P u g e t S o u n d


Bus Rapid Transit for the Metropolitan Puget Sound

Donald F Padelford



"We wish we had never started the whole thing. Fixed rail is not the answer to the transportation needs of our city. We should stop all this insanity that has gone on these past years."

Richard Riordan, Mayor of Los Angeles, on public radio program,
"Which Way LA?," June 1998.

The mayor was right, both for his city and for ours. We should stop railing at gridlock and put in place a Bus Rapid Transit system that would

The View from 20,000 Feet

The spine of the Puget Sound BRT system would be comprised of 150+ miles [1] of freeway HOT lanes (described below). Another 100+ miles [2] would be made up of arterial bus-only lanes. Additionally there would be short sections of busways, bus tunnels, and possibly elevated bus-on-rail (guided-bus) structures, similar in size and mass to the existing Seattle Center monorail. Finally Sound Transitís downtown Seattle to SeaTac Airport "Initial Segment" of light rail, currently under construction, would be converted to guided-bus technology.

HOT lanes (high occupancy / toll) are HOV lanes that have been converted to transit priority and emergency vehicle use [3], with other vehicles admitted, but only to the extent that there is room, via a variable toll. This toll is collected electronically and varies every few minutes [4]. Transit vehicles include buses, vanpools, and qualifying carpools [5]. The objective is to keep the HOT lanes in free-flow (50 to 60 mph) at all times, even during the height of rush hours. Several such systems are in place around the nation [6], and one is being implemented here on Highway 167 south of Renton [7].

Freeway HOT Lane Entrance, San Diego

Bus-only lanes would be built primarily in the parking strips of major arterials. Gates, such as one finds at parking lots, would keep out non-transit vehicles [8]. These lanes could be rush-hours-only or full time. Particularly if they are full time, it would be necessary to more than replace parking lost to merchants with pubic parking lots or garages, perhaps providing 200% of the parking formerly found on-street [9].

Arterial Bus-only Lane

Busways are simply bus streets, such as the E3 Busway south of downtown Seattle. There are times when bus through-put is such that it requires a street rather than a lane. Bus tunnels, of course, are busways that are under ground.

Downtown Seattle Transit Tunnel

Guided-bus systems are a hybrid of bus and rail systems, having some of the characteristics of each. On the one hand buses ride on (concrete) rails (like some trains in the Paris Metro), producing a smoother ride than is normally found on surface streets [10]. On the other hand they are able to drive off the rails at each end of the track to pick up and drop off passengers. Such guided-bus operations can be on the surface, in tunnels, or elevated. A first-generation guided-bus system, parts of it elevated, was built 20 years ago in Adelaide, Australia. Other systems have been built in Germany and England [11].

           Adelaide Busway                      Rubber-tired Train, Paris Metro

Finally, while Sound Transitís Initial Segment of light rail represents a foolish squandering of literally billions of scarce transportation dollars [12], it is becoming a "sunk cost" which cannot be just abandoned [13]. Therefore the right thing to do is to convert it to a more appropriate technology, such as bus-on-rail (guided-bus), which would dovetail with Bus Rapid Transit in other corridors [14].

Guided Busway

BRT Issues

Capacity vs light rail. The plain fact is that Bus Rapid Transit on HOT lanes (BRT/HOT) has more capacity than heavy rail (BART-type) let alone light rail (Portland Max-type) systems. The reason for this is that the capacity of any transit system is determined at the stations. With rail a train cannot, of course, enter the station unless the one before it has left. The time between trains is called the "headway," and it invariably is two or three minutes at a minimum, more on "non-grade-separated" systems (systems running on surface streets and therefore prone to auto and truck interference, such as Sound Transitís Initial Segment of light rail). With BRT/HOT, however, the transit unit (bus) pulls off the right of way (HOT lane) to discharge and take on passengers. Under these circumstances, you can put a transit unit down the right of way every few seconds rather than every few minutes. [15] (Note that buses on busways also have the ability to unload and load passengers "off-line" in the manner of BRT/HOT.)

Stuck in traffic? As demonstrated around the nation, HOT lanes flow at highway speeds at all times. And because buses and vanpools always have absolute priority, their travel time is guaranteed. Off the freeway, buses on bus-only lanes with traffic signal preemption move without being subject to undue traffic interference. And of course buses on busways or guided-bus structures move at their design speed. [16]

Support for "walkable" neighborhoods. Some proponents of light rail believe that Bus Rapid Transit cannot support "walkable" neighborhoods. However this view is becoming a minority one now that many examples of just such support are showing up around the nation and world. Below is a representative quote, this from a recent "Railvolution" conference in Salt Lake City [17].

"BRT-transit" Oriented Development
Is it really true that you have to have steel wheels to make a good [transit oriented development, ie a "walkable" neighborhood anchored by transit]? Many would suggest that this is a myth - one that can be dispelled by examples from South America, Australia, Canada and the United States. This session will provide an international perspective on how BRT and TOD can and do work together. Learn how TOD can be more than an accidental result of BRT development, but how with attention to design elements and partnerships, BRT actually shapes communities and achieves the same developmental objectives of its steel-wheeled cousins.
O v e r v i e w : Find out the myths vs. realities of whether you really need steel wheels to make a good TOD.

Quality of service. A persistent theme among certain local politicians is that you canít have a "world class" city region without rail. Possibly this is true if your buses are stuck in traffic, if they "belch" diesel fuel, if their stops are exposed to the weather, and if the buses lurch down the road. But none of these apply to high-quality BRT systems. Being stuck in traffic was dealt with above. As for diesel belching, our buses are increasingly first generation hybrids, and the next generation will be even cleaner and quieter. Stops can and should be the equivalent of rail stations: clean, dry and comfortable, with tickets bought off the bus and boarding level with the platforms [18]. As for lurching, this is a matter of the controls and the roadbed. With modern controls, buses can start and stop as gradually as a train [19]. And the roadbed can be as good as we are prepared to pay for. Of course with busways and guided-bus structures no other vehicles are allowed in the right of way, which aids in their maintenance [20].

Expense vs light rail. This is a big subject, but because most of the right of way (roadbed) is already in place, BRT systems are about a tenth of the cost of light rail in this region. Where light rail has been built economically, such as in San Diego or Salt Lake City, either the rails have already been in place or the topography is such that itís easy to lay them without condemnations or tunneling [21]. Because it is so much less expensive, for the same budget BRT can cover about ten times the area and support ten times as much Transit Oriented Development ("walkable" neighborhoods) as LRT.

Guided-bus technology. Guided buses are not necessary for BRT. On the other hand, given the money spent to date by Sound Transit, it makes sense to convert their Initial Segment of light rail to a bus-on-rail system (see above). The same technology, because it so accurately guides them, would allow four lanes of buses to pass each other in the Downtown Seattle Transit Tunnel, and thus permit "skip stops" with the consequent doubling of transit through-put [22]. Also, this technology, along with the relatively light weight or buses compared to trains, enables elevated bus structures to be slim and compact, making them a good neighbor to communities along major arterials, such as SR99 (Aurora). While earlier, first-generation systems use a mechanical guide-wheel to steer the buses, todayís guided-bus systems generally use magnetic or optical means of guidance (although mechanical systems may be used as a safety back-up) [23].

Through-put of HOT vs general purpose lanes. This is not strictly a Bus Rapid Transit issue. However, it is germane because it indicates where the technology underlying BRT/HOT could lead. We all know that if you put too many cars on a freeway, it bogs down into gridlock, this due to demand exceeding supply (potential through-put). When this condition prevails for any period of time, the "traffic pipes" in effect contract, causing the highway equivalent of a coronary thrombosis. The fact that this "thrombosis" happens just when supply is needed the most, at rush hours, is doubly ironic. One approach that has had success with this problem is ramp metering, which has markedly increased the through-put of our local freeways. Congestion Pricing, the mechanism behind HOT lanes, could be considered "ramp metering on steroids". If it were applied to all freeway lanes [24] (which, of course, is not necessary for BRT/HOT), it could increase the capacity of our freeway network by up to twice their current rush hours through-put [25].

Bus Rapid Transit is, or should be, the mode of choice for high capacity transit in the metropolitan Puget Sound region. It has far more capacity than competing systems, supports "walkable" neighborhoods, and provides 60 mph transit mobility to the entire metro region 24/7. Furthermore if implemented, it would set the stage for an increase of freeway capacity up to double current rush hours through-put -- with free-flow conditions guaranteed literally forever, should we decide to pursue that sometime in the future.[26]

The Puget Sound BRT System in Concept [27]



  1. 150 miles = 300 "lanes miles", ie a lane each direction for one mile. Currently 200 lane miles of the core HOV lane system are complete, and the remaining 100 lane miles have been funded.
  2. This is just a guess, albeit I believe a reasonable one.
  3. The role of emergency vehicle mobility is not explored in the OpEd, but it is extremely important. In every-day situations, ambulances, police cars, tow trucks and the like need to get to accidents to clear them off. Incidents and accidents cause somewhere between 30 and 50% of daily congestion on the regionís freeways (Niles 10/10/05, Hallenbeck 10/11/05). This is not to mention the necessity for emergency vehicle mobility in 911 or Hurricane Katrina type situations.
  4. In the case of San Diegoís I-15 HOT lanes, the toll is varied every 6 minutes. Once a vehicle is on the HOT lane, however, the toll does not change for that vehicle.
  5. What counts as a "qualified" carpool varies from HOT lane to HOT lane. On San Diegoís I-15 HOT lanes itís HOV2 (2 or more people per vehicle). In other places itís higher. The extreme situation, favored by most people who have spent time pondering the optimal arrangement, is that "qualified" be limited to "vanpools and better" (vanpools and emergency vehicles go free, everybody else pays) which would make management of the HOT much easier. All vehicles using the lanes would have a transponder (a gizmo about the size of a pack of cards which goes on the windshield and which tells the "reader" the identity of the vehicle). Of course if youíre in a two-person carpool, each person pays half to toll; a three person carpool, a third, etc. With "vanpools and better" the "mobility agency" will never have to "raise the bar" from HOV3 to HOV4 etc, the way it otherwise may have to. With that lower standard (eg HOV3) there is also the issue how do you tell who is cheating? Thereís a car on the HOT lanes, indicating it has at least 3 people in it, but does it? How do you know? There are technological solutions (a. Make everyone "check in" with a kiosk that looks like one of those remote drive-through bank tellers; a person on a remote video would randomly check the cars. b. Every car in the HOT lanes would need to carry a cell phone capable of streaming video to do the so same kind of random check-in), but they are expensive and complicated to set up and administer. However, in the end this is obviously not just a technology question. Politics ultimately decides.
  6. For instance San Diego, Houston, Minneapolis. In addition many cities are planning HOT lanes, including an extensive systems slated for the Washington DC ring freeway. Related but different are the "cordon pricing" systems in Singapore and central London, and considered for Manhattan.
  7. This is a test or prototype, although if extended to the full HOV lane system some changes would be required. For instance on SR167 HOV2ís (2 person carpools) will be allowed to drive free. If extended to the full HOV lane system, the bar would need to be raised to HOV3 or higher.
  8. Right turns would be permitted.
  9. These would charge the same as or less than on-street parking. Inexpensive municipal lots, built to support merchants, can be found, for example, on Rodeo Drive in Beverly Hills (where the items for sale, of course, are not inexpensive).
  10. Rails also allow the electronic "training" (hooking together) of buses. When in "train mode" such a system would be virtually indistinguishable to the transit patron from, eg, a rubber-tired Paris metro train. The difference is that guided-buses can go their separate ways when they reach the end of the tracks, delivering the patron directly to their destination without need of a transfer.
  11. At about 12 miles, the Adelaide system, which uses German guide-wheel technology (therefore their name for this technology, "O-Bahn", which translates as "omnibus," ie bus, "rail" Ė which I have translated as "bus-on-rail") is the longest. The UK seems to be in the lead in terms of the most number of systems.
  12. About $3,630,000,000 (2005$), per MacIsaac (8/19/05), using Sound Transit source data.
  13. Having obtained federal dollars, the region has made a commitment to the federal government to complete it the Initial Segment of the light rail line between downtown Seattle and just north of SeaTac Airport. It may be, however, that the guided-bus technology could fall within the definition of "light rail" (it does run on rails) and could thus fulfill that commitment. Of course within the political realm what ultimately counts is not so much legal agreements as it is political clout. If youíve got the votes, virtually anything is possible.
  14. It is important that the various segments of a transit systems to fit together as much as possible.
    Following this principle, Sound Transit is pushing for light rail on the I-90 bridge. However, according to the agencyís own numbers, while doing this and building out the rest of the first phase of the Eastside system would cost in excess of $6 Billion, it would have no effect at all on total trans-lake transit ridership! Compared to implementing Bus Rapid Transit on both bridges (SR520 and I-90) rail would simply shift riders from one bridge (520) to the other (I-90). This for $6,300,000,000?
  15. See http://www.bettertransport.info/padelford/DSEIS%20ST2%201204.htm
    footnotes # 17 and 18. I refer to this document as "My DSEIS Comments".
  16. HOT lanes, bus tunnels and elevated bus-on-rail structures are "grade separated" and not prone to interference from cars or trucks. (It could be argued that HOT lanes are not technically grade-separated since non-transit vehicles are allowed on these lanes. However, transit and emergency vehicles always have absolute priority and other vehicles are ejected to the extent needed to keep the lanes running in free-flow. Therefore, at a minimum, they are "virtually" grade-separated.)
  17. http://www.railvolution.com/workshops.html
  18. Ciritiba, Brazil was a pioneer in various aspects of BRT systems, including this feature. Note that the Ciritiba system is not a guided-bus technology.
  19. I donít specifically know whether systems have been deployed with this objective, but itís not exactly rocket science.
  20. One of the main benefits of a guided-bus system is the smoothness of their ride. The Adelaide authorities claim the ride to be as good as or better than high end light rail.
  21. The relative pricing of BRT/HOT vs light rail is has numerous "moving parts", not the least of which is that Sound Transit has never priced out their full, 125 mile system. And more often than not, when they have priced out a segment, costs have turned out to been lowballed to say the least. For example Sound Move, the document used to sell the light rail system to the voters, estimated a cost of $110 Million per mile (year-of-expenditure dollars) and repeatedly claimed this to be "consciously conservative." But the system is turning out to be more like a $234 Million per mile system (SeaTac to Northgate) (MacIsaac 8/19/05).
    Therefore any comparison of BRT to light rail must be largely "back of the envelope." However, when the cost of one systems are an order of magnitude (10x) higher than another, an envelope may be all thatís needed. In any event, for my take on this see
    My DSEIS Comments, footnotes #8 and 9, as well as Appendix B, footnote #3
  22. The Right of Way (ROW) in the Downtown Seattle Transit Tunnel (DSTT) is 34.8 feet. The ROW for a guided-bus system (Adelaide) is 20.3 feet for 2-way travel.
    The sidewalks in the DSTT are 14 feet.
    20.3' x 2 (2 lanes each way) = 40.6'
    40.6 - 34.8 = 5.8', the amount by which the existing ROW is too narrow for 4 lanes
    5.8 / 2 = 2.9' or 2' 11", the amount of sidewalk that would need to be taken for 4 lanes. Call it 3' 
    14 Ė 3 = 11 feet, the width of the sidewalk remaining after 4 lanes of guided-buses put into the DSTT.
    (Padelford to MacDonald 4/27/05)
    (See also My DSEIS Comments, Appendix B, footnotes #12 and 13)
  23. More precisely, test systems in Japan and elsewhere have demonstrated magnetic and optical guidance.
  24. For another view of regional "congestion pricing" see
  25. The HOT lanes on Highway 91 in Orange County, California carry twice the numbers of cars per lane and the adjoining general purpose lanes. However, a more typical number is 40% more (Burris, 10/4/05). Since the Puget Sound region is in the upper reaches of congestion, a number somewhere between 40% and 100% is probably the ballpark. The halfway point between those numbers is 70%. Whatever the exact number, if HOT technology were applied to all freeway lanes, they would carry a lot more cars per hour.
  26. While any freeway lane will stay in free-flow forever, this does not speak to the issue of how many lanes or lane-miles should be added to the regional freeway system. For instance, it may be that I-405 should acquire an additional 2 lanes each way. Congestion pricing does not change this one way or the other, except in the sense that HOT lanes will carry more traffic during rush hours than current, general-purpose lanes (see preceding footnote) and in the sense that they will flow at 50 or 60 mph 24/7, in perpetuity.

    If all freeway lanes were congestion priced several issues would arise.
    > What do you do with the HOT lanes (ie the HOV lanes turned into HOT lanes), and in particular, what do you do with the HOT to HOT interchanges, which are quite expensive. Provisional answer: turn them into truck lanes. Separating truck and non-truck traffic would make the freeways considerably safer.
    > What do you do with all the money raised by congestion pricing tolls? One thing you could do is use it to maintain and improve the highways. Another thing you could do is distribute it, periodically and per capita, to the voters of the region (Snohomish, King, and Pierce counties). This would work something like the annual distributions to citizens of Alaska from the Alaska Investment Fund, which was funded from North Slope oil receipts. In any event, deciding how to spend a surplus is a good problem to have.
  27. Not, of course, to scale.

See also http://www.bettertransport.info/padelford/DSEIS%20ST2%201204.htm  ("My DSEIS Comments") This document, which embodies my comments on a Draft Environmental Impact Statement, goes into numerous other aspects of this issue not covered here, including demonstrable bad faith on the part of Sound Transit.

Jump to more information on BRT from the Puget Sound Public Interest Transportation Forum

Jump to more information on public transportation from Seattle citizen Don Padelford