ST’s 2012 Service Implementation Plan

63 Replies to “ST’s 2012 Service Implementation Plan”

1. Fast Forward to the year 2017 in the SIP. Link will be up and running for the full year, carrying 18.4 mil riders, or over double that of today. Time to rejoice?
   A closer look reveals the very high subsidies that will be required for decades over what could have been accomplished years earlier with a bus solution, including dedicated lanes where needed at choke points.
   Operating Cost per rider in 2017 on Link will be about $10.09 per boarding [see footnote for calcs], compared to todays cost of the Seattle/Bellevue bus at less than 1/3 that amount ($3.05) [SIP pg. 171]
   Link costs in the year 2017 are based on 18.4 mil annual riders [SIP pg.108], divided into Operating costs of $73.4 mil, for a cost per rider of $3.99.
   To that, add in annual agency wide debt payments of $75mil/yr, times 43% (Links 43% share of all riders), divided by 18.4mil riders = $1.75 each
   Finally, add in depreciation on all but ROW costs for both Central and ULink of about 4 billion, divided by 50 years, or about 80 million per year. (50 is an average for rolling stock, maintenance of way, and tunnels, which last longer than track for example, so it’s a rough average). That comes out to $4.35 per rider, which can be argued as not a cost, because you don’t really have cash going out the door until it comes time to buy something – like a replacement railcar.
   Business has to account for such things, governments keep track of it, but usually ask for more money when things wear out, or new technology comes along and they need to upgrade. You can judge that one for yourself.
   The whole point here is that Link will be subsidized far higher that conventional buses are, even after the busiest segment is connected to Central Link. Ridership only falls off the further you get from Seattle’s CBD.
   It astounds me that true transit supporters can’t take an objective look at the trajectory we’re on. I’m not being mean or have any hidden agenda’s. But managers and policy makers need to face the facts in cold hard terms, then make good decisions. I fear they are all still giddy with the new train set.

   1. No, the “whole point here” is that if you look at the first year open of a 100 or 200 year system, you can shout about it.

      1. And, we VOTED to spend this money. We’ve seldom, seldom had the opportunity to vote on highway construction.

      2. As someone who frequently uses unflattering BART comparisons to describe “the trajectory we’re on” — nearly 40 years on, it’s got a system-wide $6.17 average subsidy that rises to $15 or $20 on some of the outer segments — I think Mike has a valid point.

         When you’re talking about 4-minute service to Lynnwood, it’s folly to expect these numbers to ever get much better.

      3. …Which does not mean that I think the project lacks any value (as Mike may be suggesting).

         It means that, when you build a many-billion-dollar system whose primary selling point is how far the tentacles stretch, but offer so little improvement in “connective coverage” that even those totally-excited constituents near the tentacles are going to keep drivin’ most of the time, you should probably expect some pretty bad numbers in perpetuity.

         Again, fine. You may still be serving a valid purpose. Just don’t delude yourself about what kind of system you’re building to the point where you miss opportunities to build something that works better.

      4. Mike most certainly does not have a valid point, thanks to his inability to actually use numbers properly.

         You, d.p., *may* have a valid point, as unlike Mike you seem inclined to compare apples to apples.

      5. The artillery officer can make a reasonable prediction about where shells will land, based on the initial ‘trajectory’ of the last round fired.
         Link Rail has several years of operating under it’s belt, lots of budget projections, and now a firm number for comparison of U-link costs and ridership for the year 2017. Remember, all this was hypothetical in the early 90’s when Link was being proposed over other transit solutions – that’s 25 years of trajectory Ben, which is better than waiting 200 years as you suggest to see where the shells landed. What if you missed the target?
         Using the 550 bus cost was not random, or cherry picking the numbers to find the best bus route to compare against the worst rail outcome.
         The 550 from Bellevue to Seattle is nearly identical in distance and running time for the now eliminated 194 route to Seatac. Both took 30 minutes, and had ample freeway segments. The 550 runs close headways, nearly the same as light rail, and good load factors. It is a purchased service from Metro, so the costs are fully amortized and accounted for. So this is truly what ST can purchase a bus for, that operates like a frequent LRT line. No hidden costs here. The cost of the freeway per vehicle use is minuscule in the calculation.
         Link is on a trajectory to be at least twice as expensive per rider than any bus based transit system. If this is the best ST has to offer, then I would say they have failed to provide mass transit in a responsible manner.
         All the ancillary arguments are just fluff, such as cheap electric power, and green house gases. All those factors are relative to the economy as a whole, not in isolation. When fossil fuel gets really expensive, which it will, do you really think hydro and wind will remain at bargain basement prices. Hell no, the market will prevail.
         Seattle is well on it’s way to pricing transit out of existence. Just look at your bus providers. They are cutting service to the bone. ST at it’s current trajectory won’t be far behind. That’s the real problem I see with the current crop of rah, rah blind supporters of some really wasteful practices that are happening every day.
         Somebody needs a wake up call, but apparently few here are even willing to entertain the concept of critical thinking.

   2. First of all, you’re comparing the worst possible numbers for Link with the best possible numbers for buses. There are debt and depreciation costs associated with buses too, *especially* if you consider the roads that those buses run on.

      Second, Metro could double link ridership tomorrow simply by rerouting buses to connect to Link, rather than running parallel with it. Every bus in the SODO busway could terminate at SODO or further south. Every bus to West Seattle could terminate at SODO. A ton of buses could terminate at RBS, or TIBS, or elsewhere. By dramatically increasing Link ridership (which would pretty much be free), those changes would effectively reduce the per-rider subsidy. And they would also save real money, since all those bus service hours could be saved entirely, or redirected towards more productive service.

      It’s not ST’s fault that political concerns force Metro to compete with Link, rather than cooperating with it.

      1. Maybe it is time for Board Members Constantine, Patterson, Phillips, McDermott, and von Reichbauer to have a retreat with Council Members Patterson, Phillips, McDermott, and von Reichbauer, and Executive Constantine, to work through some of these issues.

         But, um, please don’t truncate routes at SODO Station.

      2. What’s wrong with truncating routes at SODO off peak?

         Find me another city that runs so many routes parallel to a train line, all day long.

      3. “Find me another city that runs so many routes parallel to a train line, all day long.”

         Tacoma

      4. Er… I was hoping for a city that’s not part of the Seattle-Tacoma metropolitan area… and one that’s our size, not 1/3 of it… and one whose transit is *better* than ours, not worse.

         Anyway, you still haven’t answered my main question. Why is it a good use of off-peak service hours to run dozens of buses along 4th Ave S and the busway, a corridor that’s already served by Link? (And along the Viaduct, where they make no stops?)

      5. Find me a bus line that dumps riders onto a train in the middle of nowhere a mile and a half short of the CBD.

   3. “It astounds me that true transit supporters can’t take an objective look at the trajectory we’re on.”

      You know the thoughts of every transit supporter?

   4. We have buses at $20 a ride too. And we have buses at $3 a ride. To compare our ONE link line to the best bus line is silly. Tacoma link is at $3 per ride. We could compare that to the worst bus routes if you’d like.

      1. we in Tacoma don’t pay anything for the Link that runs through the city. if you went to the Sound Transit site, you would see that. can we compare to something else please?

   5. Your shallow analysis, Mike, is rendered worthless by assuming the cost of running buses will remain constant.

      It doesn’t. Have you heard of “oil prices”, just for starters? You aren’t adding in depreciation or replacement cost for buses, either (not to mention roads), so you should shut up about depreciation for rail.

      If you’re going to claim you don’t have a hidden agenda, try comparing apples to apples.

   6. Oh, and this isn’t all that on-topic anyway. :)

2. You may be surprised to hear me say this, but I now think the 574 is mostly redundant service between Federal Way and SeaTac.

   Consider the non-Normanic math: The best frequency on the 574 is 30 minute-headway, and the route takes roughly 26 minutes between Federal Way Transit Center and Seatac/Airport Station. That’s 41 minutes average wait+travel time.

   The Line A has 15-minute headway or better during most hours. I timed the trip as 32 minutes between FWTC and SAS. That’s 39 minutes average wait+travel time, not including the much faster boarding at each end.

   Dropping back from the best-case scenario on the 574 to the mid-day scenario, its headway increases to one hour most of the time. During most hours the average wait+travel time then rises to 56 minutes.

   Maybe it is time to look at canning the 574.

   1. But you are forgetting the Lakewood-TDS-Federal Way portion of the route, and the fact that the demographic that rides it (Airport Employees) would be much adverse to making a transfer to a very much slower route. Not to mention that RR “A” dosent serve the airport loop which is the major boarding location.

      1. “much slower route” Did you read the comment you were responding to? Not that your larger point isn’t valid, but…

3. Five-Sixty

   Emp-teee

4. ST seems convinced that the 512 was a cost-cutting measure, rather than the creation (or reinstitution) of a new-and-improved BRT route connecting major destinations up and down I-5. I’d still like to see the 510 become peak only, and the midday 510 trips be converted to 512s. I’d be much more likely to spend some time and money in Snohomish County if getting around didn’t involve backtracking to King County to transfer.

   1. ST is balancing expectations based on previous service and the sometimes excessive horror people have at too many stops. When the 510x and 594x were created, the only existing choices choice were 358+101, or 194+500, which took 2-3 two hours each direction. (And the 4xx which were mostly peak only.) So people seriously wanted ST to create a Seattle-Everett express, and PT had recently created a Seattle-Tacoma express. The north part of the 574 is basically compensation for deleting the 194. I agree that it’s kind of silly now, except for Tacomans going to the airport. But (I assume) it’s still politically popular. I have some vague ideas for reorganizations of the 57x, but I’m not sure what exactly would be best.

      As for the 510/511/512, I was stunned when I calculated that the 512 would only be about two minutes longer to Everett than the 510. I think that hasn’t sunk in with ST or the public yet, but we should keep mentioning this until they do it. It makes eminent sense to make the 512 to Everett a frequent all-day route, going to wherever Link ends (downtown, Northgate, or Lynnwood), and downgrading the 510 to peak only.

      1. the 59x existed for a good ten years before ST was created. I think it was created in response to greyhound eliminating the majority of their local service to the central Puget sound area by the end of the 1980s. The 574 was one of ST’s initial “launch routes”, It dident really become popular until after the TSA was formed, and it became a commuter route to the airport. This role reinforced after the 194 was eliminated a couple years ago when RR “A” was formed.

      2. Combining the 574/577 and adding a airport stop on the 578 might not be a totally devistating thing either, although from the tail ends (Lakewood and Puyallup) it would be a very long ride to seattle.

      3. Ten years? It seems to me it was more like two or three.

   2. The biggest selling point of making the 512 the all-day route should be the drastic increase in frequency it enables. If we wanted to, we could redirect the 510/511 service hours to run the 512 every 10 minutes on weekdays, 15 minutes on Saturdays, and 15-30 minutes in the evenings.

      For anyone going to Lynnwood, the frequency comes at no cost, as travel times would be the same. And for those going to Everett, the frequency would compensate for the extra travel time. And in the evenings, it would be a huge improvement. As it stands today, service from downtown to Everett is every 30 minutes on Sunday nights, but only every 60 minutes on Saturday nights. The extra 30 minutes of wait time on Saturdays is worth far more than the extra 5-10 minutes or so of travel time via the 512 on Sundays.

      1. The extra connectivity provided by the 512 (which, honestly, would take about ten minutes longer than the 510, not two minutes longer, once you count the time to pull off at Lynnwood and Alderwood) ought to enable freeing up of CT resources by truncating some of the far north commuter routes at South Everett Freeway Station or somewhere else along the way, allowing them to go into local service instead of sitting down by the Rainier Brewery all day.

         This ought to be an easy sell among Snohomish members of the ST Board.

         Be careful with the math though. You can’t just convert four hourly runs of the 511 and two hourly runs of the 510 into six hourly runs of the 512. Due to the extension from Lynnwood to Everett, that would be a substantial increase in service hours. Simply converting the four existing hourly runs of the 511 into 512s and eliminating the two 510 runs, would save service hours, but then the question of capacity would come into play.

         I know, I know, every route that has a champion here is “crushloaded”. The times I’ve ridden 51x’s, they were never even close.

      2. Looking at the schedule, the 512 seems to take about 5-6 minutes longer than the 510 at all times of day. Most of the schedule variability is around downtown and the U-District, and by the time you hit 145th, it’s very consistent.

         FWIW, I don’t think anyone claims the 38 or 42 are crushloaded. ;) But I agree with you; in my (limited) experience, the 510/511 have plenty of space at most times of day.

      3. 15-minute or 10-minute frequency full time to Everett would make a significant difference in how people view transit. People would start seeing transit as viable for trips they wouldn’t before. I’d rather wait 15 minutes than 30 minutes for going to either Lynnwood or Everett, even if it has a few more stops. And when I’m waiting for a bus in Lynnwood, I’d rather not spend the last 15 minutes thinking about the Everett-Seattle bus that bypassed me.

      4. “The times I’ve ridden 51x’s, they were never even close.”

         I don’t know when you guys are riding but I use the 511 a LOT and often have to stand. Maybe you were doing reverse commute?

         My mother who rides the 511 4 times a week has gotten in the habit of boarding it at the ID stop so she can sit down. By Westlake it’s standing room only.

      5. Worry not, Grant. We’re not trying to overload your peak-hour 511. But don’t you want to be able to travel more places quickly than just downtown Seattle?

5. I’d like to propose a modest fare increase: All multi-county ST buses should charge the multi-county fare for everyone who rides. (The 574 could be an exception.)

   Given the revelation that zone fumbling on the ORCA readers has become a block to installing backdoor readers on all the buses, ST could demonstrate the trivial solution, so the major speed improvement of all-door boarding can move forward.

   I know, distance-based fares would be totally cool, but ST and Metro are nowhere near that level of technological finesse.

   1. The 510 and 511 are extremely useful routes to get between downtown and the U-district, especially on evenings and Sundays when the 71/72/73 buses take forever to make that trip. Charging the multi-county fares for such trips seems a bit expensive.

      However, I think one could argue that using a 510/511/512 bus for those trips is, in a since, a premium service, justifying a higher price.

   2. All distance based fares require is a tap-on tap-off system. That’s why Link can do it. RapidRide would be an easy target for this.

      1. Distance-based fares could be done, but it is not a priority (or even a goal) for any agency right now. They’ve got higher emergencies.

         But consider from the point-of-view of a non-nerdy rider who doesn’t want to spend too much time thinking about the particulars of their twice-daily bus ride. They’ll want to know ahead of time how much the ride will cost. The GPS will have to be really accurate, and tied to both ORCA readers. The schedule will need to show two costs for each stop pairing (one off-peak, one peak) for Metro buses, and that will be many times more stops than those shown on the Link chart. They’ll also need a plan for routes that overlap the boundary of peak hours — one that will meet with passenger acceptance.

         Plus, it isn’t automatic that distance-based pricing doesn’t create perverse incentives that raise overall operating costs. If distance-based pricing causes express routes to end up costing less than local routes (due to the straighter line), that could be a problem.

         If distance-based pricing is based on as-the-crow-flies mileage, then savvy riders will ride too far in order to take advantage of a cutback in the route, then get off and ride backward to where they really wanted to get to. If distance is based on this principle, we’ll have people riding from Microsoft to UW on East Link, transferring, and heading back downtown, further crushloading the bottleneck in the line (which, BTW, Aleks, is one of the reasons why we wouldn’t force transfers onto Link from a whole fleet of buses just outside downtown).

         As I’ve pointed out, distance-based fares have already created a perverse incentive to ride the 577 to Federal Way instead of taking Link+A or Link+574.

6. Sound Transit’s three new locomotives should be arriving on the property sometime next year.

   1. Bah.. 2013 I meant.. MP40PH-3C’s (4000HP units)

7. The 574 is an important direct link to the airport for Pierce County travellers & commuters, really the only link. I’ve taken it many times at 1,23am and its standing room only. Perhaps they should just make it more direct.

   1. What boggles my mind is that ST provides the primary service for a pair of Metro-owned park&rides, which exist to provide free parking so airport employees don’t have to pay to park at the world’s second largest parking lot.

      Can’t the airport make some sort of arrangement for employees?

      1. Those P&Rs were built for downtown commuters, not airport commuters, silly. If they have become majority airport commuters, that’s interesting. It suggests that former 194 riders to downtown are not using the 574?

      2. It would hopefully be a simple exercise getting the most recent on/off counts, by route, at Star Lake and Kent/Des Moines.

8. That 592 no longer serving the busway will be a mess, as yes there are several people getting on and off at affected stops. I – as a 592 rider – do not support removing the 592 from the busway

   For those wondering why not all Sounder trips serve S Tacoma and Lakewood: it’s because of the single track only that is the current setup.

   ST says they expect some of us DuPont riders (and I am one who would do this) have the 592 bring us to Lakewood and divert to Sounder.

   As for the 574 change: there are pluses and minuses. There’s a lady who boards almost every day at Tukwila, gets off at the Dome, takes the 574 to Lakewood Mall, then transfers to the 2 to head home. On the other hand, airport employees need a place to park at 2 AM when the first 574s start rolling northbound. Yes, some use Lakewood Mall (the mall doesn’t want that), so ST wants to get them to park at Lakewood Station.

9. Pingback: ST's 2012 Service Implementation Plan – Seattle Transit Blog | Service Meaning and Definition

10. On elimination of the AM Sounder north/south transfer: Are you sure about that?

    I read that the south train leaves Seattle no earlier than 6:42, and the north train arrives by 6:44. Wouldn’t ST just hold the south train a couple minutes for the north passengers to transfer?

    1. Apparently not. I’m guessing not enough people actually make that transfer for them to care. Also, If the sounder train arrives 15-20 minutes late, holding the south train could cause it to arrive in Tacoma late, making it late for the peak-direction trip northbound it makes next.

    2. Would you trust a daily -2 minute transfer? If you’re going to offer the transfer, the schedule should reflect that it’s possible!! That seems blindingly obvious to me. IMHO, they should keep the 2nd AM southbound at 6:50 and make the others clock-face. A non-trivial number of people are making that transfer everyday, my guess would be 20 or so, but I’d love to have data to back it up.

    3. So right now the only transfer option be Amtrak Cascades – 7:30 A from Seattle to Tukwila to Tacoma. I wonder when that will begin, hopefully AFTER the McChord AFB Air Expo in July of 2012.

       Firmly against the idea of taking away the Sounder transfer, it will cost me $25 instead of $5 to get to Tacoma, which means $20 less to spend in local businesses. Also takes away options for commuters in difficult times.

    4. Very little thought has been put into Sounder transfers. There’s one northbound in the afternoon where you can go all the way from Tacoma to Mukilteo (and Everett of course) but since CT pays no attention to the Sounder you get to Mukilteo with about 3 minutes less than you need to catch the 113 up the hill thus making you wait for an hour for the next one.

11. Two points I don’t really understand:

    1) Why is it necessary to have so many extra buses running nonstop between Tacoma and Seattle during the same period in which the Sounder train is making the same trip in the same amount of time? It would seem we could save a lot of money by telling people to take the train during the peak period and use the buses for off-peak when the train isn’t running. The only possible reason I can see is if the trains are overcrowded as it is and buses are the only way to provide additional capacity. But, as I don’t live in Tacoma, I don’t know if this is actually the case.

    2) Can the 594 and 577 be combined into one route off-peak, similar to what the 512 does in the I-5 north corridor. Is it really worth having service every 30 minutes vs. every 15 minutes to get to Tacoma 7’ish minutes faster?

    1. 1) Wait+travel time ca. 75 minutes by train, 55 minutes by bus, and the train costs $1.75 more (and even more than that from Lakewood).

       2) Read my math on a thread above about the 512. We can’t simply convert 577s into 594s. The required service hours would increase. However, converting two 577s and two 594s into three 594s yields a small savings, if the capacity is sufficient.

       As for what to do with the 578, I’d suggest that it be altered to serve Kent Station, head up to Kent/DesMoines P&R, and then to the airport, and add a stop at Kent/DesMoines P&R to the 594.

       The 574 ought to just be an express between the airport, Tacoma, and maybe Lakewood and DuPont.

       Something tells me, though, that this paradigm shift would run afoul of subarea equity strictures.

       1. Run that 574 all day to DuPont and we got a deal ;) (yeah yeah I know, I wish)

       2. If taking the train from Lakewood takes longer than the bus, why are we paying for the train in the first place?

       3. Because voters insisted on it as a condition for passing ST1 and ST2. Because it gets so many busloads off the freeways. Because trains don’t get stuck in traffic. Because it can be expanded at less than the cost of setting it up in the first place.

    2. Ivwas on an early morning 591 (route no longer exists, use the 590 and 594 today) out of the 512 Park & Ride and there was a HUGE line when we pulled into Tacoma Dome. This is in addition to the crowds on Sounder.

       There isn’t enough room on Sounder for all the bus passengers as both can be crush loaded

       1. “I was” (darn mobile keyboard)

       2. Yow.

          Is ST negotiating with BNSF to get extra slots for trains? If you have full crush loads on bus *and* train….

       3. … and maybe rework the price structure to reward those passengers willing to take Link+574.

       4. Nathanael, ST has already negotiated four additional train easments for South Sounder. These were included in ST2, but won’t be put into service for a little while. See page 101 of the SIP.

       5. Thanks for the info, aw.

12. How is one supposed to get from Tacoma to Puyallup on a weekend, as route 400 doesn’t run during those times? Is there no alternative other than backtracking all the way to Federal Way via 574->578 (uggh).

    1. The 578 doesn’t run from Tacoma to Puyallup on weekends, IIRC. One could take PT 41 or 42 to the 409 if that still runs weekends.

Comments are closed.