few runs down before I leave. How is an occasional or new rider who doesn’t do that and doesn’t use a transit app (or know that they exist) do that? They’re going to look at the schedule at the station and see that the next train comes at 10:40, when in reality it won’t come until 11:05. And the next-arrival signs have been off the last two times I’ve taken Link. (And before that they were completely inaccurate, stating 10-minute or 30-minute headways even though it was 20 minutes at the time.)

      Metro has a valid excuse for not updating the bus-stop signs: there are thousands of bus stops across the entire county. Link has only 16 stations to worry about, so one person could easily put up all the signs in a day, and the stations have on-site security personnel who could do it. Even Metro’s excuse is not that valid because it does manage to change all the bus-stop schedules twice a year for the service changes. Sometimes it changes the sign a month before it goes into effect, which is confusing, but a one-month discrepency is better than the six-month discrepency Link is now at. ST can’t change even three station schedules in a month? It doesn’t even have to be in the original format; it can just be a typed sign pasted on top of it.

      1. The posted schedules have actually been wrong for over seven months now. But who’s counting? I guess I am.

        ST’s timing for going schedule-book free was perfect, but that needs to be accompanied with the small effort it takes to remove the existing posted schedules that will never actually happen again, and replace them with the current schedules (unique to each station and direction). For the life of me, I have no idea why the incorrect schedules are still up.

        If posting the station-specific schedules is too expensive, at least post the PDF for the whole system.

    3. 30 minutes after 9pm would be welcome if we could get back to 15 minutes Saturday and Sunday daytime and between 6-8pm. That’s when people are shopping or going to parks and appointments. Having 30-minute weekends takes away two days where you can effectively do errands, and for people working 9-5 weekdays at home, it doesn’t give them any days to do errands when Link is semi-frequent.

  6. The cost of adding an extra car to an LRV consist is very small. LRV’s aren’t buses. Buses have much higher maintenance needs due to their herky-jerky motion and rough riding surface. It is fundamentally why buses have such a short operating life compared to rail rolling stock. They basically just get beat to crap and fall apart at a relatively young age.

    There is an incremental electricity cost to adding an additional LRV, but I would be very surprised if it offset any significant portion of an operator’s salary.

    Don’t get me wrong, I would love to see increased frequency, but claiming it can be paid for with reduced LRV numbers is alternative math at best.

    1. I agree. The incremental cost savings associated with running a train with fewer cars is so tiny, I don’t think it should be done. If anything, spreading passengers out over more cars means less COVID, so should be done, even if the ridership level could theoretically squeeze into fewer cars. The only reason to run a train with less than 4 cars is limitations related to the size of the fleet.

      At one point, about 10 years ago, Sound Transit tried to penny pinch and run one car trains on weekends vs. the 2 car standard at the time. It saved all of $400,000/year which, for ST, is the equivalent of sofa change. When ridership surged, that plan was thankfully abandoned.

      At the end of day, it all boils down to labor being expensive and electricity being extremely cheap.

    2. ST at least needs to give us concrete reasons if it really can’t afford additional service. I.e., it would cost $N, and ST’s total monthly cash flow is $P and its fixed expenses and higher-priority are $Q, and these higher-priority expenses are R, S, and T. Without that disclosure ST is basically saying “Trust us, we know best.”

      1. I’m with Mike on this one. Keep in mind, it is not just the train cars, it is the track. The wear and tear on the track is of course proportional to the number of train cars per hour (or per day). You are basically ignoring what Ellen wrote above, and the experience of a similar agency (BART).

        As I wrote up above, the frequency is so bad now, it is taking a huge hit with regards to fare revenue. It is quite possible that the increase in fare revenue (which can be estimated based on previous studies) is more than enough to pay for extra drivers.

      2. @Mike Orr,

        There is no reason not to trust ST on this.

        I know there is this little cottage industry of sowing distrust of ST and everything they do, but ST’s goals regarding service delivery are in line with the public’s. There simply is no reason to believe ST is up to some sort of nefarious plot to deliver less service than they are capable of. There is no such thing as STAnon.

        And if you want data to do math with, then you can probably glean it from the public data (with a little effort on your part). If not, then file a public records request.

      3. @Lazarus — We are not accusing Sound Transit of some hidden agenda. We are accusing them of poor management. It really doesn’t take much effort to come up with examples of previous ST made mistakes — from very poor station placement (UW, Mount Baker) to bad escalators to dozens of poor planning decisions, the agency is far from perfect. They make mistakes.

        And this is one of them. They cut service because of the pandemic. They actively discouraged riding (as did Metro). The belief was that the pandemic was spread through public transportation. I don’t fault them at all for those decisions.

        But times have changed. Ridership has stabilized. We now know that wearing a mask is critical, and that if everyone wears a mask, public transportation is (reasonably) safe. From a public health standpoint, it makes more sense to encourage people to use the train instead of the bus, simply because it is very difficult to enforce a mask mandate on a bus. From a financial standpoint it is quite possible that the cutback in service is actually losing the agency more money than it is gaining from lower operator salaries.

        I seriously doubt anyone at ST has done the math. Do you really think there is someone at Sound Transit who has calculated the increase in fare revenue that would come from increased frequency and compared that to the additional cost of adding drivers? I sure don’t. There is nothing in the record of Sound Transit that suggests that kind of sophistication. Quite the opposite.

        Besides, if someone has done the work, then why not make it public? Even without the fare estimates, it would be pretty simple to show a chart showing how much extra it would cost to run the trains every ten minutes. If the agency is run as well as you assume, then Sound Transit officials should be looking at this chart every day, and debating it. If so, then revealing this document should be trivial.

        Either they are too lazy to release the chart, or they simply haven’t bothered to produce one.

  7. Ross, pandemic-wise, the ease of mask-enforcement, whoever does it, makes Link far friendlier than buses to public health. But is ST’s philosophical aversion to Link a matter of conjecture, or have you actually had Board members tell you that they believe trains spread COVID?

    Reason I’m asking is a strong feeling of my own that there’s a well-funded right-wing fake news campaign to defund transit so that good old oil and coal can Make America Great Again. Especially now that Texas has gone all Liberal over wind and solar just because George Soros says they’re cheaper.

    Meaning there’s something we all can do. Any Sound Transit Board member starts talking like they’ve drunk the Fool-Aid, online, on Twitter and on their office phones call them on it by name. Just don’t march on their suburban house. Like in Carmen’s case, result is photo-op for every idiot in driving range with a gun to pose while defending them.

    Mark Dublin

      1. I don’t see why the electorate would be more conservative in a presidential race with Dumping Trump on the ballot than in an obscure August primary that most people don’t even know exists until they get their ballots in the mail. If O’Ban couldn’t get 50% this time, he’s probably toast.

      2. I was glad to see Eyman got only 4.5% in the primary. That must have been all his donors and rabid supporters (24,653 of them). And Mr Tsimerman, the incoherent public-hearing disruptor got 0.05%, or 288 votes. Inslee got, hmm, 68% (377, 602 votes as of Friday). That’s, um, a supermajority.

      3. Oh, sorry, I was looking at the King County results, not the statewide results. Still, the statewide results only raise the other two slightly: Eyman to 6.61%, Tsimerman to 0.07%. Clark County and Eastern Washington aren’t taking the “abolish taxes” or “blow up the state” bait.

      4. Clark County and Eastern Washington aren’t taking the “abolish taxes” or “blow up the state” bait.

        Have you heard any interviews with Culp? He’s about as wacko right wing nut as they come.

      5. Inslee has 1,016,523 votes which is 49.81% of votes cast. Inslee is lucky he’s running against such a weak candidate in the general.

        Turnout was relatively low in King County and Seattle. This was a primary election, and for those on the left here, not much to get excited about. In contrast, it was a wet dream for reactionary Republicans. Inslee got three times as many votes as Culp, but Culp had way more signs. Heck, even guys like Eyman and Freed had way more signs — and they were crushed. It is like Bolt running against Tom Hanks — don’t expect Bolt to set a record in any of the early stages. He is just gonna coast to victory.

      6. I didn’t see a single Inslee sign the entire campaign cycle. Not that they’re needed, since everybody already knows who he is.

        I did see several signs for both Culp and Freed, plus a ton of signs for Republican candidates in my congressional and state legislative districts (which each ended up losing by big margins). Going by just signs, you’d think Kirkland had the politics of Moses Lake.

        But, at the end of the day, I fully expect the silent majority to re-elect Inslee, and it won’t be close. As RossB mentioned, Democrats had little reason to vote in the primary, but will definitely be voting in November. If polling is remotely accurate, Culp would need to either win 10-20% of the Joe Biden voters, on top of 100% of Trump voters, or see drastically lower turnout in and around Seattle, compared to 2016. Neither is likely to happen.

      7. My knowledge of Culp is that:
        1) He proudly refused to enforce a state initiative, raising the minimum age to purchase a gun from 18 to 21.
        2) He opposes virtually anything to contain the spread of COVID – be it mask mandates or forcing certain types of businesses to close.

        I don’t know much else, but I shudder to imagine what our case count would be under the watch of somebody like him.

      8. I haven’t seen any campaign signs in central Seattle that I remember. I saw some in Bellevue but they were for Bellevue representatives.

      9. This was a primary election, and for those on the left here, not much to get excited about.

        Yeah, I guess Seattle leftist are too busy breaking windows and setting fires to be bothered to vote. My guess, their so busy proving white privilege they didn’t have time to register to vote. I never thought white trash would apply to Washington but sadly that’s proved true.

      10. There are 740,000 people in Seattle, of whom at least 500,000 are left of center. The number of people who broke windows or set fires is probably ten or twenty. The number of people who look for white privilege under every rock may be a few thousand. Primary turnout is always low; it may be lower this year because everybody is preoccupied with the pandemic and some laid-off workers are preoccupied with making ends meet when the unemployment funds run out, and Inslee seemed in a safe position. It’s also possible that the tons of problems this year have increased voter apathy.

      11. [ad hom]

        But again, the reason so few people voted from Seattle is that very few races were critical for them. I voted (I vote in every election, unlike many of the Republicans who are ran for governor) and it was the easiest ballot I’ve filled out in years. The only race that made me pause was Lieutenant Governor. Holy smoke — Lieutenant Governor! That is a race that is normally less important than dogcatcher. Even in the primary it didn’t matter much, as one of the Democrats would win the nomination. Oh, look at that — both won — pretty much destroying your idiotic theory. When it matters, the left wing voters show up. For many, this was a boring primary. Even in contested races, it was obvious who the last two candidates would be. For most people in the city, this was an extremely unimportant election.

        Outside the city, things were different. The Tenth Congressional District was by far the most important contested race in the primary. But if you lived in Seattle, there wasn’t much excitement, which explains why voter numbers were low. Things will change in the general, when there will be real races.

      12. There were no local races or issue questions on my ballot at all, only state and congressional positions. That might have been another factor: there was no city councilmember or school board member that might have motivated people to vote for or against them.

  8. We are just about a year from having Northgate Link. Operational testing should be anticipated and some testing is probably well underway. At some point soon, ST will have to run trains through to Northgate as part of this testing. Testing at higher frequencies seems almost required soon — just to verify that the system can handle more frequent trains.

    Related to this is the purchase of new Siemens vehicles. ST has been taking deliveries for new vehicles for over a year now. I’m not sure how the vehicle testing is going but I don’t see vehicle shortages being a problem to having more trains.

    ST has been at the table as other systems are planning to stop running routes into Downtown Seattle — and to have service at half the frequency of what we had in 2011 (and 30% of the riders compared to 2019) is fundamentally a broken promise to other transit operations as well as the taxpayers.

    And where is a local elected official going public about this frequency problem? We have a some members of council who seem to be adept at creating media events with protests for their egos and credentials — yet are silent about providing decent public transportation for those same protesters. Their silence about this in contrast to leading protests suggests that their intent is to create self-focused visibility but not really care what impacts their constituents. It’s shameful that we aren’t seeing elected officials demanding more frequent Link trains — and it exposes how some of them are being mere egotists riding on a movement’s agenda.

    1. If I was Metro, I wouldn’t commit to a restructure until ST commits to decent frequencies. If you can’t run Link more often than the 41, then Metro should just keep running the 41.

      1. I agree. Under present conditions, don’t restructure anything. Ignore Northgate Link. If it only runs every half-hour, it might as well not exist.

      2. This will become an issue very soon as Metro is planning to send the final proposal to the council this fall. On the one hand, if we don’t truncate routes now when major new service is opening, we may never get another chance. On the other hand, this is an unprecedented situation where the supposed trunk is running as infrequently as buses did five and ten years ago, and ST has not committed to restoring frequency by the time Northgate Link opens. It would take a monumental effort to get Metro to rethink the reorganization, especially only a few months before finalization, and it would probably be just as hard to get the council to reject it. But what else could we do to get the council’s and ST’s and city’s attention that this is an urgent issue?

      3. As I’ve said before, Metro and Sound Transit need to work together to produce a sane transit network, as if they were one single agency. Running a subpar bus network with one agency because the other agency is too cheap to run the train frequency enough is just stupid.

        The efficiencies from truncating routes like the 41 would easily pay for a couple more drivers to bring up the Link frequency. Similarly, the money we’re talking about is a drop in the bucket, compared to ST’s capital expenditures. Even cannibalizing service hours from the buses Sound Transit runs to beef up Link would improve productivity. Even at half-hourly frequency, Link carries far more riders per hour than all of the ST Express routes.

      4. ST was never meant to be an operator of transit service. If fact it is in direct opposition to their charter to build the infrastructure. There’s a reason KCM provides the operators for Link and ST Express drivers and bus maintenance is left to the county transit agencies. I get the outcry for Link service that isn’t useless. But the funding has to come from Metro. ST spending future funds on current operations, especially when demand is in the tank, would not serve the interests of those who voted to tax themselves. Yeah, those paying taxes now don’t get $hick. That’s true for Link construction in general. Dipping into an already depleted capital budget just sets ST up to look even worse in the future when they can’t deliver on time and are WAY over budget. The time value of money will have profound effects in the future. Seems to be a fundamental problem with the “progressives” that can’t grasp the concept that there is no free lunch.

      5. What about subarea equity? The taxpayer’s left hand in the ST north King subarea is paying for extra (and relatively less safe) 7s and 106s to serve the riders abandoned by Link, while the taxpayer’s right hand in the same area is thinking it is saving a little on light rail service. Or maybe the outer portions of the KC Metro service area are actually subsidizing more 7s and 106s to speed up construction on portions of Lynnwood Link and a small section of East Link. If the latter, wealth is being transferred from the poorer areas of the county to the richer section. I’m not sure that was the intent.

      6. The August election did feature several state house/state senate races and one open congressional seat featuring centrist Democrat vs. progressive Democrat contrasts. So, the politically engaged progressives definitely had reason to vote.

        Of course, the number of people who are politically engaged enough to care about a state legislative race (beyond fill in an oval when they’re already voting for governor) is extremely tiny.

      7. (Sorry, accidentally posted my previous comment to the wrong thread)

        I suspect, from the Sound Transit Board’s perspective, that frequency cuts to Link is more about sending a symbolic message to their suburban taxpayers (e.g. outside the area where Link currently serves) that they’re doing everything they can to cut costs and be responsible stewards of taxpayer money, than actually saving enough money to make a real difference in capital projects. Of course, Link’s current service area does have at least one board member, but one board member is not sufficient to pass a budget.

        It’s like proceeding with plans to have your kitchen remodeled after getting a pay cut, and telling your family that you’re saving money by cutting your own hair. Reality: if you can afford the kitchen remodel, you can afford a haircut. And, if you can afford to build Link to Everett, you can afford to run the service we’ve got more than every 30 minutes.

        Perhaps one short-term solution is for Metro and Sound Transit to strike a deal where Metro subsidizes bringing up Link frequency, and pays for it by reducing service on parallel bus routes, like the 106. But, in the long term, Sound Transit promised to pay for Link operations, and they should be keeping their promise.

      8. The August election did feature several state house/state senate races and one open congressional seat featuring centrist Democrat vs. progressive Democrat contrasts. So, the politically engaged progressives definitely had reason to vote.

        I’m responding on the wrong thread since you did :)

        I agree, there were important races. But very few were in Seattle, which has the highest numbers of left leaning voters. Even if there was a contested race, it was often down to two people who were obviously going to advance. I also don’t think it makes that much difference how far to the left a Seattle representative is — that is not the problem. It always comes down to a handful of swing districts (typically in the suburbs) and the leanings of those people. Someone who goes along with the rest of the Democrats (even reluctantly) is far more important than whether Seattle has another “progressive” leftist, or just a regular old leftist.

        I’m not making excuses for not voting, but if you live in Seattle and are going to miss an election, this was probably it. (I can’t say the same about Tacoma.)

  9. “somewhat less dependent on UW’s planned hybrid teaching plan for the autumn quarter”

    The latest UW announcement this week is that over 90% of courses will be online. The only classes that will be offered in person will be clinical courses that can’t be offered online. So I think we should assume low Link ridership at UW Station through at least December.


    1. Pre-pandemic, how much of the ridership at Husky Station was students? I would guess the majority of the people using that station are either working at the medical/research facility or transferring to/from buses. And even though classes will be online there’s still a significant number of students that live near or on campus.

    2. Even if UW is closed we need a minimum of baseline frequency. That means at least every 15 minutes until 8pm. Metro is providing it on dozens of routes. ST Express is providing it on the 512, 545, and 550. The 255 was truncated at UW predicated on a frequent transfer to Link for downtown. The 67 (66), 71 and 73 were truncated with the same assumption, and the 44, 65, 75, and 372 don’t go downtown because of a previous trunk arrangement that Link is the successor to. The 512, 545, and 550 don’t help you if you’re in Rainier Valley, UW, Capitol Hill, or SeaTac. Why are Bellevue, Everett, Redmond, and Lynnwood more important than those areas?

      P.S. I was going to add the 577/578 because I thought they alternated for 15-minute service weekdays to Federal Way, but the current schedule has that only peak hours.

  10. both peak and off-peak headways should be short. operations take millions; the capital program takes billions. the network needs short waits. the announced midday and evening headways are too long. this is no way to run a rail system. ST is being unnecessarily cheap. the Link and bus operators are from the same union but have different training and bases.

    1. Just to be clear, do train operators get hired specifically to that job, or do bus operators pick train operator and subsequently get trained for the job?

  11. “The hourly cost of an operator is more than the hourly cost of running a light rail vehicle (LRV), but less than the cost of running two LRVs.”

    I doubt that’s true. The marginal cost of a vehicle is small. Peak deployment of Link has always been about fleet size, not cost.

    If your logic was true, ST would have been running 2-car trains at higher frequencies during evening & weekend service a long time ago. Instead, they run 3 (or 4) car trains at lower frequency because the higher marginal cost is labor.

    OTOH, if we had driverless tech, then you’d see Link running 1 car trains at max frequency during weekends, to delivery premium frequency at minimal cost. In other words, most like Translink.

  12. The 200 laid off operators are KCM bus operators, not Link operators. There are only 80-90 Link operators total. Halving headways would take many more operators than are currently available.

    You didn’t hear it from me, but the planned frequency for weekends is 15 minute daytime, 30 minute nighttime (similar to weekday but without the extra rush hour trains).

    1. So what happened to all of the old Link operators? A year ago they used to run the trains a lot more often. Either they are paying the operators to do less work, or they laid off a bunch of operators. At some point ST will have to increase frequency. Might as well get started now by rehiring those old operators, or going back to full time for the existing ones.

      1. I doubt any of them let their CDL lapse. They are all high seniority Metro employees so they will get routes the laid off drivers were operating until the next shake up. Some, seeing the writing on the wall probably opted for the Voluntary Separation Package.

      2. So, they force 200 layoffs before they consider which applicants get the voluntary separation package. Something (well, history) tells me they will be paying more overtime than ever once the vaccine works.

  13. Unfortunately you can’t just throw laid-off part time bus operators onto light rail trains. Training a light rail operator takes two months, assuming you have the aptitude and skills to pass the testing and operational exercises.

    1. You can’t even just throw laid-off part time bus operators onto full-time bus routes. They go through another class for full-time operators. Those classes, if they are happening right now, would be very small, more for indoor social distancing reasons than for hiring freeze reasons.

  14. From the link to laid off part time bus drivers, “Also, a new program announced by the Department of Community and Health Services will employ some of our laid-off transit operators as part-time King County Health Ambassadors.”

    Following their training, Ambassadors will be available with information and supplies beginning the week of August 17 at Metro high ridership stops and Transit Centers

    That’s a win win.

      1. Bernie and Brent, this announcement leaves my whole month MADE! Not even going to rag on Dow Constantine anymore for saying transit is only to get people to work, not to employ them. He’s got a lot on his mind. Also good to have Ambassadors somewhere in Government who does not have to be a rich donor.

        Mark Dublin

  15. My subject wasn’t so much Tim Eyman’s electoral chances, but individual ST Boardmembers’ motivation. As people, does a majority of the Board really believe the trains spread COVID, or are they just acting like it?

    But AJ, here’s what I’m telling my elected officials about automation: “If you want a dime of mine to get RE-elected instead of any opponent, you won’t allow anything but a human driver on the controller-handle of a train running any route whatever where human or vehicular intrusion does not take superhuman effort.

    Two years’ part time and eleven years’ full at the controls of a sixty-foot passenger vehicle on the systems’ heaviest-hauling routes tell me this: as far as MLK’s present setup or or any trackway that anyone can drive, bike, or walk across, only things blank on the negligent homicide charge are time, date, train number, and body count.

    Since the defense will likely call you for a witness, if you think driverless transit in mixed traffic is not prima facie criminal negligence, you need to get to work getting ready to save its bacon. With my geneological life expectancy, no rush about it. BTW- what’s automation’s in-traffic experience elsewhere?

    Mark Dublin

    1. What AJ describes is a Grade-Of-Automation level 4 (GoA4) system (fully automatic train operation, no operator required on board). As far as I can tell, no GoA4 system exists in any passenger system with level crossings or substantial at-grade operation anywhere in the world. Once you account for all the wrongful death lawsuits and pedestrian bridge construciton, the cost savings dissapear.

  16. If Portland Max is any example, from Mt. Baker station south, we can undercut about a half dozen main thoroughfares like Alaska, Othello, and Henderson, and make motor traffic on every lesser street simply turn right on MLK rather than cross it.

    And run a cable fence down the centerline of two side-platform stations, Columbia City and Othello, to make sure nobody’s easiest direction-change is crossing the tracks. Mt. Baker station seems like good model to follow.

    But system-wide, when drivers are no longer required to be constantly at the controls, do two things with them. One, as I think Sky-Train does, equip end car on every train with controls behind panels that somebody can uncover when the computer needs a break.

    And even more important, train and equip these personnel to the job description of “Conductor.” Whose passenger-information training and inclination will give passengers the benefit of speaking with someone who literally knows the system by heart.

    Fare-wise and just plain behavior, guarantee anybody intending trouble will find something automated to ride on and bother people.

    Mark Dublin

    1. Yes, I agree that MLK could have preventative design treatments to prevent random pedestrians from getting hit by a train — especially with fencing. It doesn’t have to be plain fencing either; there are many designs that could make a barrier more attractive. I’d also agree that some of the minor crossings can be closed. Finally, rail crossing gates could be added where lots of accidents occur or train speeds could be increased.

      The solutions could also be applied only on some segments. I’d propose reworking the crossings south of Othello or Renton Ave as well as north of Columbia City, and leave the segment between Othello and Columbia City (with a planned Graham station) with less barriers.

      Building grade separations is much harder and way more expensive. Perhaps some of the left turns for traffic could be jughandles or Michigan lefts. The places that seem to need them the most (Henderson, Othello, Alaska) are literally next to stations and that makes them tougher to fix. Unless ST has funds to move stations to one side AND put overhead or underground (and they don’t) I don’t see this happening for at least 10-20 years.

  17. Should’ve said “completely automated” or “union-wage-zero” to ride on and bother people.

    Mark Dublin

  18. UW and Westlake were up 12% versus the year-ago quarter and Chinatown / ID was up 20%, suggesting both more North end riders transferring from buses and more people exiting the system at the beginning and end of the tunnel.

    Yeah, in Q4 of last year, there was solid growth across the board, but especially in the north end. Of all the stations, ID had the biggest jump, while SoDo had the biggest drop. This suggests a lot of people who normally would get off at SoDo just getting off at I. D.

    I think it is interesting that prior to the pandemic, the three most popular stations were Westlake, Capitol Hill and the UW. The other downtown stations were also very popular. As expected, it is clear that we built the system out of order. The most cost effective section (by far) is from the south end of downtown (I. D.) to the UW, and this is with only one station close to the UW (there should be three) and no station in First Hill.

    1. Oops. I commented on the wrong post. Moderators, feel free to remove this (as it is off topic).

Comments are closed.