An excellent article by Meena Kadri at the Design Observer Group offers a tour of the new Bus Rapid Transit (BRT) system in Ahmedabad, India. While North Americans argue about whether Bus Rapid Transit can substitute for rail, the developing world is building a lot of both. India, though, made an early commitment to BRT for its second-tier cities and corridors.
Indian public policy is always intensely focused on the country’s issues of poverty and economic opportunity. So many people need improved mobility that the government is strongly encouraged to choose whatever tool will create the most urban mobility for the least cost. So while a rail transit program is underway for the busiest corridors in the biggest cities — where capacity needs require it — a massive investment is being made in Bus Rapid Transit.
What’s most encouraging is that they’re taking the speed and reliability needs seriously. An early project in Pune was too compromised, as often occurs when you try to take away street capacity for a project that nobody understands yet. Pune’s facility was only partially exclusive, so the ferocious scouring force of Indian traffic ultimately carried the day and defeated much of the project’s purpose.
I worked for a bit on a BRT project in Visakhapatnam in 2008, and it raised fascinating issues. When we first toured the proposed corridor, our first reaction was “Whoa, this is way too narrow! Where are you going to put BRT here?” Lined with street vendors, as many Indian boulevards are, the street seemed to have only space for four narrow lanes of traffic. (“Lane,” of course, is an abstract concept in India, where vehicles vary dramatically in width and thus find different sorts of lanelike spaces to suit their need.) But our guides pointed out a wall of permanent buildings lying well behind the vendors. That, they explained, is the edge of the street right of way. All the street vendors are engaged in “encroachment.” The word may sound bureaucratic but it nicely captures the barnacle-like gradualness and persistence with which life will press into any available space. The edge of the Indian street is a dynamic negotiation between the needs of transport and the needs of commerce. Pedestrians, of course, must walk along the battle-line of this enduring conflict.
So one thing the BRT program was going to do is sweep all these merchants out of the way and restore the original width of the street. (The merchants would be looked after, however; considerable budgets are set aside to manage “social impacts” from capital projects.) But then we had to consider how the entire street could operate with BRT down the middle. How would we manage access? Would pedestrians flow across the boulevard at random, or would we try — against all the natural forces of the Indian street — to force them to use signalised crosswaks?
We also had to decide on raised platforms, with high-floored buses, or low platforms with low-floor buses. I’ve always preferred the latter, because they can run through onto non-busway segments and because a high-floored bus is hard to evacuate if you have an emergency away from a station. But South American experts have enthusiastically pushed the high-floor option. It’s cheaper, and this really matters in India. It also implies high-floor stations that are a bit more visually prominent and thus a little less vulnerable to commercial “encroachment.”
If you share the North American activist view that BRT is just a front for people who want to build more roads, a Indian BRT project will probably push that button. It was clear, for example, that while city leaders were still not sold on BRT — it was funded and promoted by central government — they really liked the idea of reclaiming the full width of their street. This would create room for BRT plus two lanes of traffic, and would also create space for a “slow vehicle lane” for low-speed one-gear bicycles and human powered rickshaws. Safety, of course, was the rallying cry. Indian streets can be really dangerous, especially for exposed humans, so we faced the usual irony: A street that separated pedestrians from cars would definitely improve the safety of pedestrians and other pedestrian-speed traffic, even though it would mean the cars would be able to drive faster. The cruel fact of life is that pedestrians are a form of traffic calming. Motorists really try hard not to hit them, and this keeps them from driving as fast as they otherwise would.
One of the most interesting challenges was encouraging our Indian clients not to buy too much technology. Indians are usually good at figuring out the cheapest way to do things, but like almost everyone they do like technology for its own sake. I tried at one point to talk them out of buying automatic ticket machines. Labor is so cheap that every bus has a conductor as well as a driver, so there’s just no need for them.
Indian BRT is an interesting example of a point that I’ve often raised in bus-rail discussions. Why exactly should we spend more on a nicer, more comfortable technology when we can provide the same mobility with a cheaper one — and thus extend the product further, into more territory, serving more people? North American readers often resist the obvious point that a cheaper unit cost means you can build a larger network, partly because they doubt that cost savings would really be spent that way.
In India it’s much clearer. You have a billion people and not much money. So build big, extensive networks, by choosing the lowest-cost product that meets your standards for civilized space. The Ahmedabad BRT does have new buses — not fancy but functional, with the same next-stop information (visual and announced) that you’d expect in Europe. The BRT lane seems to be sufficiently protected from disruption. But it’s still cheap, and that means it can take you further than a more expensive technology would do. Yes, BRT has much worse emissions than rail, and eventually some sort of carbon trading system will make that problem bite. But right now, Indian planners are looking at the advent of the new Tata Nano, a $2000 car designed to be affordable to the poorer families who currently get around, dangerously, on motorbikes. The cities don’t have room for all those cars. So there’s some urgency to getting good transit built in India, and on that score their priorities make sense.
First and last photos: Meena Kadri, Design Observer Group
North American transit advocates tend to hold cynical views towards BRT because in most cases, it is a compromise (often a political one), and not an optimum solution. Take the example of Los Angeles… the Orange line BRT was originally proposed as heavy rail surface running extension of the Red line subway. That gave way to a light rail proposal, which eventually resulted in the compromised BRT. Great… except it is already at maximum capacity only a few years after operation with no option to scale up the capacity. Another shortsighted application of BRT for the sake of political expediency rather than coherent long term planning.
That’s not to say that BRT has no place in North America… It’s just that all to often, it is billed as an alternative to rail based transit when it really is a compliment to the system. For example, in the case of Los Angeles Orange line mentioned earlier, the North-South extension that is currently under construction fits the BRT mode and design perfectly to feed into an East-West rail line… except that East-West line is a BRT already operating at 110% capacity… The North-South extension will add more demand to the East-West line but at last, we haven’t found the money to upgrade that to rail.
The Orange Line is complicated by the fact that a light-rail line in the corridor is actually illegal under California state law. Some Valley politicians, in an anti-transit pique some years back, got a provision added to a bill in the California legislature actually banning light rail on the right-of-way.
Jarrett, what you say about the cost of bus versus rail has an additional complication in developing countries – namely, the difference between labor and technology cost. Once the technology for subway construction is routinized, it’s just a matter of paying people to dig, and that can be really cheap if wages are low. This is especially true in a country where growth is high, because then there’s a lot of money lying around for infrastructure and it’s better to build something more expensive now to save on labor costs later.
I think that this is partly what explains mode preferences in Latin America versus East Asia. In rapidly growing East Asia, there’s a lot of money for investment, so there’s often no point building cheaper BRT: in a few years its higher operating costs are going to be brutal, and so will the cost of building a subway to replace it. Latin America is growing much more slowly and is capital-starved, so the considerations are different.
Alon. Yes, that’s true up to a point.
Another complexity is that sometimes developing countries receive technology donations that push their mode choice a particular way. The Delhi Metro, for example, contains signs thanking the Japanese government for its donations of technology and expertise to the system’s development. These can be quite a logical way for a technology-selling country to establish dependence on its product.
The fact that there’s a huge rail transit boom in Asia shouldn’t obscure the fact that there’s a huge BRT boom as well. Like India, China has a huge BRT development plan focused on second-tier corridors and cities.
It’s so irritating to have to keep fighting with mode warriors. The reasoned answer is always that each has it’s place. The challenge for professionals is to provide some guidance about where each application is appropriate, and not to play along with the notion that any mode is superior in all applications.
Just like the availability of capital funding tends to drive mode choices in developing countries, it seems to me that the financial bubbles we’ve lived through in at least this developing country (US for me) have caused transit advocates to believe there is no budget limit that can’t be crossed. If at some point we start accepting budget limits, we’ll see that rail should be used only where activities are most dense and volumes and frequencies are highest – but that will leave a ton of other places where we still want a better level of transit service than the local bus.
And of course there’s the big obvious difference between somewhere like India and somewhere like the US. India has lots of cheap labor, but not a lot of capital, while the US has reasonable amounts of capital, but expensive labor, which really makes a difference in the relative cost-effectiveness of rail and BRT. There’s also the fact that even at $2000, the Tata Nano is probably less affordable than car-ownership in the US, and there’s much less car infrastructure, so BRT systems there don’t have to compete as hard against the car.
“I tried at one point to talk them out of buying automatic ticket machines. Labor is so cheap that every bus has a conductor as well as a driver, so there’s just no need for them.”
That’s why BRT is great for countries with low labor costs.
If we could get away with paying bus drivers $16 per hour here, no benefits, no overtime, no work rules, you could run a bus for $30 an hour. But right now Los Angeles Metro spends $120 per revenue hour, 80% for labor costs. Hence the strong advantage of higher capacity transit and automated rail. Meanwhile, India can pay $2 per hour; most of the cost of running the bus is in vehicles and fuel, not labor costs, so there is little benefit in replacing 10 buses with 1 train. This calculation may change in 10 or 20 years.
I think it’s really difficult to compare BRT in a situation like India to North America. Whereas in India building a house that isn’t up to north american code would be reasonable, would we accept it in North America? I’m not saying this to be classist or to suggest that somehow they don’t deserve the same things North Americans have, but transit is competing against something different in India than it is in North America, and the reasons why it’s important are different, too. Indian’s need improved mobility, safety, quality of life, and competing against bicycles, small engine vehicles, etc., but North Americans aren’t going to clamour to a BRT when it’s competing with cars and the issue isn’t so much mobility, but reliability, comfort, and ease of use. As well, in the developed world the added goal of being green is really important considering we produce such a high proportion of the greenhouse gases causing the problems, and that’s a large part of why we need to convince people currently driving Hummers to switch. I’m not saying India shouldn’t strive for environmental friendlyness, but building transit will prevent a lot of those emissions from cars from ever starting, as eventually cars will become available.
I don’t know if it was intended, but your comments seemed to attempt to argue that the Indian example shows that North American transit activists’ concerns are ungrounded in facts, when I strongly disagree, and I think the concern that money saved by using bus instead of rail won’t be spent on other transit initiatives is a valid and real concern. These two areas are starting from different points, and there are different goals, and I think that’s really important to recognize.
There’s more to low labor costs for BRT in poor cities than a simple economic calculation.
Labor costs are low because of the desperation of the general population. In the U.S., we’re feeling the agony of 10 percent unemployment. In a developing city, that’s the percentage of the population engaged in steady, productive work. And that’s a best-case scenario.
Poor areas desperately want to take any activity that can absorb people into productive work. A train line has more than the problem of high capital costs. If a train operator in a poor area can do the work of 5 bus drivers, the economy has little to offer for the newly redundant workers.
Labor-intensive bus operations in poor areas don’t build wealth through wages. They build stable work, which is needed as a first step on the wealth-building process.
I enjoyed reading your article. Have you been to SA lately? What are your views on our BRT system?
bzcat wrote: “the Orange line BRT . . . is already at maximum capacity only a few years after operation with no option to scale up the capacity.”
I don’t think this is correct, there are affordable ways of increasing the capacity of the Orange line. The most obvious is to run groups (or ‘platoons’) of buses together. This would probably require expanding the station platforms, but that is not so expensive. The other option is to go to longer buses, such as 80 foot double articulated buses. Run platoons of three 80 foot buses, and you have a major increase in capacity.
In contrast, here in Vancouver Canada we just opened a $2 billion light metro which is at capacity after less than a year. The station platforms are mainly underground or elevated so making them longer would be very expensive. And there is an elevated single track section that makes increasing the frequency of trains very challenging.
In comparison, increasing the capacity of a bus rapid transit or light rail line would be a breeze.
[Note, I am not saying that BRT was the best choice for the Orange Line route. It was once a rail line so probably did not have major utilities buried underneath – the potential to avoid re-locating utilities seems to be one of the most important cost saving aspects of BRT.]
Why would a platoon of busses, assuming each bus has its own driver (busses cannot safely be entrained, after all), be more reliable or efficient that busses operated independently of each other? With separate drivers, the formation may be difficult to maintain; and if a platoon of busses all enter a platform and one is done boarding before the rest, do all busses wait for the last to close its doors?
And given that the Orange Line doesn’t have signal pre-emption, how does one ensure that a platoon isn’t broken up at a crossing by the light turning red?
The LACMTA has decided to use a different solution in this case: they run an express bus down Burbank Blvd from Van Nuys Blvd to North Hollywood. This both provides a direct link to all the destinations on Van Nuys, and bypasses the busway entirely. Note that going by the scheduled timepoints, it’s not actually any slower, thus making one question what the point of the busway is in the first place.
@Tessa. I don’t think we disagree. I’m not saying that fears about what happens with the money BRT saves are unfounded in North America. But I want to suggest that this concern is a feature of the political moment rather than a settled fact of life.
I agree with all the comments on the Orange Line, and obesrve that it is profoundly compromised — in both speed and capacity — by the lack of signal priority at its frequent intersections. Having said that, remember that bus capacity is about stations rather than stops. There’s no reason you couldn’t expand capacity by, say, going to an AB stopping pattern or developing an overlay express pattern that makes fewer stops.
Yes, but stops for a traffic light impose a penalty, just as stops to pick up passengers do. The stop penalty for a traffic light is probably a greater factor for a short-dwell-time-optimized BRT such as the Orange Line than it would be for a local bus line, no?
The beauty of BRT in North American cities is that it reclaims space from the automobile. Taking space away from street vendors, an activity that has traditionally been part of the streetscape for thousands of years for BRT or any other form of transportation is really counterproductive. Streets need to be used for more than just transportation. I would also doubt that moving pedestrians to the sidewalks would improve pedestrian safety. By increasing the speed of cars, I would expect pedestrian fatalities to rise dramatically. All this really seems like is motordom, disguised as BRT.
Actually, many BRT lines in North America, especially the faster and more effective ones, reclaim space from former rail lines. Examples include the El Monte Busway and Orange Line in LA, the East and West Busways in Pittsburgh, parts of the Ottawa Transitway system, and I’m sure there are plenty more. On the other hand, “Bus Rapid Transit” like the Rapid buses in LA or the 522 in Santa Clara County don’t take any space away from cars.
EngineerScotty wrote: “given that the Orange Line doesn’t have signal pre-emption, how does one ensure that a platoon isn’t broken up at a crossing by the light turning red?”
If the Orange Line does not have effective signal priority, then it is not fully bus rapid transit (Maybe you could say BrT with a small R). And yes, if it is not really bus rapid transit then it is difficult to keep platoons together reliably. Nobody suggested that platooning would make BRT more “reliable or efficient”, it is just one of several ways of increasing capacity.
As to the doors closing, does the front half of a metro train close its doors and leave before the back half? No. And that is not considered a disadvantage, it is just the way it is with rapid transit.
The Orange Line turned off signal priority as a safety measure. There have been a couple of accidents, and the buses have to go through the intersections slowly.
The Orange Line is very quick even without it. It’s about 45 minutes across the Valley. A local bus takes about 75 minutes to cover the same distance.
As for “platooning” buses to manage crowding, L.A. does that on high-frequency local bus lines as well. It’s called bunching, and it’s a telltale sign of sloppy operations management.
Platooning on the Orange Line would actually make sense, as would a zone-express service (skipping all stops to some point, then running local), because most of the ridership comes from the Red Line, which runs at exactly half the frequency of the Orange Line. So you already end up with one bus being much emptier than the one in front of it, and informal platooning, aka bunching.
Given the comments above, it seems like the Orange Line is a fairly crude form of BrT. And it also seems like there is little political will to make it into a more advanced BRT line.
If all it takes is a “couple of accidents” for the authorities to turn off the signal priority, it suggests that the automobile lobby is still in the drivers seat in LA.
If there are too many fatal crashes, (have there been any fatal crashes at all?) it means you need crossing gates or other enhanced safety measures, not that you need to slow transit to a crawl at every intersection. (Yes, greatly reduced speeds can be a good solution but only at particularly tricky intersections).
PS. Wad – Are you trying to say that platooning should never be used to increase the capacity of a BRT line? If so, please explain why you think this.
How do you provide a positive detection system good enough for crossing gates with buses? You can’t use track circuits or axle counters like trains do.
Lots of ways to provide signal priority for busses in the absence of track circuits:
1) Use a system similar to that used to provide signal priority to fire engines in cities throughout the world. It works pretty well.
2) Use video-camera based vehicle detection. This also works well, and is rapidly replacing magnetic loop sensors in many places.
3) As a fallback, there should still be red/green lights for the bus, and the bus drivers should be trained to stop at reds like normal–in case the sensors DO fail to stop cross-traffic, or a higher-priority vehicle (such as a fire engine) happens to be coming the other way. (Here in the Portland area, a MAX train once struck a fire engine when the train diver thought that he couldn’t be pre-empted at an intersection; fortunately, the accident was nowhere near as serious as it could have been).
Crossing gates are probably not necessary; they’re necessary for rail crossings because trains AREN’T able to stop. Granting signal priority to busses shouldn’t require such extra-ordinary measures.
I’m not talking about signal priority, I’m talking about the proposals for crossing gates that some people like to make occasionally. The signal priority stuff you mention is reasonable enough and ought to have been built as part of the busway. In fact, signal preemption on many rail lines is actually triggered by road-style loop sensors rather than track circuits, and you can activate it with a bicycle.
You have a lot of ideas but this is the reality of Orange line in LA… we are already at capacity during peak hours in terms of number of buses. There is simply no room to add any more buses or “platoon” them because it will result in a line of endless uninterrupted buses from Warner Center to the North Hollywood subway station. Yes, that is a bit of exaggeration but it is not far from the truth.
California law also cap the length of buses at 60ft so 80ft buses is not a feasible option. Metro already got an exemption to run 65ft buses and that is the biggest bus we are likely to get in the near future. But the size of the bus is not the point here… it is the physical capacity of the right of way that is the issue here.
You wrote: “There is simply no room to add any more buses or ‘platoon’ them”
BRT lines routinely carry over 10,000 pphpd, is the Orange line already above this level? If I remember correctly one Chilean BRT line was carrying about 25,000 pphpd before a parallel BRT line was built to improve service quality and speed; is the Orange line getting anywhere close to this extreme and undesirable level for a 2-lane BRT line?
PS – If the size of the vehicle had no influence on capacity, then going to light rail would likewise not increase capacity.
Going from a 60 foot to an 80 foot vehicle increases capacity by 33%. Going from a 60 foot bus to a 270 foot train increases capacity by 350%. That’s why the 80 foot bus is not particularly relevant.
About three years ago I proposed a solution to alter the present Orange Line into a busway system similar to Miami’s.
Instead of platooning buses, the Orange Line would be replaced by shorter (45 feet vs. 60 feet) buses that would also operate low-frequency local north-south services.
The other benefit here is that “The Miami Option” saves riders a transfer. Very few trips originate or end near an Orange Line station. A transfer to a north-south bus is required.