Friday, December 10, 2010

111> Capacity

For some time now I have advocated designing PRT for more than strictly CBD (Central Business District) use. Even though the “flight to the suburbs” in the U.S. has led to the need for urban renewal, and PRT would be a terrific tool for just that, the suburbs and highway fed satellite communities already exist, and need service too.

Companies wishing to get a PRT product to market do not have infinite time and development dollars, and their efforts necessarily must start from the basic building blocks of stations and loops. There is little point in promoting a vision that they are not yet prepared to follow though on. Unfortunately, a scaled-back, slow version of PRT is a lot like a scaled-back, slow version of the Internet. Eighty people being able to dial up a dozen sites won’t exactly make you thunderstruck by the possibilities. Yet that is a stage that was, at one point, a future vision. 

So here’s a little peak at a PRT “trunk” line, which, of course, can only happen when there are massive networks on either end to feed it. In this picture there are four tracks going in the same direction. Reversible lanes would offer up to five possible configurations, direction-wise.   

Here are some sample numbers. With a one second headway, and vehicles traveling at 60 mph, (88 ft. spacing) and the U.S. average of 1.2 passengers per vehicle, the capacity would be 17, 280 passengers per hour. Not bad for fitting over a strip of grass no wider than a residential driveway! With vehicles that can swing forward from sudden deceleration and bumpered bogies that clamp the track for extra stopping power in an emergency, still shorter headways and higher speeds can be expected without much technological challenge, especially if platoon strategies are employed. I think what is informative here is the capacity vs. the minor amount of steel and concrete needed to achieve it. By the way, with rubber wheels inside of a sound-insulated track casing, the system would be nearly silent.

It occurs to me that when I invent, I frequently start at an endpoint or finished product and work backwards toward the present. In a world of cause and effect, one can start at the desired effect and try to envision the causes that could create it.  This is why I have been advocating a standards-based open-source PRT solution. Can you imagine a single company ever being able to successfully scale up to this kind of volume? I can’t. The only way I can see it is if the track is a simple and standardized design, buildable by local firms, the control system is infinitely extensible, and multiple manufacturers can compete for business with certified, standards compliant vehicles. The funding needs to come out of the highway budget, and there needs to be a non-profit organization specifically tasked with organizing and fostering the public/private partnerships to make the whole thing happen, as well as developing and maintaining the standards. 

I do not rule out one company building a starter network and growing substantially from there. I just do not think that the design choices that are best for the company’s shareholders in the short or medium term are the best choices for PRT as a long-term technology. PRT can have extremely positive effects on society and the planet and still be the basis for very profitable businesses, but there is no reason to believe that a system designed to be profitable and saleable today will resemble the design that has the most transformative potential. In particular, I think current designs have traded away flexibility in a rush to have a market-ready product. These designs have limitations in terms of speed, station layout, turning radius, scalability, safety, track pitch, and a host of other issues. Once deployed, these limitations become set. If they later become burdensome, there is not necessarily any way to remedy the situation while maintaining compatibility with the legacy system and its installed infrastructure. 

P.S.   I have added a Table of Contents. Now you can enjoy one-click access to all 110 posts in the archive! Also, there have been no malware alerts connected with this site per se; The one reader who has reported a problem apparently only gets a warning from the use of his “Google Alerts” redirect function to this site and not the site itself, so I’m not going to worry about it.


Andrew F said...

Hey Dan (et al)

On the topic of embryonic networks, it sounds like two cities in India are seriously kicking the tires on an ULTra-based network. ULTra has partnered with an infrastructure firm to propose two systems. One is 3.2 km and seven stations linking the train station and a popular tourist/religious attraction costed at 5 billion rupees ($112 million). The other, larger system is 105 km, 143 stations and 3000 vehicles for 50 billion rupees ($1.1 billion). These figures assume the local governments provide ROWs for guideway and land for stations.

More info here:

Andrew F said...

Here is the site for the partner:

Dan said...

Thanks, Andrew. I always appreciate links. Speaking of ULTra, they are saying on their website that they have achieved a throughput of 48 passengers from a 4 berth station in 5 minutes, which seems unbelievable to me. I wonder how many vehicles that was. Were the vehicles carrying only one passenger each, that works out to a car leaving every 6.25 seconds, or a 25 second turnaround for each berth. Come to think of it, I guess I don’t know what the capacity of the ULTra system is; it seems like the track would fill up with relatively few stations.

Bruce said...

Hi Dan, Where you say that a PRT trunk line would need "massive networks on either end to feed it", I say not necessarily. It would be viable if there were high capacity stations at each end, instead. In that scenario, it would be functioning like a commuter rail line, but offering a more flexible service. Once the trunk line is in place, a network could be built around each station.

Regarding the operating speed, 60 mph is good enough to compete with commuter rail, but a somewhat higher speed is needed if people are to be attracted out of their cars in numbers.

I think PRT should aim to take cars off highways and compete with intercity rail. Therefore, target operating speeds of 90 mph or higher on "trunk" lines.

Andrew F said...

Dan, they have a video showing that capacity test. It was for four-person occupancy with one passenger in each car a system 'expert' to facilitate the process. It doesn't seem like the average user would need an expert. None of the participants carried luggage either.

Bruce, higher speeds are nice, but they come at a cost. Dan has some posts on the topic. Your vehicles need to be capable of the speed. Higher speed means higher RPMs on your wheels, or a higher guideway dimension to accomodate larger wheels. A higher guideway increases the visual impact and limits the ability for guideways to be accomodated in buildings between standard floor heights. Maybe that trade-off is worthwhile.

I'd consider 60 mph to be pretty decent. That would put PRT at a higher average speed in most cities than any other mode of travel, including by car. Faster would be nice, but it may come at a price that could hamper adoption of the technology.

Bruce said...

Andrew F, looking at how much governments are willing to spend for faster trains, and how much individuals are willing to spend for faster cars, I think extra cost for higher speed is worthwhile.

In the UK, a lot of intercity rail operates at 125 mph, though average speed over route is more like 90 mph. On the motorways, the speed limit is officially 70 mph, but quite a lot of drivers routinely exceed that by 10 mph or more. In France, the speed limit on autoroutes is 81 mph; in Germany most autobahns have no speed limit. The trains are faster, too.

So, to compete with car and rail on interurban routes, the ability to offer a high operating speed is essential.

You wouldn't necessarily operate at that speed on all routes, but if the system has a lower speed limit built in, its usefulness will be limited to a few niches.

Conversely, if you can go faster than motorway speeds, you can not only draw people out of their cars and compete with intercity rail, you can also take this service to places that previously could not be served by trains, owing to insufficient peak traffic demand.

Andrew F said...

PRT isn't about interurban travel or long trips. PRT won't have lavatories, so that puts a certain limit on trip length without having 'rest stations'. High speed intercity PRT would be great, but if we impose those design criteria on PRT, with concomitant costs, we're going to compromise adoption of PRT as a ubiquitous urban network. This goal is far more important, as we already have pretty good interurban options, and people make interurban trips fairly infrequently. If people have to transfer from the local PRT network, with its slimmer, less expensive guideways and vehicles, to a high speed interurban network, that would not be the end of the world.

I don't know how much you've been reading of Dan's design philosophy, but I think he's on the right track. He advocates inexpensive guideway and stations (though more elaborate, expensive stations are possible where warranted) as a means of ensuring that the largest area possible could be economically served by PRT, with stations that are numerous and close to origins and destinations. High speed is nice, but incremental speed increases need to be weighed against this goal of inexpensive, flexible guideways and vehicles.

You're quoting unfettered highway speeds. Look at average car speeds in most major cities. The vast majority are far below the posted speed limit, usually less than 50% of the limit, in fact. A PRT that could cruise at 60 mph would be able to achieve an average speed pretty close to that. In the 50s mph certainly. And that would make it head and shoulders faster than any other mode of travel in a city. As an example, average speeds in London, UK is under 10 mph. Subway is not much more than 20 mph on average. So, you can see how something capable of 50 mph on average would be a totally disruptive technology in cities, where the majority of us do the majority of our travel.

Competing against cars, planes and trains on their turf, where they perform well, isn't a good strategy, at least at first.

Dan said...

Dan the Blogger Responds-
Thanks, gentlemen, for making the case for flexibility! Andrew, thanks for clearing up that ULTra thing. Bruce, the idea making longer hauls is one that is little heard, and deserves more airtime than it gets.
Most cities I visit need a network of 4-8 mile trunk lines that connect +/- one-mile- diameter, 2-4 station loops. That is because most cities have absorbed numbers of townships that are about that far apart. Spotty development means trunk lines. How this manifests in a limited, initial route design is a case-by-case thing. The components that I am designing strive to do both; block-by-block grids and high speed. There is nothing in the track design that holds the speed under 110 mph, but I also have gone to great lengths to make sure it can reach everywhere, such as very constricted corners, narrow easements or inside of buildings. The vehicles, I think, will evolve to meet the challenges of the layouts that city planners devise. It’s more important to make sure that high-speed vehicles can make it into downtown than to make sure that all vehicles must be capable of the highest speeds, at least at first.
Sixty was primarily a speed of convenience, mathematically. (A mile per minute) It is also a bit of a natural dividing line, though, between high-speed and urban, because after about that speed, wind resistance starts multiplying really fast. Motors sized for zippy acceleration and steep slopes around town will top out around that speed, so making them all capable of very high speeds might be making them all carry extra cost and weight that they rarely need.
I agree that getting people to leave their cars farther from town is a very good goal and that many PRT benefits extend to longer haul service, including being light so it can be elevated to not block traffic, being private, not requiring a critical mass of riders to justify a vehicle, not stopping along the way. (except for that bathroom/food break!) Two mega-stations on either end of a trunk line, however, begs the question of what is so special about those places as to generate that much foot traffic. I agree that it is wasteful and unlikely in any event. Let’s at least see a 3 or 4-station loop on each end… Once they see the convenience of the first loops, they’ll be back for more!

qt said...

Dan, if I'm reading Bruce's post right, he's suggesting something like a park-ride lot, or an alternative to building a commuter rail line. It could make sense--on one end of the line, at least. The loop would probably be needed on the other end, though...

qt said...

Dan, if I'm reading Bruce's post right, he's suggesting something like a park-ride lot, or an alternative to building a commuter rail line. It could make sense--on one end of the line, at least. The loop would probably be needed on the other end, though...

Dan said...

The question is how many cars a minute would you be shooting for, traveling, say, a 15 mile stretch? Say it is one every couple seconds. Now your park&ride has got to be a very big affair, with thousands of spaces. Make that 4 park&rides,a Walmart and the airport, and you've got stations cumulatively matching track capacity a bit better, don't you think? (I do like the idea of "to your car" PRT pickup service at park&rides and long-term airport parking, BTW. Try THAT with light rail!)

Andrew F said...

I think the idea of park and ride misses the point. The goal should be for you never to get in your car in the first place.

Rick said...

The goal should be to make the service faster from door to door than any other mode of transportation. This can include private automobile legs for those not within a mile or so of a station if they are traveling to a well serviced area.

People will choose the fastest method most of the time given an option (they won't walk for 30 minutes to a station if they can drive door to door in less time). This will require Park & Rides.

PRT only makes sense if it can improve mobility (minimum time for an individual door to door)and is not a replacement for the automobile.

Andrew F said...

My point is that the vast majority of trips on any reasonably sized PRT system should not involve an automobile leg. If this is not so, you're doing it wrong, and the system is likely going to fail.

Park and rides are expensive. They might exist, but they should be few and far between. PRT isn't some commuter rail system. If people need to drive to it, you've done it wrong.

Dan said...

Dan The Blogger Weighs In...
Start with Houston. Metropolitan area is over 8000 square miles. The area within the city limits is 579 square miles. More people live outside the city limits (3.6 million) than in it. (2.3 million) The vast majority of people don’t live or work downtown, or even anywhere near it. By the way, we have 7 miles of light rail. Period. Most people are not even within an easy walk of a bus route.

PRT requires pedestrians to use it. There are probably only 6 or 8 square miles in total – out of 8000, that have the foot traffic to support it. And they are not contiguous. In other words, you have to drive to get to them. And why would you travel 5 or 25 miles by car to take the last couple of miles by PRT? You might as well just go all of the way. PRT, however, creates foot traffic, just like other forms of public transportation. If people can knock 10 miles off of their driving, and arrive into a grid of destinations, that would be an attractive proposition. They would then “feed” that grid, making it more viable, and hopefully creating a demand for more local loops. The dawn of a pedestrian friendly area!

I might note that not only is Toronto unique in the way it is hemmed in by parks; it is also pretty far north. I’ll wager you don’t have a lot of experience being forced to walk long distances to get to transit when the temperature is in the 100-degree (37 C) range. I mean, you guys can bundle up, but there’s only so much we can take off. People’s cars are their means of staying cool.

Let’s get a little closer to you. (Toronto) Say someone lives near Mississauga and works in Brampton. Or lives in Markham and works at the airport. Are you suggesting that a Toronto PRT grid should only reach these folks by growing block by block from the city center? Of course in being in Canada, you might already have decent public transportation options reaching these burgs, so maybe PRT links would be redundant. If these folks are car dependent, however, I would start with a single loop each in Mississauga, Brampton, Vaughn, Markham, Pickering, connecting to the downtown grid and the airport. My hypothetical guy lives NEAR Mississauga. (Not in it) He could park at one of the Mississauga stations and have a huge choice of destinations…After all, you can’t usually choose where you work. The single loops would then grow to figure 8s, then to small grids, and so fourth. Frankly, most people prefer to raise their kids where they can have a backyard, and PRT can’t reach both these neighborhoods and the places where people work all at once.

Dan said...

BTW, Don't get me wrong.. The RIGHT way to do it would be to have a whole master planned city that is so well thought out, clean, quiet, and convenient and that people could live and work in densely-populated environments without feeling claustrophobic. I used to design such cities for fun...Giant floating Terraced pyramids and such...

Ryan Baker said...

Park and ride is possible, but probably a spotty, very situational element. Without a doubt however, you have to have a actual PRT network before you build any PRT park-n-ride.

The dynamics of a PRT park-n-ride are fundamentally different from a commuter rail park-n-ride. In addition, anywhere that a commuter rail park-n-ride is practical.. there is probably already one, or a well developed effort to create one (along with the financial investment behind the campaign).

A PRT network however, even a modest one, does some of the work necessary to make feeder routes practical that are not practical given the existing systems. There is another component thought, and that is restriction of the existing systems, specifically highway lanes and inner city parking. In some cases these may already be sufficiently restricted, but usually not. And while people have successfully identified the link behind sprawl and traffic expansion related to additional highway lanes, much less thought has been given to the parking.

Even when talking to inner city residents, who rarely drive, and environmentalists, who would love to see less driving, I find there is relatively little opposition to expansion of parking, and almost none for restrictions. Parking is detrimental to a city's design in more than one way. Most significant in my mind is the consumption of space. Consumption of space increases costs for other uses, restricts density and lowers walkability. Parking also attracts more vehicles, which bring noise pollution, air pollution, dangers to pedestrians and cyclists, and eventually consumes more space through street widening. It also slows any bus based transit through increased congestion. Lastly, it lowers the number of transit users, which reduces revenues, lowers practical service route frequency and encourages commercial construction designed to cater to vehicle traffic rather than transit/walking traffic.

In short, the point I'm making is that there should be a resistance within a community to having parking added to their community. Most of this discussion so far has focused on the needs of the people in the far flung areas who would make a decision between driving all the way, or driving to a park-n-ride. But since they are travelling to an area that is someone else's neighborhood, I'd say not only are they not the only relevant decision maker, but they are the one who should have less say in the matter.

Anonymous said...

While it's interesting to read the hypothetical discussion about fitting PRT into US infrastructure, it just makes me feel you are wasting your ammo.

There are plentyful of places in the world where PRT fits right in. Where communities would be buying them if the price came slightly down and the purchasing procedure was made damn simple. This will happen.

However, Houston - and most of the US - feels like completely incompatible with the solution. So don't start there. ( I understand this undermines your scratching-my-own-itch aspect, but maybe the PRT for US needs to be something very different. Maybe no PRT at all? )

Back to track. I see PRT arriving to US later than the rest of the world, and maybe via Mexico and Canada. Both of those countries have areas where PRT will be able to "fit in" and do a contribution to the overall transport network. US usually does not have this - the transport network being the cars.

Where I feel PRT would make sense (in US) is not the city centers, and maybe also not the suburbs (as they currently are) but the outskirts where people are poorer. Look for conditions like Mexico. Where people actually use buses. Let those people operate PRT networks for their fellows in such neighbourhoods.

Another way for PRT to penetrate US is to have suburbs become more independent - cities of their own right. This change is already happening (see "Transition towns" and "Retrofitting urbania" - links below). What that change will do is radically shorten the car-driven distances of people living in such neighbourhoods. They will become local - and PRT is good for local trips.

Dan said...

I think you guys are selling PRT short… PRT would work fine for sprawling cities. It just can’t be slow and only local in scale. Houston recently voted to expand the 11 km of light rail for 43 million USD per kilometer – not including the trains or most of the roadwork. (There’s 180 million more for that) It works out to 1.4 billion for a 25km stretch where everyone will have to stop at crossings when the darn thing goes by. Add to that the fact that this light rail system will need to have groups of waiting passengers at the stops, yet there are few pedestrians around to feed that foot traffic. So the system will run well below capacity. PRT, it seems to me, is the best alternative.

In the first place the light track means it can be elevated so it won’t block traffic. In the second place it won’t run empty (burning huge amounts of electricity or fossil fuel) just for the sake of running. Being “on demand” is exactly what is required for areas with low pedestrian traffic. Third, there are the off-line stops, which are more efficient. Fourth, because the track is so much lighter, cheaper and so “out-of-the-way,” it can be routed much more extensively, addressing the problem of not having a lot of pedestrians at any given spot. PRT is EXACTLY what sprawling cities need. Sure, it would be even better at shuttling hoards of pedestrians within healthy, bustling, mixed-use neighborhoods, but I contend that it would also work for creating good neighborhoods in the first place. Public transit enables pedestrian activities and pedestrian activities enable public transit to run profitably. It is a virtuous cycle.

Not all US cities are like Houston or Atlanta. Sprawl comes from decades of fast growth and plenty of land. Mountains or water barriers make for more careful urban land use, although most cities have at least some direction where sprawl has taken place. Making each suburban “town” more of a complete entity a laudable goal but people increasingly change jobs and most jobs are somewhere else in the suburban sprawl. If you are not a commuter within this cityscape now, you probably will be one in a few years. A few years after that your job might be close again. Unless you have each little village specialize like in China (Umbrella town, Sock town, etc.) people will still need transportation between them. The sprawl is here, the houses already built. We need a public transportation solution that addresses it. If the (above mentioned) light rail track allowance went to PRT track instead, even at 10 million per mile, it would buy 128 miles. With most commercial hubs being 5-10 miles apart that would connect quite a few. So I ask you. If not PRT, then what?

Andrew F said...

Hey Dan,

Believe it or not, but we get a good three months of hot and sticky weather each summer in the Toronto area. Downtown Toronto has extensive underground concourses not just to hide from the cold in the winter, but the heat and humidity in the summer. We regularly get temps in the 90s with high humidity so its feels about 100+ F. It's not uncommon for the weather to be hotter here than in Florida.

Dan, I think a very mature PRT system would be able to serve those inter-suburb trips very well. But those are not the low hanging fruit. Transit fails these trips pretty utterly. Cars do a decent job depending on the trip, but often require running a gauntlet of highly congested highways and city streets.

The idea I fleshed out earlier on another thread was to start by replacing some highly used bus routes with PRT, particularly those that feed large trip generators or higher order transit. The existing bus/subway system provides a ready market for a nascent system. It should be able to make a profit on day one. From there, it should be possible to rapid expand the scope of the network, and it you start increasing market share (by getting people out of cars), creating finer loops through residential areas. But still, the vast majority of your passengers will arrive on foot or from other forms of transit. Park-and-ride will be possible, but it shouldn't be necessary.

If every rider on a PRT system needed a car to get to it, you would need vast parking lots around your network to create enough demand to justify the network. And car parks are not cheap, especially multi-story affairs. Better to grab existing transit riders and current car users that live close enough to walk, and leave the drivers in their cars until the PRT can be expanded closer to their homes. The goal ought to be to get them eventually, but we should do that after we collect the lower hanging fruit.

That said, park-and-ride is fine. Just run it on a cost-recovery basis--no subsidized parking. If people are still willing to park and ride, that demand should be satisfied.

Anonymous said...

Dan - with all politeness - Europe does not have that kind of money any more. We're broke - and so is US.

Time to stop spending like that.

As to "what then if not PRT". I really don't have a solution for the US. Pass. ;)

Andrew F said...

Not sure what you mean akauppi. North America is still and will continue to be very wealthy. Yes, they can't afford to be exorbitantly stupid and wasteful so much in the future--Canada itself went through a shock like that in the early-to-mid- '90s. The US is going to have to realise that it can't afford to spend a trillion dollars a year fighting useless wars, etc. Social security and medicare will become less generous.

But PRT is a technology that has the potential to release a lot of the wealth that is currently dedicated to maintaining an enormous fleet of personal cars. If cars cost up to $8000 to own and operate, even without factoring infrastructure costs, cutting car ownership/use in half could save America $1.2 trillion per year. I have no doubt that sufficiently advanced PRT could be pretty successful in most American cities. ULTra is not going to get us there, but it's a great first step.

Anonymous said...

On the mention of ULTra, the test drives we got in Heathrow were unnecessarily bUMPy. We don't want PRT like that. They look good, but the drive needs to be smooth. Hope they address this as a top priority.

Andrew F said...

Can you say why the ride is bumpy? I would assume that on a new guideway, a vehicle shouldn't need a fancy suspension to provide a smooth ride. Is it a result of defects in the guideway? The running surface is asphalt, if I'm not mistaken, and I could see some problems like washboarding occurring.

Sean said...

OK all;

There just is not enough money in the government programs to support transit requirements. ( ) Fortunately most of the transportation needs are controlled, not by the federal government, but by locals or companies.

Some areas will need a good deal of overhaul. Others not so much. Detroit has 3 miles of elevated track to Vancouver's 40 miles. Which city serves it people with (...) public transit?

Back to Detroit. It is linked to Chicago by the ever tenuous train. Ann Arbor on the other hand is linked by a less strenuous train. So let's link Ann Arbor to Metro and Detroit, quickly and with the ability to take a car.

Now we might have independent cities like Sugar Land, Texas that want a way to get people around, especially now that a new stadium is coming to this town of 90k. So maybe a PRT system is developed. No cars, just pods that can carry 2/3 people, 4/6 people and 20+ people.

If pod systems are set up to carry cars as well, then they would have to be coordinated with with the owner of the car caring system. In this way individual regional centers can be set up, (because region to region ei. Chicago to Ann Arbor, is carried by others, say a car train?).

We get something going in one locality to suit their needs and plan to connect these locals and regions together. That might be where the government comes in.

Dan said...

Dan the Blogger is back-

Sorry guys, I’ve been out of town and other holiday stuff.

Sean, I am not completely sure what you are advocating, but it sounds like it’s pretty big if it can carry 20+ passengers or automobiles. That would definitely never happen with private funds. I, personally, am not particularly enthused by the car carrying idea. In most cases it just encourages more automobile use. I prefer to concentrate on networks that free pedestrians from reliance on car ownership and use, and discourage continued suburban expansion. At least that is what this blog is about. There are plenty of other people who are interested in improving the automobile paradigm though, such as dual-mode advocates. You might be interested in checking out the Flyway site. It’s a fusion of what I am after and what you suggest. Very light, inexpensive track is a precondition for my vision, however, so that pretty much means personal-sized vehicles.

Dan said...

Dan the Blogger is back-

Sorry guys, I’ve been out of town and other holiday stuff.

Sean, I am not completely sure what you are advocating, but it sounds like it’s pretty big if it can carry 20+ passengers or automobiles. That would definitely never happen with private funds. I, personally, am not particularly enthused by the car carrying idea. In most cases it just encourages more automobile use. I prefer to concentrate on networks that free pedestrians from reliance on car ownership and use, and discourage continued suburban expansion. At least that is what this blog is about. There are plenty of other people who are interested in improving the automobile paradigm though, such as dual-mode advocates. You might be interested in checking out the Flyway site. It’s a fusion of what I am after and what you suggest. Very light, inexpensive track is a precondition for my vision, however, so that pretty much means personal-sized vehicles.