Buses Done Right

Buses have some substantial advantages over rail as a transit modality, notably being cheaper, quicker, and more flexible. But some of these advantages can also serve as flaws -- bus systems tempt administrators into creating shoddy service and tend not to spur development. So one important issue moving forward is going to be thinking more rigorously and more creatively about buses and bus networks and what can we really do with them. The new Shirlington Bus Transfer Station in a part of Arlington County that's far from Metro seems promising in several respects. Chris Zimmerman, who's on the County board and the WMATA board, is quoted as calling it "a great example of where you can do transit-oriented development even where you don't have a rail station."

If that really works, it could be a promising model for other parts of the country where new rail construction would be infeasible but where density is sufficient to support multiple bus routes that have some frequency.

Comments

The problem with buses is that they cost more in the long term. Rail is pretty solid stuff. While ties can degrade and need regular maintenance, the lines themselves will last decades. Roads, not so much, and buses help hasten the demise of roads. Buses are certainly useful, but I think they are best used in conjunction with other forms of transit.

Ideally, as the new suburbs return to urban development and leave the 70s behind, buses will have a limited role in transit as communities tie themselves to the rail lines of the core cities (and eventually are annexed by said cities).

Posted by freddiemac | June 27, 2008 11:05 AM

Well, except that because the station is tucked away against the 1-395 retaining wall in a hard-to-reach corner with narrow turn radii, my 10B bus now has to winds it way through Shirlington to get to the station, which, thanks to several traffic lights and stop signs, adds about 10 minutes to my twice daily ride ...

Posted by lcindc | June 27, 2008 11:09 AM

I hate it! It will never work.

Posted by ixmner | June 27, 2008 11:15 AM

freddiemac,

But in cities and towns of modest size, the volumes on even main transit routes may never be high enough to justify the sunk costs of rail.

And even in the more diffuse parts of larger cities, the need to make a transfer from a bus to a train can severely hamper the adoption of public transit, particularly by commuters. So if you can't serve certain customers from the start with rail, it can make more sense to provide a dedicated expressway for buses (and just for the part of the route where congestion on the mixed-use highways becomes an issue). That allows those customers to take express buses from their local area to the busway and then straight onto the expressway without a transfer.

Incidentally, I think the latter sort of approach can in fact help spur development in communities served by the dedicated expressway--perhaps not quite to the same degree as a rail line would in the areas immediately adjacent to the expressway, but more so than rail in the communities that would first need to take a shuttle bus to a rail station, and now can instead be served by buses without a transfer.

Posted by DTM | June 27, 2008 11:20 AM

Ideally, as the new suburbs return to urban development and leave the 70s behind, buses will have a limited role in transit as communities tie themselves to the rail lines of the core cities (and eventually are annexed by said cities).

I think not. You can't possibly build a rail net dense enough to serve all development -- there will always be feeder bus lines.

In this case, most of the bus lines presumably connect this rapidly-growing high-rise concentration with the Pentagon Metro by way of the bus lanes on I-395 -- not more than a mile or two as the crow flies. This is NOT the boondocks, to all intents and purposes it is part of the downtown core.

Posted by roac | June 27, 2008 11:30 AM

Buses have some substantial advantages over rail as a transit modality, notably being cheaper, quicker, and more flexible.

Which one is better for throwing people under? I hear all this talk about throwing people under buses, but throwing people under trains doesn't seem to get the attention it deserves.

Posted by washerdreyer | June 27, 2008 11:43 AM

Buses quicker than trains? On what planet? Maybe if you had a dedicated bus lane or something ... otherwise buses will always have to stop with traffic.

Buses certainly are necessary as feeders to trains/subways. Unless you are in Manhattan (i.e. even in much of Queens, Brooklyn and the Bronx), the point about rail not being dense enough applies. But, IMHO, it's ridiculous that the most efficient way to get from NYC to certain parts of NJ is to take the bus all the way there -- the bus is certainly slower than even taking a car and if only a few more rail lines were added, it would be quicker to take a train.

Posted by DAS | June 27, 2008 11:49 AM

I think MY meant "quicker to set up and start operating."

Posted by roac | June 27, 2008 12:09 PM

I'm glad for the expanded transit. But this quote from the article was silly:

"Shirlington really was hampered for decades - the only way you could get there was to drive and spend 15 minutes looking for parking," he said.

I've been going to Shirlington regularly for the past, oh, 10-15 years, and I don't think I have ever had to drive that much to find a space. It's a little harder now that there is more stuff there, but if you go to the massive old parking garage next to the stores, or the really big new one next to the library, it's not such a challenge. Oh, and they are free.

No dummies, over in Shirlington. When they moved Signature theatre, built the library, and expanded the shopping and dining, they thought about the cars too.

Posted by Brad L | June 27, 2008 12:12 PM

Where the patronage on the line is too low for heavy rail, you use light rail. Where its too low for light rail, you use buses.

Gosh, now wasn't that simple?

There is, of course, nowhere where the potential patronage exists that "rail cannot be put in" in a technical sense or in terms of full economic costs ... if conventional light rail or tram/trains aren't suitable, then a technology like Aerobus (light rail track on suspension cable) will certainly be a possibility.

However, quality Express buses are an important transitional stage, and for some routes they will be upgraded to electric trolley buses rather than to light rail, just because the potention patronage of the route does not justify the light rail.

And for those services, anchors on the Express route will include not only transfers onto light rail, intercity rail, and mass transit, but also transfer stations into the network of "last mile" connections via buses on the regular public right of way.

Posted by BruceMcF | June 27, 2008 12:12 PM

"I think not. You can't possibly build a rail net dense enough to serve all development -- there will always be feeder bus lines."

Yeah, I know. That's what I meant by limited role.

I admit I didn't follow Matthew's example very closely because I don't know much about DC metro, I was more generalizing. Apparently they are trying to build more walkable and dense in certain parts of Texas. Isn't that crazy?

Posted by freddiemac | June 27, 2008 12:13 PM

And even in the more diffuse parts of larger cities, the need to make a transfer from a bus to a train can severely hamper the adoption of public transit, particularly by commuters.

Not if the Buses are done right. It will if the bus routes are not designed to integrate well and the transfer is an after-thought ... but that is just saying that if it is not designed to do the job, it won't do the job well.

In places, like Perth Australia, where a new rail system includes a redesign of the bus route system so that the bus route system serves as an effective feeder, the redesign increases the patronage of the bus service. In particular, a part of that segment of the market that just won't ride the bus is somehow able to say to itself, "I'm stepping on the bus to catch the train", and that makes it OK. The Maggie Thatcher quote about any man over the age of 30 on the bus is a failure comes to mind ... its somehow OK if its a short bus ride to catch a train.

A common transfer model is that a well-designed transfer is equivalent to twice the waiting time. If the transfers are just tacked onto an existing bus route system, and if fares are collected by mode of transport instead of by trip, the wait time to be certain of catching any given train can be substantial, killing off demand for the transfer. However, a well designed, well integrated bus route system can easily offer travel time savings to commuters on a wide range of trips compared to bus-only options that offsets the perceived inconvenience of the transfer.

Its just as easily a political/instututional problem than a strenuous design challenge ... a problem of the "bus people" and the "rail people" with strong institutional incentives to fight for their mode rather than co-operating to provide more effective trips opportunities to riders.

Posted by BruceMcF | June 27, 2008 12:27 PM

Brad L. is right that the quote from Jim Moran is stupid. Hardly anybody is going to take transit to Shirlington to eat or go to a movie, because parking is easy and free. The point of upgrading transit service is to give the people who live in the new residential high-rises a quick way to get to the Metro.

Arlington has made itself a model by encouraging high-rise development around the Metro stations, thereby moderating automotive congestion and keeping its property tax rates the lowest in the DC area. As Chris Zimmerman says in the linked article, this is an effort to expand the model to areas that the subway does not serve directly. (As I noted above, the dedicated bus lanes are already there -- though I do not know first-hand that they connect Shrlington to the Pentagon.)

Posted by roac | June 27, 2008 12:33 PM

Not only do buses cause more damage to streets than streetcars (which only put pressure on their track, not pavement), they're also reliant on diesel fuel which has rapidly increased in cost. In addition they have slower acceleration rates than streetcars, and buses last from 10-15 years while streetcars have life expectancies ranging from twenty to thirty years and heavy rail cars can be expected to last for thirty to forty years with proper maintainance. Plus, it's considerably easier to maintain an electric motor over a diesel engine. So while rail may not be the best solution for location for various reasons, one simply cannot dismiss the efficiencies that can be found in rail transit applications.

Posted by David Alexander | June 27, 2008 12:34 PM

David Alexander,
Not that this undercuts your entire point, but DC buses primarily run on Natural Gas.

Posted by Rock On | June 27, 2008 12:44 PM

You people are all delusional! I have seen the future, and it is Loudoun County, and it works!

Posted by ixnermay | June 27, 2008 12:44 PM

Buses only have advantages over trains when the options are "take the bus," and "walk because there's no train going where you are headed." Not only do they have to stop with (frequently crowded) traffic, but they also make frequent stops at, um, Bus Stops. As a result, a trip that may take 20 minutes by car and maybe 10 by train can take an hour or more by Bus. Unless you have no absolutely no alternative, or you simply like being on a sweaty, crowded slow moving monstrosity, riding the bus makes almost no sense.

Trains go fast, don't stop with traffic, are far more comfortable, and lets' face it, more aesthetically pleasing. And in addition, while they do have more up front costs, long run they're just cheaper than maintaining a massive fleet of buses. Also, they have the added advantage of being impossible for an incompetant or angry driver to accidentally slam into an oncoming car because they didn't pay attention to, or simply disregarded, a red light. (This happens all the time in LA.)

Where buses do make sense is, as has been pointed out in this thread, where bus lines act as augments to train lines and other forms of transit. Even so, it's a bit much to refer to something that can add hours to what ought to be a short trip as "advantageous."

Posted by Ross Lincoln | June 27, 2008 12:53 PM

Buses only have advantages over trains when the options are "take the bus," and "walk because there's no train going where you are headed." Not only do they have to stop with (frequently crowded) traffic, but they also make frequent stops at, um, Bus Stops. As a result, a trip that may take 20 minutes by car and maybe 10 by train can take an hour or more by Bus. Unless you have no absolutely no alternative, or you simply like being on a sweaty, crowded slow moving monstrosity, riding the bus makes almost no sense.

Trains go fast, don't stop with traffic, are far more comfortable, and lets' face it, more aesthetically pleasing. And in addition, while they do have more up front costs, long run they're just cheaper than maintaining a massive fleet of buses. Also, they have the added advantage of being impossible for an incompetant or angry driver to accidentally slam into an oncoming car because they didn't pay attention to, or simply disregarded, a red light. (This happens all the time in LA.)

Where buses do make sense is, as has been pointed out in this thread, where bus lines act as augments to train lines and other forms of transit. Even so, it's a bit much to refer to something that can add hours to what ought to be a short trip as "advantageous."

Posted by Ross Lincoln | June 27, 2008 12:59 PM

Not only do they have to stop with (frequently crowded) traffic, but they also make frequent stops at, um, Bus Stops.

Of course, this is taking a common type of bus service and generalizing. In Europe, one of the successful approaches to increasing public transport patronage using buses is a mode of service halfway between ordinary city buses and Bus Rapid Transit that they often call "Quality" buses ... limited stops service with strategic investments in bus priority measures at bottlenecks along the route.

BRT rarely makes sense at the capital/labor costs of high-income nations, with the patronage levels that can justify building them almost always also supporting the establishment of light rail services ... but Quality Bus routes can be established for levels of patronage that do not justify establishing a light rail line.

Posted by BruceMcF | June 27, 2008 1:11 PM

Youse guys is missin' da humor by not clickin' on da link.

It turns out the 'transit oriented development' here refers to the actual building of a bus station- or at least, that's what we used to call them, 50 years ago. In a move that seems almost radical to suburban Virgina, the transit authority has provided space for the buses to pull off the road, toilets and shelter from the rain for riders, and a kiosk with an actual attendant to hand out bus schedules (some restrictions may apply). And, as suits the flexibility and low start-up costs of buses, all of this (we learn from comments) was done in a mere ten years!

Yes, they managed to build a bus station in two years less than Seattle will have built a 15-mile double-tracked light rail system from the city center to the airport.

Matt lives on wells of parody and satire too deep for the average reader to plumb.

Now, let's talk a minute about 'transit oriented development'. What is it? Development spurred by the provision of transit. How do you do it? You provide, in a coordinated approach, a serious upgrade of transit with a believable commitment to stay, coordinated regional funding for public development based on the transit line, and rezones to facilitate new private investment based on the transit line. Does it work? Again, we need only look at Seattle, where development is happening along the transit line that has yet to open. Developers aren't reading tea leaves here, they're just looking at several billion invested in a new rail line that is currently in the testing operations, with cars running up and down the line.

Do you see any of that in Shirlington? No, you don't. The transit agency bought some cheap land outside of the shopping district and built a bus station. With luck, they might add a feeder bus to the stores and library.

If this is the kind of forward thinking we have to try out before we implement the proven modalities of trams and light rail, you can have it.

Matt likes to think that the whiz kids of his cohort can bring "rigorous and creative thinking" to any problem, little realizing that for about 120 years omnibus, streetcar, and motor bus operators have also been attempting to do just that. He might more profitably ask why after all this time it is still necessary to look for 'gee whiz' solutions. The odds would seem to favor some option from the past, implemented with more vigor than rigor, producing solutions.

Posted by serial catowner | June 27, 2008 1:31 PM

It turns out the 'transit oriented development' here refers to the actual building of a bus station- or at least, that's what we used to call them, 50 years ago. In a move that seems almost radical to suburban Virgina, the transit authority has provided space for the buses to pull off the road, toilets and shelter from the rain for riders, and a kiosk with an actual attendant to hand out bus schedules (some restrictions may apply).

Dude, take your own advice ... read the link more carefully.

an indoor waiting area, outdoor plaza and Commuter Store

From what it sounds like, that aint just a BUS STATION. That is, for all intents and purposes, a TRAIN TRANSFER STATION WITHOUT A TRAIN.

Now, living in outer suburban Ohio, obviously I don't get to see any actual train stations at the moment (nor any actual bus stations ... there's a main bus route and a local circulator, but nothing that involves a bus station). But I did live in Australia for a decade, and its the "commuter store" that really marks that out as a train station.

So now all they have to do is to include that bus transfer station on an Aerobus line through to link with the metro, and, voila, they have an integrated bus / light rail system.

At the rate they are going, I'm thinking 2028.

Posted by BruceMcF | June 27, 2008 2:02 PM

BruceMcF,

Just to clarify, I wasn't saying bus-train feeder systems never make sense. As with everything else, which approach makes sense depends on the circumstances (indeed, in Australia I believe they have busway systems in at least Sydney and Brisbane).

Posted by DTM | June 27, 2008 2:34 PM

With luck, they might add a feeder bus to the stores and library.

Huh? I dunno where this facility is, exactly (I intend to swing by tomorrow and take a look), but it can't be much more than a quarter mile from the furthest reaches of the development.

Posted by roac | June 27, 2008 2:40 PM

Ha ha, no I did not miss that. Of course, you might think a Commuter Store was where you go to buy commuters, or where commuters go to buy umbrellas, but if you check the hip with-it transit lingo of today, you find a Commuter Store is where you go to get...bus schedules. Maybe even a baseball cap with the transit agency logo, or a t-shirt or bumper sticker.

We have Commuter Stores out west, too.

Posted by serial catowner | June 27, 2008 2:41 PM

What people buy at a Commuter Store are SmartTrip cards, which is what Metro calls the plastic cards with the transponder inside that collect and dole out your fares. (Don't leave home without it!) Also the various long-term discounted passes for the elderly and disabled.

Posted by roac | June 27, 2008 2:54 PM

While people are mentioning Australia, can I at least suggest looking at the Adelaide O Bahn. Right of way separated from traffic. I'm not sure we need the fancy sunk rail-track-lookalike, but it's possible to build a busway.

Posted by Sherman Dorn | June 27, 2008 3:52 PM

David Alexander,

You list many supposed advantages of rail and none of its disadvantages. In a few places, rail transit makes sense instead of buses, or as a complement to buses. But most or all of those places already have it. Something like a third of all rail transit travel in the United States takes place in just one state--New York. New small-scale urban light-rail systems may be built in response to population increases in some cities, but there'll never be another major new subway or urban heavy-rail system in the United States. The population density to support it just doesn't exist. The most likely long-term future trend is a continuation of the current one, which is the gradual loss of transit market share over time to private motor vehicles as old dense transit-oriented cities continue to lose population to new sprawling car-oriented ones.

Posted by Mixner | June 27, 2008 4:22 PM

Buses have some substantial advantages over rail as a transit modality, notably being cheaper,

Cheaper? This may be true of the original capital cost (and only maybe, because buses wear down and need to be replaced much faster than light or heavy rail).

But not for operating costs. Here in Toronto, the new streetcars will hold upwards of 5-6 times as many people as buses. This means 1/5 as many drivers (and vehicles) to carry the same number of people.

Posted by McKingford | June 27, 2008 4:59 PM

Actually,
Buses are very effecient and can be if done right very fast.
Key facets include dedicated roads/lanes, double + length buses and stations that are designed at bus level for swift loading and unloading.
It is already being done in some cities, even in America.
It is called BRT. Some of the earlier adopters were cities in South America like Bogata and Curitaba, whose ex-mayor James Lerner is an urban visionary.

Posted by nama | June 27, 2008 7:29 PM

Well, actually...from the AP wire, June 19:

"And there are some early indications that homes near urban centers, and subway, train and bus stops are often selling faster and at better prices than those in the distant suburbs.

On Wednesday, a survey of 900 Coldwell Banker agents showed a remarkable 96 percent said that rising gas prices were a concern to their clients, and 78 percent said higher fuel costs are increasing their desire for city living."

It's almost like Mixner was frozen in 1967 and has just been thawed out.

Posted by serial catowner | June 27, 2008 7:31 PM

It's almost like Mixner was frozen in 1967 and has just been thawed out.

Whereas you were frozen circa 1900, have never been thawed out, and in your cryogenic stupor imagine that Americans today still want to live in small houses and get around on buses like they did then.

While the AP report is interesting, you really need to put it in perspective. I suggest you take a look at the long-term trends in transit use and population change in urban centers.

Posted by Mixner | June 27, 2008 8:52 PM

More accurately, Mixner was unfrozen in 1967 then refrozen in 2000, because he refuses to recognize any trends contrary to his claims that have occurred after 2000.

Posted by DTM | June 27, 2008 11:17 PM

Sherman Dorn | June 27, 2008 3:52 PM

While people are mentioning Australia, can I at least suggest looking at the Adelaide O Bahn. Right of way separated from traffic. I'm not sure we need the fancy sunk rail-track-lookalike, but it's possible to build a busway.

The Adelaide system is an interesting one ... it definitely outperforms regular city buses in terms of passenger appeal, as when Adelaide privatized its public transport about two decades back, it was the O-Bahn that maintained its patronage, while patronage on the rest of the system decline.

It is, of course, a hybrid system, with buses running onto the busway from the city streets (and every once in a long while a car wanders into the busway), so while the O-bahn portion of the route may service light-rail patronage levels of 20,000+ trips a day, the individual bus routes sharing the O-bahn seem likely to be at bus patronage levels of under 10,000 trips per day each.

OTOH, it has not seemed to have been any easier to extend the O-bahn than it is to extend rail lines, and in the event it was the tramline (light rail) that Adelaide decided to extend.

Certainly in terms of the ease of extending a line through an already built up area, an Aerobus system would have the edge over something like the O-Bahn.

Posted by BruceMcF | June 27, 2008 11:22 PM

DTM,

because he refuses to recognize any trends contrary to his claims that have occurred after 2000.

Right. Like the trend of declining population in Pittsburgh and other rust-belt cities that I recently brought to your attention and that you refuse to recognize.

Posted by Mixner | June 27, 2008 11:40 PM

Mixner,

Good illustration. The best data available on that subject contradicts your claims.

Posted by DTM | June 27, 2008 11:48 PM

DTM,

Good illustration. The best data available on that subject contradicts your claims.

Official population figures are provided by the Census Bureau's Population Estimates Program. Here's the data for Pennsylvania.

Here's Pittsburgh sorry decline:

2000: 333,804
2001: 330,576
2002: 327,458
2003: 325,075
2004: 320,402
2005: 316,299
2006: 312,819

And here's Philadelphia's:

2000: 1,513,655
2001: 1,497,897
2002: 1,485,249
2003: 1,473,364
2004: 1,465,475
2005: 1,456,350
2006: 1,448,394

Notice the trend? All the other cities you claimed had grown are also in decline. Read it and weep, DTM.

Posted by Mixner | June 28, 2008 12:11 AM

Mixner,

As I explained to you before, the best population data available for these cities comes from the ACS, because there is an invalid assumption underlying the data you are citing.

Posted by DTM | June 28, 2008 12:47 AM

All swans are white. Therefore, the black swans reported in the Antipodes are clearly the product of a delusional mind.

Posted by Great-Great-Great Granpa Mixner, c. 1790 | June 28, 2008 1:13 AM

DTM,

As I explained to you before, the best population data available for these cities comes from the ACS,

No it does not. The Census Bureau specifically warns against using the ACS population estimates for years prior to 2006 because they are incomplete:

Persons in group quarters are included with persons in housing units (HUs) in all 2006 ACS estimates based on the total population. All ACS population estimates from previous years include only persons in housing units.

The Census Bureau's official population data is provided by the Population Estimates Program:

Although the American Community Survey (ACS) produces population, demographic and housing unit estimates, it is the Census Bureau's Population Estimates Program that produces and disseminates the official estimates of the population for the nation, states, counties, cities and towns and estimates of housing units for states and counties.

The Census Bureau's official population estimates from the Population Estimates Program contradict your claims for every one of the cities you cited. All of those cities lost population between 2000 and 2006.

Posted by Mixner | June 28, 2008 2:23 AM

Bruce,

I'm not sure that "extending the line" is the best way to judge an individual BRT choice (such as a busway). If a set of routes gets people off the roads, it can be worth it even if extending the system is difficult.

Posted by Sherman Dorn | June 29, 2008 10:36 AM