The Frequent Network Plan

Frequent, fast, and affordable bus transit for Seattle

Archive for the ‘STB Reposts’ Category

Frequent Network Plan – Questions and Answers

with one comment


Here are the two versions of the FNP map:

Here are three reference documents covering the Frequent Network Plan in more detail:

  • A short list just listing each route, its frequency, and its key corridor or destinations.
  • A detailed route-by-route list of all the routes in the proposed all-day network, with basic descriptions of each route and a listing of technical hurdles each proposed route would face.
  • A cross-reference with current service, explaining how riders of every current all-day route would be served by the proposed network.

Key Questions

Why are you doing this? You don’t work for Metro. Metro doesn’t want to do any more “big bang” restructuring projects. And if it did, it wouldn’t use your ideas.

This is really about showing people what is possible with current resources, not about turning Metro’s network into this exact proposal next year or in 2021.

Metro has many talented, well-informed people on its planning staff, and they have done a fantastic job of improving some parts of the network in recent years. Metro has moved toward the principles underlying this plan in its own recent restructuring efforts; you will see that this plan leaves many of Metro’s newest creations almost totally unchanged. But, in the end, even the best planners at Metro remain accountable to top management, which in turn is accountable to the County Council. As a result, it is politically difficult for them to float larger-scale proposals or to show what is possible if we disregard political constraints. Since I’m unconnected to Metro (other than as a former operator), it’s easier for me to do that. I can start a discussion without raising the sort of alarm — “My bus is going away next month!” — which makes reasoned debate difficult.

In the end, my goal is to raise awareness among activists and ordinary bus riders about how a new frequent network could improve people’s lives. In advocating a frequent network, I don’t mean to dismiss some riders’ fears. Change is hard because people have adapted their dwelling places, commutes, and personal lives to the current network over a period of years. But the current network creates huge limitations on the growth of the city. Spontaneous trips by transit are not possible when so many key routes only run every 30 minutes, so people can’t live without cars and the space and expense they imply. Transfers are a nightmare. Transit is not competitive with driving when lines are so indirect and slow. This proposal is about showing that all of these things can be substantially improved without a magic source of new money.

Even though I’m less attached to these particular proposals than to the general idea more speed and frequency, this presentation is highly detailed for two reasons: first, to show conclusively that a much more useful network is possible at current funding levels, and, second, to present some specific new ideas that I think are worthwhile. The key takeaway is this: Transit in Seattle can be much easier, faster, and more fun.

Are these ideas all yours?

Certainly not. Many ideas in the Frequent Network Plan are mine, but many aren’t. Metro planners, SDOT (through the TMP), and frequent STB posters all deserve credit for some of them. Within the STB world, I should particularly thank Bruce, Zach, Martin, Adam, Andrew, Brent, Matt, Mike Orr, d.p., Anandakos, and Aleks. All of them (and more folks that I know I’m forgetting) have contributed valuable ideas to the community that are reflected in this proposal.

This is only for weekdays during the day. But riding Metro is worst at night. What about nights? Can we improve there too?

With the right core network in place, improving service at any time becomes easier. The goal of this project is to put forward an idea about what the right core network would be. Creating serious transit proposals is time- and labor-intensive, and simply designing this all-day network and preparing it for public presentation has consumed pretty much all of my spare time in recent months. (Yes, I’m a transit geek.)

I can tell you in general terms that it’s impossible to continue 10- and 15- minute levels of service on all of these frequent routes into the night with current resources. But night service is not that resource-intensive, because it runs fast. With a bit of extra revenue, we could get to frequent night service everywhere. And, even on today’s resources, we could have 15-minute night service in quite a few more places than we do today.

What about peak hour? How would you change the peak-hour network? Do people’s peak-hour commuter rides have to go away to make the Frequent Network Plan work?

I haven’t proposed a new peak-hour network simply (again) for lack of time, but there are many ways to improve Seattle peak service as well. Peak-hour riders are most concerned with speed, and many commutes could be faster than they are today. If I have time in the future, I may add a peak-service overlay to the Frequent Network Plan.

The all-day network in the Frequent Network Plan uses only the resources used for the current all-day network (at current all-day frequencies), along with hours from a tiny amount of peak-only service that would be made entirely redundant by the new all-day network. It would not require taking any resources from other current peak-only service (including current peak frequency improvements to all-day routes), although significant changes to the peak-only network would be necessary for peak service patterns to make sense with this all-day network.

Some of these lines go outside of Seattle. What is the scope of the Frequent Network Plan?

The Frequent Network Plan is primarily about the city of Seattle. It includes the city, and unincorporated areas south of the city which the city could annex at some point: North Highline and Skyway. It also includes Shoreline and Lake Forest Park, because both Metro and Sound Transit include them in the subareas that include Seattle. A few routes that extend outside Seattle are included because they are the primary local service for communities in Seattle.

Similar improvement would be very possible both in South King County and on the Eastside. If I have the time and the inclination, I may eventually propose a network for South King County, which I think is a much higher priority than the Eastside. (I invite others to do the same!) As a longtime resident of Seattle, though, I have to put the city first.

Is this all about Metro? Are you proposing changes to Sound Transit services?

This is almost all about Metro. For now, the Frequent Network Plan assumes almost no changes from Sound Transit and no changes from the City of Seattle. It includes the two already-funded streetcar lines; already-under-construction portions of Link; and existing ST Express service patterns. Only one change from Sound Transit (the addition of two stops to ST Express route 522, at NE 80th St and NE 115th St, for transfers to newly frequent Metro routes) would be needed to make the network work.

When Sound Transit makes a decision about station locations for the Lynnwood Link extension, I may update the Frequent Network Plan to include a comprehensive restructure of service in far north King County.  Such a restructure would necessarily imply significant revisions to ST Express route 522.

Why doesn’t the Frequent Network Plan mention RapidRide? Existing RapidRide routes revert to numbers, and there are no new RapidRide routes.

Because I’d rather see RapidRide-style improvements applied to every frequent bus route, without the separate branding. RapidRide is not true BRT, and riders are disappointed because they expected BRT. Had  RapidRide-style changes been made to ordinary frequent routes, as funding permitted and without the over-the-top marketing, they would have been considered a success, not a failure. When possible, those improvements should be made across the system, first to the 8- and 10-minute corridors, then to the others.

I recognize that, in the real world, federal obligations require keeping the RapidRide brand on the existing RapidRide service. Thus the RapidRide colors used on the route map for the three existing RapidRide routes. Also, all three are upgraded to 10-minute frequency under the plan, and all three see at least minor routing improvements (major, in the case of RapidRide D/FNP route 15).

Which routes get extremely frequent service?

Link, of course. Beyond Link, lots of bus routes, as follows.

A bus would come every 7-8 minutes on the following key corridors:

  • Part of FNP routes 2 and 10: Westlake to Pike/Pine Corridor via Pine St
  • Part of FNP route 3 (partial): Belltown to First Hill via 3rd Ave, Yesler Wy, and Jefferson St
  • Parts of FNP routes 5 and 40: International District to Fremont via 3rd Ave, Belltown, and South Lake Union (alternating between Dexter Ave N and Westlake Ave N, but with evenly spaced departures northbound from downtown and southbound from Fremont)
  • Part of FNP routes 13 and 24: Pioneer Square to Uptown via 3rd Ave, Denny Wy/Broad St, and 1st Ave N/Queen Anne Ave N
  • Part of FNP routes 14 and 34: S Jackson St through the International District
  • Part of FNP routes 34 and 35: Jackson St to Beacon Hill Station via 12th and 14th Aves S
  • Part of FNP routes 50 and 55: California Ave SW between Admiral and Alaska Junctions
  • Parts of FNP routes 65/67 and 73/75: U-District Station to University Village via UW Campus (changing route numbers at UW Hub)

A bus would come every 10 minutes on the following key corridors:

  • FNP route 7: South Lake Union to Rainier Beach via Fairview Ave, Boren Ave, and Rainier Ave S
  • FNP route 8: Uptown to Madison Valley via Denny Wy, John St, and Thomas St
  • FNP route 12: Colman Dock to Madison Park via E Madison St
  • FNP route 15: Downtown to Crown Hill via Elliott Ave W and 15th Ave W/NW
  • FNP route 44: UW Hospital to Ballard via 15th Ave NE, NE/N 45th St, N 46th St, and NW Market St
  • FNP route 48: U-District to Mt. Baker TC via UW Hospital and 23rd Ave
  • FNP route 54: Downtown to Westwood Village via Alaska Junction, California Ave SW, and Fauntleroy Ferry Terminal
  • FNP route 58: International District to Aurora Village via 3rd Ave and Aurora Ave N
  • FNP route 120: Downtown to Burien via Delridge Wy SW, Westwood Village, and Ambaum Blvd SW

What does this mean for my neighborhood?

In the near future, I’ll be writing a series of posts for STB highlighting the advantages (and occasionally the disadvantages) of the Frequent Network Plan for particular areas of the city, and zooming in on some of the specific restructure ideas in the plan. It improves frequency in a huge number of places, and makes quite a few new connections as well. You can see and read about the entire network today, though, by clicking the links above.

How did you choose the route numbers?

In most of the network, if the large-scale corridor a FNP route serves is broadly similar to the large-scale corridor served by an existing all-day Metro route (or is similar except for an extension), I kept the existing route’s number for legibility. Where large-scale corridors have been reconfigured, I chose new route numbers.

The only place I broke entirely with current numbering is in north King County. Numbering routes in the 300 series no longer makes sense, for two reasons. First, many 300-series routes serve Seattle extensively or even primarily, so the borderline for choosing a 300-series number is very hazy. Second, Metro’s service area is now divided administratively into three components: Seattle and North King, South King, and Eastside. It makes sense to have just three series of route numbers to match. Fortunately, there are more than enough unused numbers that all current all-day and peak 300-series routes can be renumbered. 358 (and RapidRide E) becomes 58, while 330- and 340-series all-day routes get new numbers in the 80s with the same last digit as the existing routes.

Why doesn’t this plan break more through-routes?

Metro has seen major reliability improvements from breaking through-routes on long, frequent corridors. Examples include the former 7 (now 7 and 49), the former 43 (now 43 and 44), the now-separated 10 and 12, and older through-routes involving previous Aurora, Delridge, and U-District core service. The Frequent Service Plan has several through-routes that could negatively affect reliability, although none of the remaining ones tend to have as severe an impact as those broken in the past. (Difficult through-routes still in the FNP are the 3/13, 5/21, 16/124, and 24/132.) The plan only breaks a couple of current through-routes, although more could be broken with increased resources.

I believe firmly that increasing frequency is more important than breaking through-routes for improving the rider experience, provided that the through-routed routes are not too long. Just like increasing frequency, breaking through-routes is enormously expensive in service hours; for instance, it would require more service hours to break the through-route between the current RapidRide C and D Lines while keeping the current 15-minute frequency than it does, in this proposal, to increase frequency on both lines to 10 minutes all day. Breaking through-routes also results in more congestion in already-congested areas.  For long through-routes, there should also be a focus on small capital projects, such as transit signal priority, queue jumps, and BAT lanes, which can improve reliability through the worst areas of congestion.

What would be the effect on reliability? Are you cutting layovers to the bone to make this happen?


In this post, Bruce gave a great example of real-world scheduling, and showed how improving frequency gives disproportionate bang for the service buck — an effect the Frequent Network Plan relies on very heavily. But, often, past increases in frequency have been accompanied by shortening layovers (known as “recovery time”), which means that buses may not get to the end of the line before they are supposed to leave for the next trip. This can have catastrophic effects on reliability. On top of that, Metro has dramatically cut recovery time across the system in recent years, partly in response to the famous (or notorious) 2009 performance audit. The Frequent Network Plan does not restore pre-2009 recovery time, but it should allow for slight improvements on many routes, and it does not achieve its higher frequencies by cutting recovery (with one or two exceptions on generally reliable routes). Given the straightening of routes and the breaking of two through-routes, I would expect the Frequent Network Plan to be more reliable on balance than today’s all-day network.

What would be your first priorities for further improvement with more resources?

More… frequency! Based on ridership and load-factor numbers, the following improvements come quickly to mind:

  • Improve route 58 to 7-8 minute frequency
  • Improve routes 14 and 70 to 10-minute frequency
  • Improve the portion of route 59 between Westwood Village and Upper Rainier Beach to 15-minute frequency
  • Improve route 1 to 15-minute frequency, possibly in conjunction with a restructuring to connect West Queen Anne with central Queen Anne
  • Improve route 131 to 15-minute frequency

Also, significant frequency improvements in far north Seattle and Shoreline will become possible when the Lynnwood Link extension opens.

I would also like to see one more route extension, if additional resources were available:

  • Extend route 78 from East Green Lake to Ballard via N/NW 65 St and 24 Ave NW. This would require substantial physical street improvements on N 65 St between Linden Ave N and Phinney Ave N (which cannot currently handle buses), but would create a new one-seat crosstown connection between Wedgwood, Roosevelt, Green Lake, and Ballard.

Give me some numbers!  I’m as much of a transit geek as you are, because I’ve read this far!


How many daytime buses are required for the current all-day Seattle network, and for the Frequent Network Plan?

There are enough quirks in the network that these numbers are necessarily a bit inexact.  With that said, I am using a baseline of 324 all-day buses in the part of the current all-day network that the FNP would replace.  The FNP calls for 337 all-day buses.  The hours for the extra 13 buses are from the following peak-only work that would be made entirely redundant under the proposal:

  • Some (not all) peak trippers on RapidRide C and D
  • Peak trippers on route 8
  • Peak trippers on route 11
  • Peak trippers on route 12
  • Peak trippers on routes 26 and 28 (local only; express trippers not used)
  • Peak trippers on route 41
  • Peak trippers on routes 43 and 44
  • All route 48 Express service
  • Extra peak trippers on routes 71/72/73/74 that allow for route 74 service
  • Some (not all) peak trippers on route 120
  • Peak trippers on routes 131 and 132
  • All route 330 service
  • All route 373 service

Some additional peak trippers would be devoted to improving recovery time on all-day routes during PM peak.

How many daytime buses are you devoting to each route?

The number of buses below covers the base all-day frequency.  Additional buses would be added to some routes during peak hours to improve either frequency, recovery time, or both.  The hours for those additional buses would come from peak service not in the list above, typically on routes where the proposed all-day service level is not as high as the current peak service level.

Route Number of buses Cycle length Notes
Route 1 3 buses 1:13
Routes 2, 10 9 buses 1:56 Cycle includes complete roundtrips on both routes
Interlined so all coaches get longer recovery at the route 10 terminal
Routes 3, 4 9 buses 1:54
Routes 3 (TB), 13 9 buses 1:58
Routes 5, 21 14 buses 3:04
Route 6 10 buses 2:12
Route 7 14 buses 2:00
Route 8 9 buses 1:10
Route 12 8 buses 1:02
Routes 14, 70 10 buses 2:10
Routes 15, 54 17 buses 2:22
Routes 16, 124 13 buses 2:49
Routes 24, 132 16 buses 3:38 Some coaches will interline with route 131 at Burien to take advantage of long route 131 recovery downtown
Route 28 3 buses 0:33
Route 31 10 buses 2:00
Route 34 8 buses 1:44
Route 35 9 buses 1:56
Route 40 15 buses 3:03 Route 62 will be replaced with partial route 40 trips to improve reverse-peak frequency between Ballard and Stadium
Route 44 10 buses 1:19 Will add 1 bus during mid- to late PM peak for extra recovery, from peak tripper on current route 65
Route 48 10 buses 1:22
Route 50 9 buses 2:01 Will add 1 bus during peak hours for longer recovery, from peak trippers on current route 50
Routes 52, 55 8 buses 1:37
Route 58 15 buses 2:09
Route 59 6 buses 2:48 Will add 4 buses during peak hours for 20-minute frequency and longer recovery, from peak trippers on current route 7X
Routes 65, 67, 73, 75 19 buses 3:55 Cycle includes complete roundtrips on both sets of routes (65/67 and 73/75)
The routes are interlined for longer recovery
Route 69 6 buses 1:14
Route 71 11 buses 1:34
Route 78 7 buses 1:25 Would require approximately 5 additional buses for N 65 St/Ballard extension
Routes 81, 85, 87 8 buses 3:38 Cycle includes 5-minute pulse in both directions at Northgate TC
Routes 86, 88 5 buses 2:21 Cycle includes 5-minute pulse in both directions at Northgate TC
Will add 1 bus during peak hours for longer recovery, from peak trippers on current route 316
Route 106 4 buses 0:50 Will add 1 bus during peak hours for longer recovery, from peak trippers on current route 106
Route 107 (and 148) 5 buses 2:17 Will add 1 bus during peak hours for longer recovery, from peak trippers on current route 7X
Route 120 15 buses 2:07
Route 128 3 buses 1:26 Will add 1 bus during peak hours for longer recovery, from peak trippers on current route 125
Route 131 5 buses 1:58 Some coaches will interline with route 132 at Burien to allow more coaches to take advantage of long route 131 recovery downtown
Total 337 buses Note: Cycle length includes minimum 5-minute layover at each terminal (except live loops)

Copyright 2013 David Lawson. This material is licensed under the Creative Commons CC-BY-SA 3.0 license. You may freely modify and share it as long as this attribution remains.

v. 1.1.0


Written by David L.

July 29, 2013 at 16:41

Posted in FNP, STB Reposts