KC Metro route 49 bus / photo by Paul Kimo McGregor

Sound Transit recently released its proposed 2021 Service Plan, in which it prepared for the pandemic to continue through the duration of 2021, by continuing the suspension of ST Express routes 541, 544, and 567 indefinitely, continuing to have pared-down service on most other routes, and making 15-minute off-peak headway on Link Light Rail the plan for the foreseeable future.

There are a couple of categories of service savings, related to Link connections that have not been fully utilized by the collective transit agencies.

The first category of savings comes from the extra space freed up at Northgate Station’s bus transfer facility during the weekday peak periods, for the duration of the cuts. King County Metro routes 67 and 75 are about to get their peak frequency downgraded. Other routes are getting surgical cuts. While some of the routes and numbers will change with the September 2021 North Seattle bus route restructure, the reduced level of service probably will not, unless the pandemic wraps up and Metro manages to climb out of the budgetary hole. In particular, Metro might free up even more bay space by pre-mothballing proposed route 68 (Northgate to Lower Queen Anne express).

The freed-up bus bay space could conceivably be used to have more of the Community Transit commuter trips terminate at Northgate. CT wouldn’t have to change route paths, just the number of runs on 800-series routes vs. 400-series routes. However, it would mean using more of their artics and fewer of their double-talls. While the number of additional trips moved to Northgate beyond the current plan may be small, each trip so moved is a non-trivial savings.

The second category of potential savings is reducing frequency on all-day Link feeder buses, to match Link’s schedule. Unlike the first category, these savings do not have to wait until Northgate Link opens.

King County Metro has several Link feeder buses that run more often than every 15 minutes during the weekday mid-day off-peak period, including:

In addition to about-to-be-synced routes 48 and 49, more of these routes could be pulled back to 15-minute headway, and timed to Link’s schedule. They might get better reliability and a more even spreading out of passenger loads as a result. Some of them are using 40-foot buses, such as route 36. There ought to be more 60-foot buses available now to upgrade capacity in order to synchronize better with Link.

One smaller opportunity for savings may be to splice together routes 45 and 48 again, if it does not result in capacity issues.

The sad reality is that these savings, if taken, will probably not be used to save other service, but rather to reduce the mounting deficit in the various agencies’ operating budgets. They will also have spin-off effects as agencies can further downsize maintenance and other support staff, making the ramp-up to post-pandemic service levels more difficult, or at least lengthy.

24 Replies to “Getting more savings from Link feeder buses”

  1. The logistics (and time) of getting in and out of Link stations makes timed bus service dIfficult. It can take up to 3 or 4 minutes to get between leaving one vehicle and waiting for the next. Some people can be pretty fast (like run up escalators) while others need time. Neither the bus nor the train driver can determine if riders have left the other vehicle. At best, a bus that has a layover of at least 5-9 minutes and is perfectly timed to the exact minute can get riders both to and from a connecting train but that’s challenging to implement.

    The right solution is really to have Link trains every 10 minutes all day. Even 12 minutes would be better. That way, the transfer issue is not particularly critical.

    1. The additional point is that scheduling a layover still requires paying a driver (the majority expense when operating a route). If a bus is sitting at a stop, the driver still gets paid — so it’s better to just have more frequent buses that are always moving rather than to add stationary time and schedule less frequent buses.

      I would suggest that exceptions are when a driver needs a break or when a system is so infrequent (say at least a 20-30 minute spacing between both buses and trains) that a timed bus layover makes sense.

    2. AJ, is there no chance that drivers can be trained, equipped, and ordered to communicate with each other regarding connections? And every major transfer point staffed by a supervisor to coordinate operations?

      Wills and Ways, Wills and Ways. No? But since I flunked Numerology, I need something explained. Between ten, twelve, and fifteen minutes, in both operations and passenger experience, what’s the difference? Many thanks.

      Mark Dublin

      1. Even if drivers could talk to each other at transfer points, it would still be a problem to know how many people getting off one vehicle want to transfer to another one.

        The only viable way I see on how to do that would be for a rider to “reserve “ the transfer on a phone app. That could easily be overkill.

  2. I can’t believe you’re suggesting reducing the A, 7, 8, 36, 44, 45, 48, and 49 to time them with Link. Probably 90% of the trips on those routes are not to Link but to other stops, or to destinations that happen to be near the Link stations. The 7 especially has a lot of people riding it from one part of the valley or another or Jackson Street, and gets overcrowded at the current frequency.

      1. I agree. Both of you beat me to it. None of the buses listed are feeder buses.

        There really aren’t many feeder buses right now. The 635 is one of the few (it is even called a shuttle). I suppose the 107 could be called a feeder bus, but only for a section (on Beacon Avenue). Neither bus runs very frequently. For the most part, the buses that go by stations would follow the same route, even if there wasn’t a station. At most they are making small detours to connect to Link (e. g. the 50). Even then, that bus also connects to the 101 and 150, so if Link is running less often, those that transfer can still get downtown. The same could be said for the 635, really. It is a feeder bus, but it feeds to both the A Line and Link. The 345/346/347/348 could be considered feeder buses, but they feed the 41. In other words, there is more bus to bus feeder buses, than bus to train feeders.

        That will change with Northgate Link and East Link. At that point, both Northgate and Roosevelt Stations are “fed” by lots of buses running from I-5, SR 522 and surrounding neighborhoods. Mercer Island will have ridership almost entirely dependent on feeder buses (from I-90).

        But buses like the 7, 8 or 44 operate largely independent of Link.

      2. The 107 is also a feeder from Rainier View and Renton West Hill. It took over the 42’s tail and an earlier 107 to downtown.

      3. The 107 is also a feeder from Rainier View and Renton West Hill.

        Yeah, but again, it feeds bus routes as much as Link. The 107 has this bus stop here: https://goo.gl/maps/HQQ9qMsy5qPgzn4HA. That allows for a same stop transfer to either the 106 or 7 (both of which go downtown).

        As I see it, there are several set of 107 riders:

        1) Those who transfer to Link heading downtown.
        2) Those who transfer to the 7 or 106 (whichever comes first) to go downtown.
        3) Those who transfer to Link for its Rainier Valley stops.
        4) Those who transfer to the 7 or 106 for its (different) Rainier Valley stops.
        5) Those who take a one seat ride (e. g. to Beacon Hill).
        6) Those who take it to SeaTac.

        My guess is that first and third set is relatively tiny, and may gravitate to the second when Link is infrequent. If you tried and time the train for those riders, you likely hurt every other set of riders (including those that ride Link the other direction).

      4. All of them are tiny. My friend who lives on it doesn’t use it because it’s so infrequent and short (north of Rainier View) and slow (from Rainier View to Renton). I always thought of it as a Link feeder. Probably nobody takes it to the 7 or 106 because they live in large-lot houses with cars, and they’re not going to take one bus a mile and transfer to another bus for another couple miles and it’s not an urban center.

  3. If connecting to Link is part of truncating buses, then Link needs to be frequent. There are rail lines all over the world operating at 3-5 minute headways. 10 minutes is the bare minimum to be considered frequent. With all of the capital that we have invested in Link, and the fixed costs of operating the tunnel stations, providing security, etc. It’s ludicrous that ST is saving a few dimes in operating costs by running at 15/30 minute headways and making these transfers more time consuming, even more ludicrous to keep doing it after Northgate opens.

    We and our leaders should demand that ST puts Link back at 10 minute headways 7 days/week from opening until 9pm, and 15 minute headways afterward.

    1. Thirty minute Link headways are acceptable under one condition.

      That the top official in any chain of command that decrees them, is required by law to ride them. To a job where tardiness is not only grounds for but guarantee of termination.

      Sauce for the Goose, sauce for the flea-bitten overstuffed Turkey.

      Mark Dublin

    2. Actually, per zoning and parking, the city of Seattle does consider 15 min headways to be “frequent”. That is the criteria they use for setting parking minimums on new developments.

      That was the issue with the old Kort Haus redevelopment on Greenwood Ave. The developer claimed that since Metro was scheduled for 15 mins or better they didn’t need to provide much parking since frequent service was being provided, but the anti-development neighborhood activists claimed that since Metro was incapable of actually meeting their promise, therefore the developer had to provide parking.

      But the 15 min frequencies on Link are just temporary. Once things start to recover from CV-19 the frequencies will improve.

      1. Maybe 15 is frequent for a bus service. It is not frequent for a transit rail line in any city in the world.

    3. “If connecting to Link is part of truncating buses, then Link needs to be frequent.”

      We don’t have the power to do that if ST doesn’t want to. Metro is easier to convince than ST. We’ve been asking ST for two months to restore 15 minute service and it’s only doing it now, and now it’s planning to keep 15-minute service through the next three service changes. If ST won’t budge we have to work around it until we can get a board who puts riders first.

      1. Then let’s get a board that supports service.

        It is insane to spend billions on capital and millions on maintaining stations and elevators and escalators and then dribble the service to save a few hundred thousand.

  4. The first category of savings comes from the extra space freed up at Northgate Station’s bus transfer facility during the weekday peak periods, for the duration of the cuts. … The freed-up bus bay space could conceivably be used to have more of the Community Transit commuter trips terminate at Northgate.

    Is that really what drove the decision to keep the express routes to downtown? That seems presumptuous, especially since Metro hasn’t decided how many Metro buses are going to use the transit center. Has there been any thought to how many buses can use a bus bay? What effect would off-board payment have on that number (which will certainly be considered, if not adopted)?

    This sounds like yet another rumor that is spread by folks thinking of possible issues (similar to the deleted Montlake vent shaft, an issue that was repeatedly raised in the comments, but turned out to be not an issue at all).

  5. Al S., to be considered “qualified” on a route, drivers should know their every connection from memory. Same for coordinators. And Might be worth the wages to hire, where necessary, “coordination specialists” in addition to present staff.

    But the main requirement for making transfers work is for the whole fleet to run as close to schedule as possible. Meaning that money spent on reserved lanes and prioritized signals might save a lot more than a tight budget will deny.

    Same for general instruction and training for all operating personnel. And most important of all, it’s also equally incumbent on both the Company and the Union to be sure that anyone who’d really rather be a private contractor be given contact info for “Personnel” at Uber and Lyfft.

    Mark Dublin

  6. The 2021 plan may be off-track. The virus has its own calendar; no one, not even ST, can know it; we do not know when an effective vaccine will be distributed. So, transit ridership may rebound more quickly than they apparently expect. ST should be nimble and ready to improve service quickly. ST is a capital intensive agency. They tend to be overly cheap with regards to service. Their partner agencies, ET, CT, Metro, and PT, are service agencies. The recession is causing them to reset their capital program – fine. But that does not mean they should reduce service. In fact, the network would be much better off if ST restructured its network and increased service (or provided shorter headway) at off-peak times. Of course, ST has weak routes that could be deleted, but their buses and hours should be shifted to ST trunk routes 512, 522, 545, 554, 574, and 594, so that waiting is reduced. ST has buses sitting in the bases at off-peak times as it has peak only routes 513, 532, 541, 544, 567, 590, 592, and 595. So, they have buses to run the trunk routes more often. The service budget is millions; the capital budget is billions; how much would the capital delivery be delayed if ST spent more on service today to run Link and its trunk routes more often at off-peak times? I suspect not much at all. Ridership attracted today is worth more in cost benefit terms than ridership attracted years from now. Just as it makes sense for the partner agencies to restructure bus networks to meet Link, so should ST bus routes be similarly restructured; it is past time that Route 545 serve the UW Stadium Link station and not run on the I-5 general purpose lanes; likewise, with Link serving Roosevelt and Northgate, routes 542, 555, and 556 can be truncated. For both Covid physical distancing today and ridership attraction tomorrow, ST should provide short headway service. Why did we spend billions on Link or millions on the bus network if not to run it frequently and attract riders?

    1. No more bus truncations until Link has 10 minute service from start of service until 9pm and 15 minute service rest of the evening – 7 days/week. None of this 15/30. Adding that transfer to the final 10 minutes of a bus ride can double the travel time.

  7. 100% right, eddiew. But somebody please confirm or deny: The people drawing up these plans…at some level, have they really made up their minds that failure is easier than trying to make transit work?

    Mark Dublin

Comments are closed.