"The state has since reduced restrictions on cars allowed in the bus lane, and will soon even permit single-occupant vehicles to use it."

once again, only south of Edsall. From there north it will remain Bus/HOV3. the part that will be open to SOV's will be a HOT lane, and will still be free flowing, at least in theory, and so should still be a desirable route for buses. Yeah Shirley highway is not true BRT - boarding is off route, almost entirely at conventional busstops - its still added to transit usage in its corridor, vs no transit accommodation.

I realize this is somewhat orthogonal to your point, but I do get tired of the notion (advanced by Dan as well) that somehow the Shirley highway has been completely dismantled as a transit facility, when it is in fact still a very important transit facility for large numbers of NoVans - so important, that Arlington County went to considerable trouble to fight HOT laning it within the County

by AWalkerIntheCity on Jul 24, 2012 10:35 am • link • report

This post points up some of the reasons that BRT is a terrific group of techniques to upgrade bus service but not a substitute for light rail.

If bus stations are set away from intersections, as recommended here, transfers at points where lines cross are inconvenient.

If buses run at high frequency, it becomes impossible for pedestrians to cross the street and for vehicles to make left turns except at widely spaced signalized intersections. This makes it an obstacle similar to a freeway. Light rail, on the other hand, can carry several hundred people in a single train and therefore can carry a heavier passenger load while allowing cars and people to pass between. Also, light rail can be put underground in downtown areas much more easily than buses.

When buses run frequently, signal priority becomes difficult or impossible. The signal priority on the Cleveland Health Line was turned off, due to protests from motorists on cross streets.

If regular buses are forbidden to use the bus lanes, BRT loses its vaunted flexibility.

Bus lanes, off-bus fare collection, etc. are great features for buses. And if your city is already defaced by an elevated expressway, putting dedicated bus lanes on it can improve mobility. But BRT is no substitute for rail in creating a good urban environment.

by Ben Ross on Jul 24, 2012 10:51 am • link • report

In 2010, the I-395 HOV roadway carried close to 9,000 persons in the morning HOV-3 restricted period in an assortment of buses (Metrobus, Alexandria DASH, Fairfax Connector, PRTC, Martz, Quick's and probably a few others).

That's not bad, and would probably have been higher had the decision not been made in the 1970's to turn nearly all Metrobus service in the corridor back at the Pentagon as soon as the Blue Line started operating. As I suggested in response to Dan's recent post, it did not help that the HOV-3 restriction on the express lanes was removed north of the Pentagon in about 1988 or 1989.

Additionally, the two I-395 HOV lanes also carried over 23,000 persons in car-pools, van-pools and yes, some single-occupant hybrid vehicles.

by C. P. Zilliacus on Jul 24, 2012 10:57 am • link • report

Ben, I think "Frequently" here means a bus every two or three minutes, which looks *empty* as far as a road is concerned, but is very good service for bus.

by Michael Perkins on Jul 24, 2012 10:57 am • link • report

BRT, like Communism, is one of those things that sounds like it would work in theory, but doesn't work in practice. And then when it fails, its advocates complain, "But REAL [Communism/BRT] hasn't been TRIED!"

I'm sure that there are circumstances under which BRT could work, but those circumstances will never exist, and if they do exist, circumstances will quickly change to degrade BRT service.

I think the best bet is to stop trying to push BRT in the hopes of "getting it right" this time. It's just not going to happen.

by JustMe on Jul 24, 2012 11:04 am • link • report

Michael - A bus every two minutes in each direction is a bus passing an intersection every 60 seconds. Signal cycles on main roads where I live are 100 to 150 seconds. So that's more than one bus per signal cycle. Buses stack up at signals and must wait for the light, rather than getting automatic priority. Pedestrian crossings away from signals are dubious at this frequency and impossible when frequencies get much higher.

On the other hand, a bus every 5 minutes is 12 buses per hour. If they carry 60 passengers each, they carry 720 passengers per hour. A highway lane with stoplights can carry more people, so it is hard to justify a bus-only lane next to a backed-up general-purpose lane. (If the general-purpose lane isn't congested, there's no need for the bus lane.)

So we find that "real" BRT can fit into the city like light rail when buses run every 3 or 4 minutes, but no more and no less.

by Ben Ross on Jul 24, 2012 11:14 am • link • report

"A highway lane with stoplights can carry more people, so it is hard to justify a bus-only lane next to a backed-up general-purpose lane."

to add conitinuity to the transit system (which may mostly be rail and conventional bus) - to have more tranit users who tend to be bike/ped users more when not using transit - to impact auto use externalities.

While the total persons moved is an important metric, I am not sure its the only one that should impact the BRT decision.

by AWalkerInTheCity on Jul 24, 2012 11:30 am • link • report

This raises a point (perhaps an obvious one, but as I live on the other side of the river I've not been following the BRT discussion): how will provisions be made for turning across the bus lane? And what about for pedestrians? For the latter I'd be an advocate of some kind of pedestrian bridge crossing. I know the arguments against them, but really - they DO serve a purpose and can be made quite useful.

by Ser Amantio di Nicolao on Jul 24, 2012 11:40 am • link • report

A bus every two minutes in each direction is a bus passing an intersection every 60 seconds... Pedestrian crossings away from signals are dubious at this frequency and impossible when frequencies get much higher.

@Ben: Could you explain this a little more -- this doesn't seem to track with observed experience. Looking around at D.C. streets, 15 second gaps between traffic tend to spark widespread jaywalking on a wide 2-lane street with parking on both sides; 25 second gaps do the same on a 4-5 lane street with parking.

So a sixty-second frequency between buses from either direction seems like more than ample time for pedestrians to cross away from signals.

by Arl Fan on Jul 24, 2012 11:57 am • link • report

There are a couple of differences. To be "rapid," buses on a suburban BRT lane should travel at the speed of adjacent cars, 40 to 50 mph. Unless there is a pair of medians between the bus lanes and the regular lanes, you need to have a gap in both regular traffic and the bus lanes.

by Ben Ross on Jul 24, 2012 12:28 pm • link • report

@JustMe

"Real" BRT has been tried, and has a pretty good track record for success. Jarrett Walker writes pretty extensively about successful Australian BRT systems on his blog.

However, you may be correct to be cynical that a functioning BRT system can come out of the American political process. However, it has been done successfully outside of the US.

by andrew on Jul 24, 2012 12:47 pm • link • report

One of the things I like about "quasi-BRT" is its flexibility. To me it is not a matter of whether it is better or worse to have BRT, LRT, or the "quasi-BRT" but more of what is more appropriate to the given situation.

To me a good example of this comes in Pittsburgh on the MLK East Busway. This is a pretty much no-frills high speed bus transportation facility that has proven very adaptable. There are routes that service just the busway and those that use the versatility of the buses to fan out in the lower density areas to enable a broader reach of curb to curb service and one seat rides. At the same time, there is a very good volume of park and ride usage along the line. The buses are simply off-the-shelf models and fare collection is no different than the rest of the network.

It is certainly an example of where I simply can't see rail or even "true BRT" having the same level or combination of effectiveness, convenience, and efficiency.

Would rail or "true BRT" have had a bigger impact on economic development? Possibly, but doubtful. One need only look at the Burgh's Overbrook line to see that development density and rail alignment are not directly related.

by AJ on Jul 24, 2012 2:28 pm • link • report

Annie is right that the Cleveland HealthLine is a great example and very successful. It is a great project that could be copied across the country. You could move more people up and down Rhode Island Avenue on this type of system than you could even fathom compared to the present situation or a streetcar at a fraction of the cost. And as in Cleveland, you would have economic development to go along with it, probably more so than the planned streetcar given it would be more practical for someone to actually commute by mass transit from upper RI Ave.

by neb on Jul 24, 2012 4:36 pm • link • report

I've never heard of the term "Shirley Highway Busway" and never heard that it was thought of as a BRT system. As far as I know, it was always considered to be an express lane for buses and car pools, later renamed a High Occupancy Vehicle (HOV) lane. As an express lane, it works relatively well. But to be a true BRT, it would have needed stations along it - so I think you chose a bad example.

by John Flack on Jul 25, 2012 10:01 am • link • report

These self-styled "standards" seem to reflect an auto-first way of thinking:

"Stations, placed at intersections, leave insufficient space for free left turns (the equivalent of free right turns in the U.S.), and left-turning vehicles yielding to pedestrians back up traffic."

No, the "space" for free left turns can come from having a dedicated pedestrian-only crossing phase in the signal cycle, or from timing a protected left-turn signal phase before the cross traffic with the bus.

"If bus stations are set away from intersections, as recommended here, transfers at points where lines cross are inconvenient."

Very true: by making the BRT less convenient for passengers, these "standards" make its failure more likely.

"This raises a point (perhaps an obvious one, but as I live on the other side of the river I've not been following the BRT discussion): how will provisions be made for turning across the bus lane? And what about for pedestrians? For the latter I'd be an advocate of some kind of pedestrian bridge crossing."

Pedestrian bridges are often not for pedestrians, for whom they often represent a gross inconvenience of extra time and effort spent going up and down in a confined space that is vulnerable to crime, but rather to allow more cars to go through at a higher rate of speed.

Plus, these bridges are absurdly expensive in the United States, because the Americans with Disabilities Act has led agencies to include elevators on both sides in order to avoid liability. One planned pedestrian bridge over a normal street intersection in Los Angeles (Lankershim/Campo de Cahuenga) is estimated to cost $19 million! How much transit can that buy? Depending on an agencies cost structure, up to 200,000 revenue service hours!

by These BRT Standards Need Fixing on Jul 25, 2012 12:50 pm • link • report

Pedestrian bridges are the worst. Inconvenient for pedestrians, more dangerous than being on the street (people often ignore them late at night to avoid running into someone while alone), and expensive to build since we have to make them ADA compliant. Maybe they're acceptable as a last resort for allowing people a path across a highway, but on a BRT corridor? There are going to be traffic control devices (stoplights) along it, and in order to be transit-friendly it must also be walkable, so pedestrians should have a signal where they can cross at grade.

by MLD on Jul 25, 2012 1:44 pm • link • report

[This just out from the GAO]

Projects Improve Transit Service and Can Contribute to Economic Development GAO-12-811, Jul 25, 2012

From the summary:

U.S. bus rapid transit (BRT) projects we reviewed include features that distinguished BRT from standard bus service and improved riders’ experience. However, few of the projects (5 of 20) used dedicated or semi-dedicated lanes— a feature commonly associated with BRT and included in international systems to reduce travel time and attract riders. Project sponsors and planners explained that decisions on which features to incorporate into BRT projects were influenced by costs, community needs, and the ability to phase in additional features. For example, one project sponsor explained that well-lighted shelters with security cameras and real-time information displays were included to increase passengers’ sense of safety in the evening. Project sponsors told us they plan to incorporate additional features such as off-board fare collection over time.

The BRT projects we reviewed generally increased ridership and improved service over the previous transit service. Specifically, 13 of the 15 project sponsors that provided ridership data reported increases in ridership after 1 year of service and reduced average travel times of 10 to 35 percent over previous bus services. However, even with increases in ridership, U.S. BRT projects usually carry fewer total riders than rail transit projects and international BRT systems. Project sponsors and other stakeholders attribute this to higher population densities internationally and riders who prefer rail transit. However, some projects—such as the M15 BRT line in New York City—carry more than 55,000 riders per day.

by C. P. Zilliacus on Jul 25, 2012 4:57 pm • link • report

Leave it to transit advocates to only consider a system true BRT if it is the most expensive per mile and 'whoops' you might as well build light rail.

What should happen is a system of improvements for increased speed and reliability for the bus system.

The biggest improvement people can make is to end cash transactions on boarding the bus. That means more than introducing an electronic pass, it means differential pricing so that everyone is forced to use it.

In London people bitched and moaned when they made cash fares twice the price of Oyster card fares, but within a month 95% of all bus boardings had switched over to electronic payment.

Just this change has made a huge difference to the speed of bus travel. You notice it when you drive behind a bus in Suburban London. There are bus will pull over for a few seconds to pick someone up and be off again, contrast that across the border in some small suburban town and getting stuck behind a bus for 30 seconds while it picks up just two passengers(no flat fares in the outer suburbs either) is a contrast in frustration.

Build some nice bus stops, just a little shelter with a bench and electronic next bus indicator will do wonders. Many councils in the UK have a rolling programme of level boarding rebuilds where a stretch of the curb is built higher to an exact height above the road so that when a bus is properly parked there are just milimeters in height difference between the floor of the low floor bus and the sidewalk.

This is not only good for the mobility impaired and parents with strollers it also speeds boarding.

Again it depends on the US city, but all busses should be low floor with a simple extending ramp from the middle door for disabled travelers.

Then we come to the difficult bit: Bus lanes. They don't have to be the full length of the route, just the parts that get congested. People seem to want them to be fully segregated to keep traffic out, but all you need is some political bravery, video cameras on every bus and an automatic fine of $200 and after you've issued a few hundred thousand fines no one in the city will ever drive or park in a bus lane again. People learn, you just need politicians who are not lily livered. That also means that the cops or city workers are not automatically exempt. Once those departments have to set up a process to get them overridden then they will clamp down on their own workers flouting the rules.

The number one difference between the US and Britain, is that transport departments are willing to go much further in the UK in reducing traffic lanes for cars and switching them to bus lanes.

There is no reason a decent bus system has to cost tens a millions a mile, you just need some paint, some good buses and decent fare system to get you most of the way there. If you want a higher capacity system with an artic bus every minute then you may need fancy middle street lanes with mini stations and multi door loading. But it's not necessary. I feel that much of the cost of US systems is to do with avoiding conflict with the car lobby.

So Do you know how most curb side bus lanes were painted on British roads? They just banned street parking on those stretches of road. I'm not sure how far that would get you in the US.

by Rational Plan on Jul 26, 2012 5:09 am • link • report

SerA: to answer your question, look at what Alexandria is building near Potomac Yard: there'll be separate left turn lanes and protected-only signal phases for left turning traffic crossing the bus lanes on Route 1. And not every intersection will be full access...only some.

by Froggie on Jul 26, 2012 8:42 am • link • report