135 Replies to “Bringing Frequent Service to South King County”

1. Twenty years ago, Route 107 came out of the Tunnel, and ran down the south end of Lake Washington to Renton, via the neighborhood at the west end of Renton. I always thought that at least at rush hour, the stretch along the lake would make an excellent route- maybe because I enjoyed the scenery so much.

   Maybe this route would not have to go downtown- it might better just terminate at the Rainier Beach LINK station.

   Also wondered about continuing the Route 7 wire from 62nd and Prentice a block up to Renton Avenue, and at least over to Skyway, if not all the way down to Renton. I also thought the 7 would be a good Tunnel route, and indeed Metro considered doing this, with a ramp off the ramp from I-9o down to Dearborn Street at Rainier Avenue.

   Would have been a route I would have driven to the end of my career. But then again, always wished new Tunnel coaches could have kept their wire down there. Proving that the American Psychiatric Association made a mistake saying there’s no such thing as Asperger’s Syndrome.

   Queen Anne Counterbalance used to be another good example long ago.

   Mark

2. Though I don’t live in Washington, I have an ORCA card that I use when I visit Seattle. However, there seem to be a fair number of riders south of Seattle that still pay cash and use transfers to get between bus routes, which under the current fare scheme would discourage them from transferring to LINK.

   A fair number of places that have light rail treat transfers the same, weather they are on light rail or a connecting bus. Would the residents along these routes be more encouraged to take LINK if such a policy was in place?

   While one-seat rides are nice, really good planned and executed timed connection between routes can make a huge impact on weather people would support having their route broken up or not. The Skagit Transit -> Island Transit connection between Mount Vernon -> March’s Point -> Anacortes to get to the San Juan Island Ferry Terminal works so well that while a one-seat ride would be nice, it really isn’t that big an issue.

   1. There are multiple reasons the fare system encourages cash payment on Metro buses, but that is a topic for another day. Suffice it to say that the fare system ought not drive the routing, unless the agencies are actually trying to incentivize routing choice (which I don’t think they are — another pet peeve for another day).

      1. Agreed.

         Pie in the sky idea. What if Metro/ST encouraged transfers by give people and extra hour if they do transfer between bus and Link? Do people think that changes anything or does it just complicate things?

      2. Adam, assuming we can’t get rid of cash transfers entirely, I would *love* to see longer transfers with ORCA for either bus-to-bus or bus-to-Link.

         Currently, the system is set up so that you will never get a shorter transfer by paying with cash than you would with ORCA. I want to see that the other way around. Give 3 hours with ORCA, and instruct drivers to give cash transfers for no more than 2 hours from the route’s end terminal.

      3. Come on, guys, just agree that everything should be done so that everybody gets an ORCA card ASAP, including kicking bales of them out the doors of C-130’s from McChord. Cash is a pain in the rear bumper all the way around, delaying service and costing a fortune to count and secure.

         And anything that makes it easier to use service is good especially between 107 and LINK. Like George Orwell would have had the sheep say: “ORCA good, cash bad, delays (whatever 1940’s British slang was for “sucks!”

         Mark

      4. In the coming months, if the low-income Metro fare survives the April 22 election, ST will have to make a decision on whether and how it will honor the low-income ORCA. I’d assume ST would set the low-income fares the same as youth fares, but I’m not counting on anything. If ST were to not honor the low-income fare (unlikely, I am projecting), the Link transfer plan would be a much tougher sell.

      5. If ORCA cards gave you a longer transfer time, it would only be fair. My experience with paper transfers is that if you’re polite and smile and show the driver a transfer that’s the right color for that day, the driver is happy. S/he doesn’t care if it expired an hour or so ago. But not ORCA. It’ll change another fare simply because the clock tells it to. Your smile and cheerful greeting to the driver are no help at all. Bummer.

      6. A little something to consider for that future discussion about these changes:

         It wasn’t exactly cheap to accomplish, but here in TriMet land they completely eliminated the newsprint style transfers. Instead, they now use the same paper stock as they use in the MAX ticket machines, and the date is printed on the ticket, which also has a hologram on it to discourage foraging of the ticket:

         http://www.oregonlive.com/commuting/index.ssf/2013/04/trimets_new_ticket_printers_ma.html

         When I first saw that this was happening, my first thought was that this was going result in system apocalypse because of the time required for a ticket to print vs the time required to tear off a transfer from a metal bar. They’ve done this system wide since August 1st, and in reality it isn’t anywhere near as bad as I thought it would be. The printers are actually extremely fast, and so by the time the first bill gets sucked into the farebox, the printer has already printed out the desired ticket. So, in reality, it doesn’t really take any longer since the time limiting constraint remains the passenger putting the cash into the farebox.

         The bad news, of course, is the cost of buying ticket printers for each bus in a system the size of King County Metro.

   2. Tri-Met used to have ticketed printers back in 1984. Didn’t last long.

3. It would be part of my transit dream come true to have the 169 become both more frequent AND terminate at a LINK station! (And also extend its hours to 2am).

   This would give me all week access with a nearby, single bus, to both Seattle, and SeaTac!

   I would call it an instant hit for many our riders here on East Hill.

   1. Oh, and I just noticed what you did to the 164 and 168!

      Yes, fantastic. Direct to SeaTac and replacing the pain of changing buses for the 150 with a seamless route!

      Getting to Southcenter from here still seems like a chore though, which is really odd for such a singular and important destination. But perhaps that is as much their fault as anyone’s.

      1. If Southcenter had been arranged in a way that had considered transit or pedestrian access in the slightest, it would be much easier to design transit to serve it well. As it is, Southcenter is pretty much impossible to serve in a satisfying way.

      2. Improving the 169 from 30-minute headway to 20-minute headway would theoretically reduce the travel time between East Hill and Southcenter by 5 minutes when connecting via the F Line, or 10 minutes when connecting via the 164/180, at least at mid-day.

      3. Why don’t you consolidate a lot of the Kent Valley expresses by making a Kent Station to Angle Lake express?

         If you can get people quickly on LINK, any time of day, it removes the problems of an infrequently running Sounder. It gets people to SeaTac and Seattle.

         Again, Southcenter is problematic. All they would need really is a gondola from Tukwila Station right to inside the food court!

      4. John, could you list the routes to which you are referring?

      5. Are pigs flying? Is it 72 degrees in Seattle on New Year’s Day? Are the Bailoman and the urbanists actually agreeing on something?

         A Kent-Link express would be ideal, but given that Kent is so far south from Link’s terminus and Metro doesn’t have spare service hours, it will probably have to wait until Link reaches Kent-Des Moines Road. The 180 is already going about as fast as an express could go, and it takes 20 minutes. But a route on KDM Road to KDM Station would take much less time.

         As for it going to Angle Lake Station, people wiser than me think there’s little advantage over SeaTac Station given the lack of direct roads from the east (and the large number of riders going to the airport anyway).

   2. They seemed to have really pushed out the schedule on the 180 since I last rode it to the airport.
      It used to give out at like 6pm on Saturday and 4pm on Sundays!
      Now it’s 6am to 2am on weekdays and 3am to midnight on weekends.
      That’s usable (except the local routes from Kent Station give out much earlier).

      Long term I’d agree with KDM road vs Angle, but short term, just run it to Angle.

      That would really seem to take care of all Seattle-Kent express needs.

      You’ve got Sounder for the heavy rush hours.

      And you’d have LINK to Kent Station for all other times of day and weekends.

      Could be done as soon as Angle opens up.

4. Overall this is pretty great — it seems to be much better than what’s currently there (though I admit I rarely go to this part of the county).

   My primary piece of feedback is 20 min frequencies are awkward. That’s too long to just show up at a station for most people.

   Is there a way to get the frequencies down to 15 minutes for some of those routes by eliminating peak only routes that are duplicative (e.g. offering 1-seat rides to Seattle).

   Another thing, wouldn’t it be better to have the 184 continue south of Federal Way rather than the 181? That way riders there get a 1-seat ride to Link, rather than a 2-seat ride (since the 184 connects to link but 181 doesn’t).

   1. These 20-minute headway routes are (1) designed to interline for 10-minute headway to key destinations (Renton TC, Southcenter, and Kent TC); and (2) would not mesh well with Link schedules at 15-minute headway. I am assuming the times are mid-day headway, and that any of the Link-connecting routes could be adjusted to be more frequent during peak.

   2. Every word of Brent’s comment is correct.

      I started out with 15-minute frequencies, but it turns out that expected wait times are significantly shorter with 20-minute service and timed connection. If you take Link to Auburn, you will either wait 0 minutes or 10 minutes, due to the timed connection therefore, your expected wait time is 5 minutes. But if the bus came every 15 minutes, your expected wait time would be 7.5 minutes.

      1. Aleks… that doesn’t sound right. 15 minute frequencies can still have a some timed connections. Assume that every third Link connects to a bus — so a timed connection every 30 minutes — and the other half of the buses are scheduled to provide an overall 15 minute headway. Then you have a 1/3 chance of no wait, a 1/3 chance of a 5 minute wait, and a 1/3 chance of a 10 minute wait. Expected wait time should still be 5 minutes.

      2. I think you’re leaving out the “timed” part. In the 20 minute case, the bus will always hold for the train, even if the train is running late. In the 15 minute case, it’s possible to miss the bus that leaves a minute before your train arrives.

         Also, upgrading from 20 to 15 minute frequency costs a lot. (This network is affordable using the existing service hours only because of 20 minute frequency.) Why upgrade service if it would lead to the same effective wait time? Let’s save those service hours for places where we’ll improve things for most riders, rather than just staying the same.

   3. The 181 wasn’t really part of my restructure. That is, I didn’t seriously look at it, and I’m not really making a recommendation about how it should work. Oran added it to the map mostly for context.

   4. My primary piece of feedback is 20 min frequencies are awkward. That’s too long to just show up at a station for most people.

      20 minute headways aren’t too bad, so long as they are consistent. 15 minute timetables aren’t too much different if you are waiting out in the rain either – it is far better to check the timetable and show up at the stop when you need to. Memory based timetables are used by a number of transit agencies overseas, and with 20 minute timetables the worst you wind up with something along the lines of 7 minutes past the hour, 27 minutes past the hour, and 47 minutes past the hour. What throws a wrench in the works is when someone muddles with the consistency and you get something like a 7 minute headway sometimes, and 18 minutes some other time, and maybe a 38 minute headway some other time.

      The few times that I have been into that part of Seattle, the lack of timed transfers proved terrible. This looks like it would be a huge improvement.

   5. Brent and I and Aleks were discussing this 20 vs 15 minutes issue when she was writing this article. 15 minutes is the outer boundary of what people consider “frequent” and “convenient”, and the 150 already has it weekdays and Saturdays. But timed trasfers to Link creates another ballgame. It effectively “extends” Link. And as Link’s north and east extensions get built out, timed transfers to it will gain an even larger market share and new riders.

      Aleks’ proposal actually is 10 minutes to Kent Station and Renton TC because two routes are paired on each of them. So every train would have a bus to/from the transit centers. It’s only going beyond them that the half-frequency occurs.

      1. The idea of timed connections from Link sounds wonderful. Unfortunately, though, I have to have a certain amount of skepticism as to whether the promise that buses would hold for a couple minutes if the train is late would actually be kept.

      2. In large part, my post is trying to illustrate how a network based on timed connections to a very frequent, high-capacity trunk service can lead to better service for less money. My hope is that if enough people understand the need for timed connections, it will make it easier for Metro to propose a change that uses them more heavily, and harder for them to justify keeping the current network that doesn’t provide their advantages.

      3. The 116 currently has timed connections for the Vashon ferry with specific instructions on the run card to wait for transferring passengers. It’s relatively trivial to do so since you drive the bus out onto the dock and turn it around to wait for the passengers, who are the first ones off of the boat. Because of this, the 116 is a rare (only?) example of a route that is still pay as you leave. Everybody boards the bus and then pays as they leave. In practice, it gets confusing as people further along the 116 route board – at which point it turns into PAYWYL: Pay whenever you like

         Assuming you can solve the when do you pay problem you’ll likely need capital improvements to the area around Ranier Beach Station to make transfers as smooth as possible and aid drivers in knowing when they have their passengers.

         Great ideas… We need more of this.

      4. Another thing that is important for timed connections with every other train to work is that the posted schedule needs to clearly indicate what time the train that connects with your bus leaves each downtown station – which, in turn, means Sound Transit needs to start posting schedules for Link. A model where it’s a coin flip whether you have a timed connection or a 10-minute wait at Ranier beach is probably not going to fly.

         That said, I really like the idea of timed connections, and I would like to see it done elsewhere. For instance, The 221, 226, and 240 should all have timed connections with the 554 at Eastgate. Instead, all three of these routes have awkwardly long waits. Similarly, the 208 and 927 should each have a timed connection with every other 554 at Issaquah. In reality, neither does. Sometimes, it feels like Metro is intentionally scheduling connections with longer-than-necessary waits with the trunk route they are supposed to be connecting to, as a way to reduce ridership, in order to make it easier to cut the coverage route when funding gets tight.

      5. It does have a bus at every train. It’s only if you’re going further than Kent Station or Renton TC that it matters which bus you’re on. If you’re on the “wrong” bus it’s a 10-minute wait at the transit center for the other one. That’s is pretty good compard to the existing connections at those transit centers, which can be anywhere from 5 minutes to 25 minutes, or 45 minutes if you’re catching the 168 when it’s hourly.

      6. In Miami, there is a signal that indicates when a train comes in. That might be helpful for places like Tukwila International Blvd Station and the future Northgate Station.

5. Great work, Aleks. Even as you argue that suburbs are impossible to serve with frequent grids, I like how much more gridlike this arrangement is than today’s.

   One thing that was not in the scope of your post but that would make this whole network work better, and which really needs to happen, is off-peak express bus service between Kent and Seattle. Right now, the Kent-Seattle connection is tortuously slow whenever Sounder isn’t running, and there’s nothing changing Metro’s network can do about that.

   The frustrating part is that Sound Transit could add this service with an extremely minimal number of new hours, although there would be some ancillary cost. The idea is to reroute the 578 to become a true Sounder shadow (except for Tukwila), replacing the Federal Way stop with a Kent stop. Off-peak service to Federal Way would be provided with a new stop on the 594. The problem is that there is currently insufficient capacity on the 594 to add all of the Federal Way riders. Pierce Transit would have to operate larger buses, something it has been reluctant to do until this point. In the long run, frequency on the 594 might need an upgrade. But it would be worth the small investment to add Seattle-Kent express service, which would be a major improvement to access between the Seattle and South King County universes.

   1. I agree about peak service. Even with Link, time competitive express service is key for commuters serving 2nd tier or employment destinations that are otherwise hard to serve with frequent, all-day routes. I think the loud and clear lesson is any proposal that restructures core route to improve frequency of service and connectivity needs to also look at ways of improve the popular commuter routes.

      1. The current 150 route would be much more efficient (and popular) if it skipped the segment on Interurban Ave between Southcenter and Tukwila P&R. There’s very llittle that is a destination or origin along that road and the 150 would get riders from Kent and Southcenter to downtown much faster by avoiding Interurban Ave.

         Metro could run a local bus every 30 minutes from Southcenter to Rainier Beach Station via Interurban and the Tukwila P&R that would likely be adequate for the number of boardings generated in the corridor. During peak hours there would need to be direct service from the Tukwila P&R to downtown, however.

   2. A couple questions:

      1) What of Sound Transit’s subarea equity rules? It seems this would be more political than administrative.

      2) Doesn’t the MCI have more seats than an articulated bus?

      1. Could you explain in more detail how subarea equity could impact the 578/594 restructure proposal?

      2. We already have plenty of ST routes that serve trips outside the subarea that is paying for them — 522 trips between Lake City and downtown, 574 trips between the airport and Federal Way, 560 trips between the airport and West Seattle. I’m not too worried about the politics.

         The capacity is a different issue. The MCI has 57 seats with no real ability to accommodate standees, and at this point I think PT is operating only MCIs on off-peak 594 and 578 runs. The 594 is close to capacity while the 578 has a lot of extra room, so you don’t need that many more seats. A ST-configured artic has 60 seats with ability to accommodate about another 30 standees, but standing on a route as long as the 594 is obviously sub-optimal. I think the best solution here would be more 77-seat Double Talls.

      3. I suppose a biarticulated would take up too much space in the terminals.
         http://pt.wikipedia.org/wiki/Ficheiro:Expresso_Biarticulado_Curitiba.jpg

      4. Sound Transit still has subarea equity and according to the 2014 Service Plan has South King limited to just over 70,000 hours. In the examples you cite, I have a hunch that there was some kind of political machinations. My point is, unless there is a political discussion on subarea equity, how would South King pay for this? What would you have to give up in South King to pay for a stop on the 594 in Federal Way?

         Secondly, I don’t think there are 60 seats on a Sound Transit articulated bus although you are correct there is more standing room on an articulated bus. While it would be nice to have a double-deck bus on the routes, I believe there is a low bridge at Tacoma Dome.

      5. Glenn in Portland, dual trailers are not allowed in Washington. Given how disfunctional our legislature is, it would be a heavy lift to get them approved, even for a transit vehicle.

   3. It takes over 1 hour and 20 minutes for the 594 under the current configuration to travel from Lakewood Station to Eastlake and Stewart in Seattle. The 594 often fills to capacity even on weekends, leaving no room for those getting on in Federal Way. No other routes except for the 592 (to Olympia) and 595 (to Gig Harbor) are longer, and those routes aren’t fully funded by Sound Transit (Pierce Transit subsidizes part of the 595 and Intercity Transit pays for the 592 extension).

      I find it a huge disservice to those in Pierce County to have an additional 5 minutes in Federal Way, to make up for the reduction in service just so Kent can get it’s express bus.

      Ever been on route 150? Ever notice how half the bus empties at Southcenter? Either they are shopping, or they are looking to transfer to another bus there. These folks have no interest in going to Seattle. So I really don’t see a need to reroute the 578 to Kent. A huge waste of money, in my opinion.

      FYI: I live in Kent.

      1. There are things the 594 could do that would make up the time lost by a Federal Way stop and then some.

         Staying on I-5 all the way to the Secena St. exit and skip SODO is one obvious time-saver. Going from Lakewood straight to Tacoma Dome Station, skipping the slog through downtown Tacoma is another. If done right, I don’t think for Tacoma riders, this needs to be a loss. On the other hand, these same changes could happen without adding the Federal Way stop, so if you look at it that way, it is a loss.

         While the 578/594 restructure looks good on paper, I’m concerned that, in practice, it’s an us-against-them situation, pitting the needs of Kent against the needs of Tacoma, which makes a change like this unlikely to happen.

         It should also be noted that the 510/511/512 restructure, which did happen, has two big differences over the Seattle->Tacoma corridor. First, there are more riders going to Lynnwood than Everett, so fewer people seeing longer trips. Second, even for Everett riders, whatever they lost of travel time, they gained back in frequency. The only 510 didn’t run better than every 30 minutes off-peak, every 60 minutes in the evenings. The 512, by contrast, is every 15 minutes. On the other hand, with the 594 already at 15-minute midday headways on weekdays, it is difficult to justify the increased travel time with additional frequency. (On evenings and weekends, when the 594 is down to every 30 minutes, though, it might be different, especially since many Pierce Transit local routes have extremely poorly-timed connections with the 594).

      2. The problem with skipping Busway for st 594 is that it kills transfers between Tacoma and South King County. It’s a good way to transfer between Tacoma/Lakewood and Kent, and it’s also important for those in Lakewood and Tacoma. As it is, it would still be over an hour end to end, so adding the stop in Federal Way on ST 594 to accomodate Kent is not the best idea. Any express from Kent to Seattle would still have to reach I-5 somehow. Don’t forget that.

      3. There was a reason the busway was created, and I assume it was to keep buses away from I-5 and its entrances near downtown. The freeway is faster there when there’s no traffic, but bottlenecks are a frequent and unpredictable occurrence, so having a dedicated busway seems like a good thing. When I took the 578 a couple months ago, I assumed it would go on the busway but it didn’t, instead it spent ten minutes sitting in traffic around 6th & Airport Way waiting to get on the freeway and I thought, “Why doesn’t it use the busway?!”

6. This all looks good, but the devil is in the details. Some of the actual street routing will need a major re-think, particularly the F in Southcenter, and the 106 the not-depicted 101 in Renton.

   – The 106 and 101 (southbound for this example) should serve MLK, then Renton TC, then along Grady Way and terminate at S. Renton P&R. The route would run basically in reverse northbound, using S. 2nd and a jog onto MLK to get going northbound. (Can we also PLEASE re-assign the bus bays at SRP&R to eliminate the loops??)

   – RR F in Tukwila…I don’t even know where to start, and this isn’t even an operational route yet!

   (Also +1 for adding a 578 stop at Kent Station.)

   1. Based on previous conversations with Aleks, I think the idea was not to serve S Renton P&R with either the 106/105 through route or the 101/169 through route at off-peak hours, although the 101/169 would get within a couple blocks of the P&R on its way to Benson Road. The peak 101 and 102 would continue to use the routing they do today.

      I’ve had a larger restructure of Renton-area peak-only service in mind for awhile which I’ve never proposed, and which is beyond the scope of Aleks’s off-peak work. I would do the following:

      1) Combine the 111 and 114 into one route serving Renton (via S Renton P&R) rather than I-90;
      2) Reroute the 143 to serve S Renton P&R rather than Renton TC;
      3) Keep the 102 as is;
      4) Keep a few peak 101 trips, just between S Renton P&R and downtown, and just enough to keep peak frequency between those two destinations the same as it is today.

      Off-peak, though, I would have the 101 go directly from Renton TC to RBS. Serving S Renton P&R off-peak with Renton’s core downtown route wastes a ton of time and is blocking downtown Renton, which should be redeveloping, from decent transit access.

      1. David is right about my proposal, and also that I’m not making any recommendations about peak service at this time.

         While I don’t really understand the peak network that well, I do completely agree with David that any routes which serve both Renton TC and South Renton P&R should give preferential treatment to Renton TC.

      2. Would making service to South Renton P&R less direct, in exchange for making service to Renton Transit Center more directly, even be politically viable?

         Many would argue that it boils down to a captive-rider vs. choice rider issue. The service between the P&R and downtown has to be direct, or else anyone with a car to leave at the P&R would just drive directly to downtown. On the other hand, so called “captive-riders” would ride the bus anyway, no matter how glacially slow the service is.

      3. Reducing service at P&Rs is always unpopular. But Metro can ascertain whether drivers really use it midday and weekends, and if not, Metro can point that out. Whenever I’m there, there are two or three people at the stop and most of them did not park there; they’re either transferring buses or somebody dropped them off. So moving the all-day stop a couple blocks would not affect these.

         Re mid-day drivers, I’m not sure about Renton P&R but most P&Rs are full by mid-morning so later drivers can’t use them anyway. And if anybody comes back to their car in the early afternoon, they can walk two blocks.

      4. I think it is politically viable. Does Renton want to revitalise its downtown or not? Does it even want to fully utilise its never-full parking garages? RenTC isn’t that much–if at all–out of the way for riders who want all-day and express service. Frankly, kill S Renton park & ride.

   2. adam,

      Sorry if I’m missing something. Is it possible that you haven’t clicked on the link to the full-detail map, which shows my proposed street-level routings for each route? The full map contains much more detail than the schematic.

      The 101 is not depicted because it will no longer exist, at least during off-peak. In large part, that’s what makes this change possible. Or am I missing your point?

      The 106 will skip South Renton P&R. All-day service to South Renton P&R will only be provided by the 107, though there will continue to be service on Grady Way provided by the 169. This was an intentional decision, motivated by three things:

      – It’s really hard to efficiently serve South Renton P&R.
      – Ridership there is actually quite low outside of peak hours.
      – The main purpose of South Renton P&R is as a connection point, but with the new network, that role is no longer needed; all the connections happen at Rainier Beach Station instead.

      You’re correct that I didn’t depict the situation inside Renton TC (or any other TC). I don’t know enough about the operational challenges of these transit centers to make any useful recommendations.

      1. Quite right — I missed the street routing. It looks good to me.

         We’re mostly on the same page, and I was thrown off by the complete elimination of the 101 — I’m definitely okay with retaining it only as a bi-directional peak service.

         South Renton P&R is definitely a challenge to serve, and the current routings get very low marks for efficiency. I like David’s proposal for the 143 as all-day service for the P&R, either with or as an alternative to your 107 proposal.

         (Have a look at the loop for the 101 at S. Renton — that’s entirely caused by the bus bay assignments. Switch sides, and the loop is eliminated for a much simpler in-line deviation.)

      2. As I mentioned this in the article, I’m not proposing any changes to the peak network. The 101 would remain as a peak-only bus. I haven’t shown it in my map because it’s a map of all-day frequent service.

   3. The South Renton P&R issue is another interesting piece in this puzzle. Effectively Renton has two transit centers, because the nominal one doesn’t really serve southeast Renton, and with Metro’s current routing the most common transfer (169+101) is timed for and is shortest at the P&R. Aleks had an earlier draft with an “MLK-TC-northeast” route and a “Renton Ave-P&R-south route”, which would effectively make two transit centers with little overlap. Two transit centers in small Renton? That started me thinking of where the two lines cross at Sunset & Rainier, could we put the transit center there instead? That would require buy-in from Renton and rezoning.

      So what Aleks did in the final was to have both routes use the transit center, and then have the south route go near the P&R but not to it. Because it still serves the same walkshed of city hall, Uwajimaya and surrounding businesses, kiss-n-ride from houses on the hill, and park n riders (most of whom are peak hours). It looks like it would work well as long as there’s good shelters and a load/unload space at the Grady Way stop pair.

7. If they were to ever bury Link through Rainier Valley, I could even see getting rid of the 101 and 150 peak hour too. If Link were able to travel the Valley slog at comparable and consistent speeds to the 101/150, it would be a no brainer.

   1. The extra travel time is more about the 5-minute passage under Beacon Hill than the extra travel time along MLK. Even with burying or elevating the track, I just can’t see how more than a couple minutes could be saved.

      1. When the difference between the 101 (MLK P&R) and Link (Tukwila Station) to University Station is 10 minutes (25 vs 35), those minutes count big time, especially when it would provide reliability and safety improvements.

         It’s similar to all the work WSDOT is doing for Amtrak. People not in the know wonder why we are spending millions to shave minutes off of a 4 hour train ride. What they don’t realize is that the improvements are also reducing the things that cause those four hour train rides to become 6 hour train rides.

      2. Actually, the real time-sink of Link is not the Ranier Valley, but downtown and SODO.

         With regards to Ranier Valley, rebuilding Link to run elevated all the way would NOT save anywhere near 10 minutes. The track distance Mt. Baker station and Ranier Beach Station is just under 4 miles. If you do the math, this would be 7 minutes at 35 mph, or 4 minutes at 60 mph. Given that Link already has signal priority, the actual amount of time that could be saved by running link elevated or underground would be a mere 3 minutes.

         If you rerouted Link to travel straight, as the crow flies, from Tukwila International Blvd. Station to SODO station, maintaining 60 mph the whole way, with no intermediate station stops, then you probably could achieve a 10-minute time savings over the current route. But the costs of doing this would be billions, and all the ridership from the Ranier Valley stations would evaporate.

      3. Little late in the response, but note that I never claimed we can save 10 minutes. I implied that the existing 101 service is 10 minutes faster than Link from a similar location. It’ll be hard to convince people to get rid of their one seat freeway ride and transfer to a light rail that runs along city streets at city street speeds, plus has the occasional mishap that comes from being at grade.

         I have no idea what the time savings would end up being. Probably less than five minutes. But if you read my second paragraph, you’ll see it’s not so much about the time savings, but about reliability. Plus, even 3 minutes on a 35 minute ride is an 8% time savings, if you want to stick with just that stat.

8. Once again, I offer the suggestion that the 101 and 150 should terminate at SODO Station, instead of Rainier Beach. There’s very little additional time needed to run the buses to SODO versus RB and the trip from SODO to University St. is 11 minutes on LINK. The LINK trip from RBS to University St. is ~28 minutes, so making SODO the transfer point could actually save a significant amount of time for riders. Also, if the northbound RapidRide lines were to begin/end their routes at SODO Station, off-peak transferring riders would have a LINK train every 10 minutes or a RR bus every 7.5 minutes to complete the 11 minute trip to downtown. Transferring at RBS only offers a LINK train every 10 minutes (plus the 28 minute trip to downtown).

   1. The choice of Link connection point impacts different riders in different way, and doing it at SODO would yank some of the connectivity improvement of allowing these routes to connect into Rainier Valley. Sure, we could put downtown rides above all other connections all day, and let the buses get stuck on I-5, killing the savings that were meant to improve frequency on the local portions of the route. And if the buses go to SODO, those riding the route will only be the downtown riders, and they will insist it continue on into downtown, as the SODO transfer is dismal, and Metro has no plans to invest in the comfort of that transfer.

      What we’ll end up with, going down that path, is the status quo, with service hours being killed in I-5 gridlock, and the span-of-service and frequency death spiral slowly taking its toll on what is left of the 101.

      1. Well, it’s not that far from SODO Station to Rainier Valley versus Rainier Beach Station to Rainier Valley.

      2. Agreed with Brent. Aleks’ proposal is virtually creating branches of Link into Kent and Renton and significantly improves access between those two suburbs and the Rainier Valley. Sendings buses to SODO would not achieve that.

      3. And, those “branches of Link” are in lieu of real Link branches to Kent and Renton which would cost billions of dollars more and are too controversial to consider.

   2. I’d like to see Aleks calculations for truncating the 101 and 150 at RBS. Every time I’ve done it in the past, it’s always a big time penalty to transfer. Even at SODO, and until buses are forced out of the tunnel in 2020?, it costs half the headway of Link, plus a couple of minutes to exit the bus and queue up for Link, with nothing in it for the rider.
      The other consideration are the platforms at both RBS and SODO. I don’t think they were ever designed to handle several fully loaded artics arriving at the same time, in the wind and the rain, and expecting everyone to transfer ‘for the greater good of the network’, with smiles on their faces.

      1. https://seattletransitblog.wpcomstaging.com/2010/12/29/transfers-at-rainier-beach/

      2. I don’t think I’m arguing that anyone should transfer for the good of the network. What I’m trying to argue in this post is that there *is* something in it for the rider. The service hours saved from not running buses along I-5 are directly funneled back into the South King network.

         For all riders, this will lead to a dramatic improvement in reliability and frequency. For the subset of riders who already transfer in Kent or Renton, this will also lead to a dramatic improvement in trip time as well, and without forcing riders to transfer any more often than they do today.

         Here are my rough notes from the calculation process. If anything doesn’t make sense, feel free to ask.

         Note that it took me some effort to design the network so that it would lead to travel time improvements for all riders. Here are three key observations I made:

         – You need 10-minute frequency on the core segments (Link to Kent, and Link to Renton) to match the wait-inclusive travel times of today’s network.
         – The bus to Kent needs to take I-5 between Rainier Beach and Southcenter. Any other route takes too long.
         – Elsewhere in the network, 20-minute frequency actually leads to better expected wait times than 15-minute. With 20-minute frequency, if you’re coming from Link, your expected wait time is either 0 or 10 minutes, or an average of 5. With 15-minute frequency, your average expected wait time is 7.5 minutes. So yes, it’s actually better to have less frequent service, if that service has a timed connection with Link.

         Finally, don’t forget that the biggest improvements from this change are for riders coming from routes like the 169 and 180. Today, they transfer from an infrequent suburban route to an infrequent I-5 route. In the proposal, they transfer from a more-frequent suburban route to very-frequent Link. Those riders — and my calculations suggest that they’re probably a majority of all 101/150 riders — see travel time reductions of 20% or more.

      3. Nice workup Alexs. MT150 leaving Kent at noon, arrives 62 minutes later at Westlake, so the extra time over the transfer at RBS would be about 8 min. Not a game breaker!
         I like the grid like feel to it, and some of the new OD pairs created, such as giving Auburn back a through route to Westfieild Mall, in exchange for the transfer to Link at RBS. That’s more useful than going to the airport. The 164 becomes more productive also.
         The higher frequencies on other segments is a major plus, so on balance I have to give you TWO THUMBS UP on your proposal.
         When buses leave the tunnel, the additional time on 3rd will wipe out any differences that exist, so what the heck, just do it!
         mic.

      4. MT150 leaving Kent at noon, arrives 62 minutes later at Westlake, so the extra time over the transfer at RBS would be about 8 min.”

         That’s what I thought too because people normally calculate travel time starting at the first vehicle’s departure rather than when we arrive at the stop. But Aleks is suggesting we include the wait time for the first vehicle. That erases the “Rainier Beach overhead” and actually turns it into an asset over the status quo, because the 150 drops to 30 minutes evenings/Sundays and the 101 drops to 30-60 minutes. Waiting at a stop or adhering to a 30-60 minute schedule is the #1 thing people hate about transit and what most depresses ridership. And if you’re transferring (say 168+150 or 180+150), you can’t avoid that waiting time so in that case it’s better to go to Rainier Beach than to stand waiting bored at Kent Station.

      5. The key to making all of this work is timed train->bus connections in the outbound direction (a.k.a. buses that hold for trains). Without the guarantee of the timed connection, things don’t work.

         There is also the possibility that the additional Link riders this restructuring would generate would justify increasing the frequency of trains, say, to run every 8 minutes, instead of every 10.

   3. There’s very little additional time needed to run the buses to Sodo vs. RBS

      This isn’t true. The difference is six miles on I-5, which is not free-flowing at many times of day. That’s at least seven and often more like ten minutes in each direction, which is enough on routes of this headway to require an additional bus.

      1. Metro schedules the 150’s trip from Tukwila P&R to Busway & Spokane at 11 minutes off-peak. Tukwila P&R to Rainier Beach Station would likely take 5 to 7 minutes. If we allow 2 minutes to get from Spokane & Busway to SODO station, there’s a difference of about 6 to 8 minutes. How do you define “very little additional time“?

   4. There’s very little additional time needed to run the buses to SODO versus RB

      For Renton, this definitely isn’t true.

      Getting between Rainier Beach Station, and the first stop on MLK after exiting the freeway, takes about 5 minutes. It’s on local streets that aren’t particularly congested, and so that’s likely to be a reliable time for most of the day.

      Getting between SODO and that same stop takes 10 minutes on the schedule. Of course, if there is any traffic on I-5, it could be much worse than that.

      However, the bigger problem is interlining. You need to have a route on MLK, and a route on Renton Ave, because there’s demand on both. The route on Renton Ave naturally goes to Rainier Beach. Two 20-minute routes combine for 10-minute service between Rainier Beach Station and Renton TC. If you extend the MLK route to SODO, you’ve actually doubled average wait time, unless you also reroute the Renton Ave route to SODO (which would be awkward and circuitous).

      1. Looking at the current 101 schedule, the trip from S. Renton P&R to SODO Station would take about 25 minutes off-peak, plus 13-18 minutes to transfer, wait and complete the trip to University St (38-43 total trip time). I would estimate the trip time from S. Renton P&R to RBS at 20 minutes (allowing 8 minutes from the MLK/129th timepoint to RBS) plus 30-35 minutes to transfer, wait and complete the trip to downtown (50-55 minutes total trip time). Much faster via SODO.

   5. There’s very little additional time needed to run the buses to SODO versus RB

      I disagree. Until we build left-side transit-only ramps to the busway at Spokane in both directions (absolutely needed, not in anyone’s budget), delays on this particular stretch of I-5 will continue to be guaranteed at rush hour, and to appear unpredictably the rest of the time.

   6. The way the network is set up today, the 150 acts as a trunk route. That means it’s a route people transfer to to get to Seattle. Many people transfer once to get there. Some poor souls may have to transfer more than that. So adding another transfer point at RBS would just make their trip even lonnngeeerrrr. You can restructure all you want, but you’ve got to make sure that you get people to their destinations in a timely manner.

      For instance, if you’re in Enumclaw….take bus to Auburn…..next 578 45 minutes away dang it missed it so take the 180 to Kent, catch the 150….enough transfers already dang it let me get to Seattle already.

      1. John, that’s why Aleks’s proposal restructures almost all the South King area. Virtually every route will connect to a Link station, so riders can transfer once to Link without having to take the 150 at all. Granted, your rider from Enumclaw is one of the few exceptions: he’d need to take the 180 from Auburn to Ranier Beach before transferring to Link. But even today, he’d need to take the 180 from Auburn to Kent before transferring to the 150 (or to Seatac before transferring to Link.) (Both of these are assuming, as you did, he missed the 578.)

         So, I don’t see how this restructure exacerbates the problem you bring up.

9. Federal Way to Kent?

   1. A or 184 to 168, or alternately 181 to 180.

      The one-seat ride on today’s 183 is gone, but that’s a completely terrible way to get from Federal Way to Kent anyway.

      1. The 183 is adequate, but it does need some straightening up. What it needs even more is expanded hours!

         But is it worth giving up the direct(ish) connection between FW and Kent for a route that simply mirrors the existing A-line?

         Come to think of it, the proposed 184 is somewhat inherently redundant, because most of the time, the route is within walking distance of the A-line anyway. Would it make more sense to route the 184 from Federal Way TC to Kent (more directly than the current 183), then farther northeast, to either Renton TC or Fairwood center (w or w/o Fairwood loop)?
         It does fill in a weak spot in connecting to Federal Way. By that I mean direct north and direct east connections from FW are strong, but northeast travel would always require at least one transfer. This is a big deal for Federal Way because (since FW is in the corner of the county) all places significantly west and/or south of the FWTC would require a transfer to go beyond that. So for NE travel from these places, we would be looking at 2 or 3 (!) transfers instead of 1 (transfers, not buses).

10. Getting to my doctor’s office in Kent from my frequent western Burien route goes from a 2-seat ride (120 -> 180) to a 3-seat ride (120 -> F -> 164/180).

    But wait, oh, maybe I can still do it with 560 -> 169… assuming the 560 survives in it’s current form for much longer

    No further complaints, looks like a pretty solid plan to me.

11. Aleks, a couple of questions.

    Why would you change frequency to the Renton highlands from 30 to 20 minutes? Do you have data that shows current service is overloaded?

    Also, why would you still have the F Line going through the South Renton P&R? If you want to shave 5 minutes off of unnecessary deviations, that would be the place.

    1. Sam, if overloads were the only reason to increase frequency, our night network would still have 120-minute frequencies.

       Aleks’s 105 is through-routed with the 106, which has 20-minute frequency to connect with every other Link train.

       1. David, don’t you think it’s important to look at what type of capacity a route is experiencing throughout the day before suggesting the headway be changed? You can’t, on the one hand, say you’re concerned with a route’s alignment efficiency, then turn around and say a bus’s capacity efficiency doesn’t matter. If a bus is running one-third full during off-peak hours, isn’t it inefficient to provide more service during those times?

       2. Sure, capacity is a factor in frequency decisions, but it’s not the only factor. If it were, we’d have virtually all of our service hours go to peak service, when there is currently a lot of overcrowding. But the goal is to make a network that’s usable all day. There is a happy medium between providing way more service than is warranted in a corridor (such as upgrading the 183 to 15-minute service) and making entire areas hard to reach because the frequency is too low. The 105 is the major service to the Renton Highlands, a geographically isolated area with a lot of low-income residents, and it has pretty good ridership for a suburban 30-minute shuttle route. It’s a good candidate for more frequency.

    2. Sam does seem to have a point about the F-line, though.

       Also, why does the F line keep its existing jog east just west of Southcenter, when the 180/164 take a straighter path on 61st?

       1. William, I mentioned this idea back in April. I said the F Line shouldn’t go into the South Renton P&R. I think every regular Metro route that goes to the P&R also goes to the TC, so riders can connect there. Keep the F Line on Renton Ave and it will shave five minutes off its time. And people who need to go to or near the P&R can get off on Renton Ave and walk. It’s just two blocks away.

       2. As I mentioned above, I tried to avoid proposing any changes to routes that don’t act as feeders to the 101/150.

          I think there are lots of routes in Metro’s network that could be improved (of course), including in South King. But I mostly want to convince readers that moving the nexus of SKC-Seattle service from Kent/Renton to Rainier Beach is not only possible, but highly desirable. Changing routes like the 140 doesn’t really prove that point one way or the other.

    3. Sam, Metro answered your first question. They proposed to boost the 105’s off-peak frequency to every 15 minutes in the F Line Restructure. Also, the 2013 Service Guidelines Report says the 105 is underserved during peak.

    4. Regarding the 105, aside from what everyone else said, it actually wouldn’t be any cheaper to reduce Renton Highlands frequency to 30 minutes. The reason is that there’s already a bus on Renton Ave every 20 minutes, and the marginal cost of sending the bus east is actually very low.

       Regarding the F, I think you’re right that it would be better not to loop into the park-and-ride. However, I tried to avoid changing routes any more than necessary. The only routes I’m proposing to change are the 101, 106, 150, and other routes that are designed as feeders for those routes. To the extent that a bus already serves a “crosstown” market, I tried to leave it the way it was. Even then, I think that it would be possible to implement the basic idea of my proposal with significantly fewer routing changes, albeit at the cost of some reliability or span of service.

12. Title 6.

    1. Not sure what you’re trying to say. I believe this proposal strictly improves the speed, frequency, and coverage of the existing network for every single existing rider, including (especially!) along several corridors with a disproportionate number of low-income residents. If you disagree, can you be more specific?

       1. Title VI isn’t specifically about income, but analyzing income demographics tends to be a useful approach to analyzing ethnic demographics. Transit agencies have to do a Title VI analysis with every restructure. The only ways I know for a restructure to run afoul of Title VI is removing or reducing service along a path with stops where there happen to be ethnic minorities who ride transit. Aleks’ proposal improves frequency or leaves it as it for nearly every route path in South King County (other than express portions). It also reduces travel time for nearly every major destination pairing.

13. Honestly what should happen is to build an elevated Link Line along I-5 from SODO to just south of Boeing Field, and extend the MLK segment down to Renton. (and discontinue the short stretch between Rainier Beach and the I-5 crossing).
    Link, under the current configuration, is just too slow to replace Freeway Buses. Therefore, it should be reconfigured into two lines. One that serves TIB, the Airport, Angle Lake, and potentially Federal Way (Grade Separated, much like north link), and another line that runs along MLK and then reaches Renton.

    A Street running Light Rail Line will never be a viable replacement for Freeway Buses.

    1. The problem with that idea is, in a word, frequency. Would you be willing to cut headways in half to support it?

    2. A Street running Light Rail Line will never be a viable replacement for Freeway Buses.

       Not true, even today.

       The existing line, indirect routing and all, is only a tiny bit slower to the airport than the old 194 when traffic is clear, and faster whenever there’s the teeniest bit of traffic.

       1. One thing about freeway routing that is important to remember, is that even though the cars may be doing 70 mph, Metro buses usually top out around 50-55. Going up the hill along SR-518, I doubt the old 194 did better than about 35 mph.

          Stuff like this is one reason why the difference in running time between Link and a freeway-routed bus is often less than one might think. Then, factor in congestion, dwell time at stops, and the fact the the train usually has a lot more consistent headways and less bunching.

       2. Knock on the nearest piece of wood furniture before a car crashing into a post brings the whole Link system to a smooth halt. As part of the analysis, it seems fair to consider what the back-up plan is when Link is blocked. The most straightforward answer may be to continue some of those routes from their connection point on to downtown. But can this be put into play reasonably quickly, with extra buses thrown onto the impacted routes?

          I’m up for the idea of elevating Link through the valley, and in the SODO, to someday remove all at-gradeness with other modes, not because of the 1-2 minutes in travel time it would save, but because it would open up the option of driverless trains, with more frequency, and a lot fewer accidents to boot.

       3. Given time and will, we should be able to have driverless trains anyway, grade separation or not. Driverless cars are almost here and the technology for a driverless train down MLK has got to be easier than the technology for a driverless car in arbitrary traffic conditions.

          If it doesn’t happen, it will be largely the union’s doing (e.g. bus drivers threatening to go on strike if trains are coverted to driverless systems). But, grade separation or not, we would have to fight the union to go driverless anyway, so I don’t see what difference it makes.

          Then, there’s the reality that adding grade separation on MLK now would cost a tremendous amount of money, not to mention severe disruption of service for everyone that uses it for several years. Yes, it has some value, but is it worth spending several billion dollars and forcing Link riders to slog it all the way downtown on the 7 or 36 for 3 years? In my opinion, the answer is no. I would much rather see the money spent on future expansions, such as rail to Ballard.

    3. How much frequency does anyone really need to the airport anyway? When you go there, you know you are going to have to sit in a security line for a while, plus all manner of other nonsense that means you have to spend a fair amount of time waiting.

       If you wanted to actually get faster service to the SeaTac airport from downtown Seattle, you could “build” a second line by dropping a ramp from the existing line down onto the BNSF main line. It’s a three track + line from there to SODO, and you can add a station for the King County Airport too, which has no transit service at all now. Lots of effort has been put there at improving speed and capacity to for Amtrak and Sounder as well.

       I know, I know, FRA regulations prohibit light rail on freight main lines. However, CalTrain was granted a waiver for having European equipment operate over the Union Pacific main line. It’s part of their electrification process. Collision avoidance signals will have to happen at some point anyway, as that is a mandate from congress, and when that happens the FRA should be a bit more lenient on what it allows on the main line.

       That assumes, of course, you have the budget to build something.

       1. Frequency to the airport is actually a fairly big deal for two reasons.

          First, most people (including everybody north of downtown) do not live within walking distance of a Link station and have to take a connecting service. So good frequency on Link means not getting stuck with a 20-30 minute wait in the middle of the trip.

          Even if someone is driving you to the Link station, the potential for random, unpredictable traffic delays still make frequency of the train important.

          Similarly, on the way home from the airport, it is impossible to predict exactly when you will get off the plane and get your bags, so it’s difficult to time your departure to match limited schedules.

          Also, Link is not only about travel to the airport. A lot of people use it just to get around town.

       2. Yes, but every 10 minutes, so that you have to break the existing LINK to a two – prong line in order to get a direct express to the airport?

          Furthermore, wouldn’t the taxi companies and airport shuttle companies have a fit about having a publicly subsidized competitor?

    4. With what money? Seattle has higher priorities than a SODO bypass, namely a Ballard-downtown line, a Ballard-UW line, a downtown-West Seattle line, and something to Lake City. A bypass would serve the South King/Pierce areas who aren’t paying for it. And why are they more important than people going to/from Rainier Valley? The purpose of a subway is to connect neighborhood centers where the pedestrians are, and that’s what the Rainier Valley routing does. Splitting the south segment into a Y may make sense in the long term, but the tight South King funding and the lowish Renton ridership don’t justify it at present.

       1. Of course, the SODO bypass is not a priority for Seattle, as it isn’t Seattleites who would directly benefit (unless they work at the airport, in Federal Way, or in Pierce County). It is the Federal Wayans and Pierce Countians who would want the bypass.

14. Quick! Someone at King County Metro hire Aleks before some big money consultant back east grabs her up, and replaces a few letters in her name with dollar signs.

    1. Is Alex actually a “she”, or is it just the picture?

       1. I am transgender, and my preferred pronouns are ‘she’, ‘her’, and ‘hers’. That’s all I’m going to say on the subject for now.

       2. Not that it really matters!
          Still, really brilliant work crunching the numbers, Alex.
          Mic, former Metro So.King Co Sounding Board Co-Chair

       3. Indeed!
          Anyone with this type of vision could probably make quite a lot of money attempting to untangle the mess at SEPTA, or a few other places.

15. Thanks for upping the role of Rainier Beach station in this vision!

    One minor modification to the map might help: Show Route 7 on the diagram! Rainier Beach station is served by Route 7, which is a major trunk route for SE Seattle. Today, it’s really hard for SE Seattle residents to get to Southcenter unless they live near a Link Station (which still requires them go out-of-direction to TIBS to get to the Southcenter area).

16. [ot]

    1. [ot]

       1. [ot]

       2. [ot]

       3. [ot]

       4. [ot]

       5. [ot]

       6. [ot]

17. [ot]

    1. [ot]

       1. [ot]

       2. [ot]

       3. [ot]

       4. [ot]

       5. [ot]

    2. [ot]

       1. [ot]

       2. [ot]

       3. [ot]

18. John Bailo mentioned an express to Angle Lake Station in one of the above threads that was actually related to this post.

    I’d like to hear how some ideas for what such a route should look like. I’m assuming it would serve Kent Station, hopefully with 10-minute frequency during the day, since creating some sort of faster service between Kent and downtown Seattle is a primary point of the route.

    What is the quickest path between Kent Station and Angle Lake Station, each way? Should other stops be served? What would be the travel-time cost of these stops?

    Should there be one or more neightborhood tails?

    Would this route end up being competitive with the current 150, Aleks’ proposed 164/180+Link, or the oft-proposed re-routed 578 coming half-hourly? Remember in your calculations that Kent Station is a transfer point, not where anyone actually starts their trip.

    1. Taking a quick look at the map, I think such a shuttle would be best non-stop, as there doesn’t seem to be any compelling destinations along the direct route (Kent Des Moines Road to I-5) that would justify a stop, especially when every stop would mean getting off the freeway and meandering through stoplights.

       That said, if you’re going to run the Kent shuttle to Angle Lake Station, you may as well just run it to SeaTac Airport station. It’s only minimally further and eliminates the transfer for people going to/from the airport.

19. The more I dig into Alexs’ proposal, the more I like it. Having a one seat and faster ride between the Highlands, Fairwood, E.Kent Hill, GRCC and SE Auburn, to RBS and Link sure makes a lot of sense.
    I’ve always started with the departure time from Kent Station to figure if truncating at RBS was justified.
    Having lived within walking distance of Kent Station, I can say there aren’t a lot of us walking. Most arrive by bus, and that was the kicker for me. Eliminating the KTC transfer makes it come out good.
    I was concerned about the times transfer at RBS (a lot), knowing it’s congested there, and not a lot of curb space for buses to lay over, waiting for it’s load of passengers from Link. Each train, every 10 minutes midday, would need a Renton bus (106 or 169) or a Kent bus (164 or 180) that has moved up from the layover spaces further away, and is ready to ‘load and go’. So that’s two buses queued up.
    Add to that a Fairwood bus (107) every third Link train, and hopefully a 7 bus on new wire every other train or more often, and that’s a lot of curb space, but doable. It seems like a multi-lane bus queue, off street would be an optimal solution as part of this, maybe something under the power lines.
    Have you talked to Service Development about this much, maybe Doug if he’s still doing the south end?

    1. One other comment, then I’ll shut up. Early iterations of Link Operations showed every other train terminating at RBS, which is why the center pocket track was built, just south of the station. This sure makes some sense to increase frequency on Link to at least RBS when ULink starts, with only half the trains required to continue on nearly empty to S200th (Angle Lk Stn).
       Everyone wins on the improved headways through MLK.

20. Pingback: What We’re Reading: SLU Is A Boomtown | The Urbanist

Comments are closed.