String Rail - a low cost, low impact, high speed transport alternative

[Second_in_pie]

Careful slagging PRT. A lot of proposals are pretty half baked, like this one, but the technology has a lot of promise, especially for urban transport.

Umm, no. PRT is a terrible idea.

The only reason you'd build tens of kilometers of rail for use by 1, maybe 2 people per vehicle over rail for vehicles that can hold way over a hundred people is that your people want to live their separate lives away from anyone else. It's slightly better system capacity than cars, but the same mindset, and for it to actually work you'd have to build these lines all over your city. Imagine if we built subway all around the city instead! PRT comes from an American philosophy that says you shouldn't have to interact with others, and when you do it should be in the shortest and least demanding of your attention as possible. Why would we build transportation that's basically the same as a car, except that it's totally incompatible with car infrastructure and vice-versa? It's not even a good idea as an airport people mover.

 

[afransen]

PRT is not a substitute for subways. It's a substitute for cars especially and perhaps buses.

Yes, it would carry fewer people than heavy rail, but then, it could cost orders of magnitude less per km. PRT is not exactly like the car: there are shared vehicles (very little to no parking), inexpensive infrastructure, the vehicles are light and energy efficient, and it is less prone to congestion because traffic is managed at a system level. Certainly a PRT-oriented city would look nothing like a car-oriented city.

I think a lot of people are knee-jerk anti-PRT on a moral level, which I find an exceedingly odd dynamic to be injected into an infrastructure discussion . I look at it from a utilitarian perspective: is it technically feasible, would people use it, would it be cost effective. There is nothing inherent in PRT as a technology that requires we answer one of these questions in the negative. Foolish proposals in the past do not mean every future proposal be foolish.

As far as how it works as a people-mover, we'll see how it performs at Heathrow. I haven't seen any indication that it will be a disaster, there, and it will be entering public service in the upcoming weeks. Certainly the PRT used by Heathrow is far from ideal, but it is useful at demonstrating the principles of the technology.

 

[EnviroTO]

I think transit plus walking / cycling / autoshare makes far more sense than PRT. PRT infrastructure may be cheaper than heavy rail but how can it be cheaper than the roads which are already there for purposes of emergency vehicles, delivery trucks, etc. Perhaps eventually cars will be smart enough to drive themselves and autoshare programs will be more popular, but the infrastructure of PRT makes no sense.

 

[Second_in_pie]

PRT is not a substitute for subways. It's a substitute for cars especially and perhaps buses.

Yes, it would carry fewer people than heavy rail, but then, it could cost orders of magnitude less per km. PRT is not exactly like the car: there are shared vehicles (very little to no parking), inexpensive infrastructure, the vehicles are light and energy efficient, and it is less prone to congestion because traffic is managed at a system level. Certainly a PRT-oriented city would look nothing like a car-oriented city.

But it's not. PRT may be more efficient than cars, but for PRT to be as effective as cars, it needs to go everywhere. Now obviously we're not going to be building little monorails or whatever down every street, so what does that leave us with? Major arterials? Once you get to major arterials, things like subway or LRT become better, and that leaves gaps for people to chose their cars over PRT because there'll still need to be road infrastructure for lazy people to get to the PRT stations. PRT won't be able to handle the crush loads that subways or regional trains could, so there'd have to be a switch to those modes as you get further down. It becomes very inefficient very fast, and loses a lot of it's pros once you factor in things like density or open space.

I think a lot of people are knee-jerk anti-PRT on a moral level, which I find an exceedingly odd dynamic to be injected into an infrastructure discussion . I look at it from a utilitarian perspective: is it technically feasible, would people use it, would it be cost effective. There is nothing inherent in PRT as a technology that requires we answer one of these questions in the negative. Foolish proposals in the past do not mean every future proposal be foolish.

My knee-jerk anti-PRT on a moral level is the same reaction I give cars. Engraining a subculture that's based around total isolation from others into our culture is a bad idea, whether the vehicle runs on tarmac or rails. I like public transit and bicycles because they require a human interaction with others. Humans are social creatures, yet in the huge metropoles that we have today we are as isolated from one another as ever. This may not factor into whether one transit system is more effective than another, but if it's tied in effectiveness but is radically different from another, I think the broader social effects that it has are important.

 

[afransen]

I think transit plus walking / cycling / autoshare makes far more sense than PRT. PRT infrastructure may be cheaper than heavy rail but how can it be cheaper than the roads which are already there for purposes of emergency vehicles, delivery trucks, etc. Perhaps eventually cars will be smart enough to drive themselves and autoshare programs will be more popular, but the infrastructure of PRT makes no sense.

We'd certainly need fewer arterials, and streets would be more pleasant with less car traffic. Nonetheless, cars have and will always have the problem of operating on a plane, which means frequent stops/starts, low average speed and congestion. Cars that drive themselves will probably never be suitable for use in urban environments, particularly because they are a liability nightmare. If you think PRT is unrealistic, letting computer-guided vehicles operate at high speed near pedestrians in all weather conditions is positively nuts. Even though roads already exist, traffic does impose a degree of wear and tear so the marginal car is not 'free'. Additionally, cars require parking, which gobbles up an astonishing amount of an urban area. If we could reduce parking demand by half it would make an enormous difference.

I'd say PRT infrastructure makes sense if its fare revenue can cover the capital costs of construction in addition to its operating cost. Regardless of how you feel about how likely this is, I think we can agree that if it can achieve cost recovery, the infrastructure would make sense. It's more than you can say about any existing transportation infrastructure beyond toll highways, and perhaps the busiest subway lines.

 

[afransen]

But it's not. PRT may be more efficient than cars, but for PRT to be as effective as cars, it needs to go everywhere. Now obviously we're not going to be building little monorails or whatever down every street, so what does that leave us with? Major arterials? Once you get to major arterials, things like subway or LRT become better, and that leaves gaps for people to chose their cars over PRT because there'll still need to be road infrastructure for lazy people to get to the PRT stations. PRT won't be able to handle the crush loads that subways or regional trains could, so there'd have to be a switch to those modes as you get further down. It becomes very inefficient very fast, and loses a lot of it's pros once you factor in things like density or open space.

So a grid of subways of equal spacing as PRT would always make more sense? Maybe we live in different cities, but I don't think we'll be building $300 million/km subway on Mavis any time soon. Like I said, these are not substitutes for each other. As far as crush-loads go, that's partially a symptom of having very expensive infrastructure: you tend to funnel everyone down it to maximize utilization. With a network rather than a branching structure, you would be spreading the flow out over multiple, parallel routes rather than funneling everyone from many kilometers in either direction to one line.

My knee-jerk anti-PRT on a moral level is the same reaction I give cars. Engraining a subculture that's based around total isolation from others into our culture is a bad idea, whether the vehicle runs on tarmac or rails. I like public transit and bicycles because they require a human interaction with others. Humans are social creatures, yet in the huge metropoles that we have today we are as isolated from one another as ever. This may not factor into whether one transit system is more effective than another, but if it's tied in effectiveness but is radically different from another, I think the broader social effects that it has are important.

How much human interaction occurs on a bus? My experience (and I honestly don't mind riding buses at all!) is a group of people studiously avoiding eye contact, blaring music in their ears, reading books, or generally doing their best to pretend that no one else exists. I think your social engineering effort is based on a premise that may not actually be borne out in reality. My concern with cars is more an efficiency argument. They gobble up remarkable amounts of resources, not only energy but also just the sheer cost of the infrastructure, the wasted space, and unpleasant cities. We're never going to make much progress replacing cars with modes of transit that struggle to top 25 kph in average speed. Beyond that, I'm not convinced that traditional transit is always the most economical option. Most transit service requires massive subsidy, and is not particularly user-friendly except in the very densest areas. So, I think it's worthwhile not to rule out a potentially useful technology for transporting people or goods more cost effectively and in a way that reduces environmental impact. Although I don't see it as the solution to every problem, I see no reason to ban it from our toolbox of transportation technologies . Especially if that ban is motivated by some pleasant fiction about community-building on buses.

 

[Second_in_pie]

f

So a grid of subways of equal spacing as PRT would always make more sense? Maybe we live in different cities, but I don't think we'll be building $300 million/km subway on Mavis any time soon. Like I said, these are not substitutes for each other. As far as crush-loads go, that's partially a symptom of having very expensive infrastructure: you tend to funnel everyone down it to maximize utilization. With a network rather than a branching structure, you would be spreading the flow out over multiple, parallel routes rather than funneling everyone from many kilometers in either direction to one line.

Yeah, that's kind of the point. Do you really think that PRT will be drastically lower cost than subway? And when I say drastically lower cost, I mean half at the best. We're obviously not going to be building $300 million/km subway on Mavis, but will we be building $200 million/km PRT on Mavis for maybe half of the service? The problem is that basically what PRT is advocating is a car. So why don't we a) just funnel that money into a hydrogen/electric network, or b) just clear the roads and put in BRT or some other type of RT in the space that PRT would take up?
The point is that if you're

How much human interaction occurs on a bus? My experience (and I honestly don't mind riding buses at all!) is a group of people studiously avoiding eye contact, blaring music in their ears, reading books, or generally doing their best to pretend that no one else exists. I think your social engineering effort is based on a premise that may not actually be borne out in reality.

So you've never seen two people meet by chance on the subway and discuss what's happening? Or at least enjoy other people's company as you take your trip home? Or perhaps have someone comment on your clothes or phone or something? I'll admit that part of that problem is the fact that we've got a broader social phenomenon coming from suburbanization and auto use that public transit won't solve as soon as you get on it.

My concern with cars is more an efficiency argument. They gobble up remarkable amounts of resources, not only energy but also just the sheer cost of the infrastructure, the wasted space, and unpleasant cities. We're never going to make much progress replacing cars with modes of transit that struggle to top 25 kph in average speed. Beyond that, I'm not convinced that traditional transit is always the most economical option. Most transit service requires massive subsidy, and is not particularly user-friendly except in the very densest areas. So, I think it's worthwhile not to rule out a potentially useful technology for transporting people or goods more cost effectively and in a way that reduces environmental impact. Although I don't see it as the solution to every problem, I see no reason to ban it from our toolbox of transportation technologies . Especially if that ban is motivated by some pleasant fiction about community-building on buses.

Okay, so if public transit isn't as cost-effective as PRT, the only conclusion I can come to is that PRT will have a distance-based fee after you buy your $5k/year membership to pay for the cars that you use. That's really the only way for it to be actually cost effective, and it's the same deal you'd be getting using a car on a private toll highway.

And you do realize that PRT would run on a pretty heavy subsidy too, just as cars do? I'm guessing that even a cost of maybe 50 cents/km and a $5k/year user fee could be optimistic for it to be totally on it's own, but would you really pay for that? Especially if you need a car anyways to get groceries or go up to your cottage? I honestly doubt it.

PRT just wouldn't work in a city situation. Even if you're going to be putting it on arterials, it's going to have a hard time keeping up with cars and the amount of traffic regular roads would be able to handle, and you'd still have to walk, drive, or take the bus to your local PRT station. And if you actually want it to replace cars, then you'll have to at least double track it on an arterial like Mavis. For that cost right there, you could probably stick a subway up Mavis instead.

I won't say to ban PRT from our transit toolbox, but just that if you were to try to fit it into almost any transit route, you'd find that there are much better options.

 

[afransen]

fYeah, that's kind of the point. Do you really think that PRT will be drastically lower cost than subway? And when I say drastically lower cost, I mean half at the best. We're obviously not going to be building $300 million/km subway on Mavis, but will we be building $200 million/km PRT on Mavis for maybe half of the service? The problem is that basically what PRT is advocating is a car. So why don't we a) just funnel that money into a hydrogen/electric network, or b) just clear the roads and put in BRT or some other type of RT in the space that PRT would take up?
The point is that if you're

Alright. PRT won't cost $200 million/km. That's the point. The company that installed a pilot system at Heathrow has designed some systems for other cities as part of a RFP process, generally quoting $10 million/km in capital costs. To be fair, they have only built their Heathrow system, but if we guess they're off by a factor of two, then PRT is still far less than a tenth the cost of subway per installed km. Not that it is directly competing with subway. I'd say it is competing with suburban LRT/BRT/bus applications to a much greater extent. And I would certainly say that their system is only a baby step in terms of what PRT is capable of.

PRT is not a car. It's closest analogue is a taxi, with point to point service, low/no parking requirement, and a pool of vehicles shared by users.

As far as space utilization, PRT generally has to operate in a separate ROW. So when running along streets, it would generally need to be elevated. This is one strike against the system employed by Heathrow in my opinion: it's guideway would be a bit imposing for my taste. I'd lean more toward a suspended system, which has more flexibility. My point is, though, that the footprint would be supports every 20 or 30 m and stations.

So you've never seen two people meet by chance on the subway and discuss what's happening? Or at least enjoy other people's company as you take your trip home? Or perhaps have someone comment on your clothes or phone or something? I'll admit that part of that problem is the fact that we've got a broader social phenomenon coming from suburbanization and auto use that public transit won't solve as soon as you get on it.

I'm not saying this doesn't happen, I'm saying that it is rare. Furthermore, I'm not sure we should choose transportation options to optimize social interaction. I think sidewalks, stores, restaurants, pubs, cafes, bookstores generally serve this purpose pretty well.

Okay, so if public transit isn't as cost-effective as PRT, the only conclusion I can come to is that PRT will have a distance-based fee after you buy your $5k/year membership to pay for the cars that you use. That's really the only way for it to be actually cost effective, and it's the same deal you'd be getting using a car on a private toll highway.

No membership required. It would work just like a taxi. The operating cost would be lower than a taxi, since the vehicle would consume 75-90% less energy and would not require a driver. Also, travel times would be faster than taxi, especially in congested areas. As we know, people are pretty willing to pay high fares to use taxis, and PRT could provide taxi-like service at significantly lower cost. I think there would be a market, but it remains to be proven.

And you do realize that PRT would run on a pretty heavy subsidy too, just as cars do? I'm guessing that even a cost of maybe 50 cents/km and a $5k/year user fee could be optimistic for it to be totally on it's own, but would you really pay for that? Especially if you need a car anyways to get groceries or go up to your cottage? I honestly doubt it.

I would oppose large-scale PRT deployment if it required a subsidy, particularly if that subsidy is more per passenger kilometer than typical transit. I imagine that PRT would work quite well in conjunction with car-share programs. I think you're putting up a straw man with your $5k/year structure. No business would try to set up such a pricing structure. To give you an idea of what you're suggesting, if we had three million users paying what you suggest for access to this service, that would be equivalent to $45 billion per year, assuming 20,000 km per user which is typical for car use. Do you really imagine that a system would cost 45 times what it costs to run the TTC, just to serve the GTA? My point is that you're deliberately choosing absurd prices in order to conclude that no one would be willing to pay them. This is the straw man fallacy.

PRT just wouldn't work in a city situation. Even if you're going to be putting it on arterials, it's going to have a hard time keeping up with cars and the amount of traffic regular roads would be able to handle, and you'd still have to walk, drive, or take the bus to your local PRT station. And if you actually want it to replace cars, then you'll have to at least double track it on an arterial like Mavis. For that cost right there, you could probably stick a subway up Mavis instead.

I won't say to ban PRT from our transit toolbox, but just that if you were to try to fit it into almost any transit route, you'd find that there are much better options.

PRT doesn't have to stop between the origin and destination. Even if PRT has a lower top speed than cars are permitted, the fact that cars have to stop frequently significantly lowers their average speed. PRT generally emphasizes many, small stations to minimize walking. The key here is inexpensive stations where there is low demand. It is certainly conceivable to have a system where most homes and business are within a few hundred meters of a stop, which is only a few minutes walk. The walk is less of a disincentive if you don't have to wait for ten or fifteen minutes at the stop for a vehicle to arrive.

I'd agree with you, that if PRT cost $200 million/km it'd be a tough sell anywhere. Thankfully, the actual number seems to be closer to 1/20th that figure.

PRT isn't quite ready for prime-time yet, but I'm hopeful that the baby-steps it is currently undergoing can quickly spur more investment in developing the technology and larger deployments to build the case for its effectiveness.

If you're curious about some of the pitfalls and design problems with PRT, this blog is pretty interesting:Open PRT Specification project It's written by an engineer trying to design a cost-effective, flexible, open-source PRT standard.

 

[EnviroTO]

We'd certainly need fewer arterials, and streets would be more pleasant with less car traffic. Nonetheless, cars have and will always have the problem of operating on a plane, which means frequent stops/starts, low average speed and congestion. Cars that drive themselves will probably never be suitable for use in urban environments, particularly because they are a liability nightmare. If you think PRT is unrealistic, letting computer-guided vehicles operate at high speed near pedestrians in all weather conditions is positively nuts.

I'm not sure we would require less arterials, just less lanes on the existing ones. Cars have the problem of operating on a plane but if used only to get to the railway or subway there isn't as far to go. I'm not sure that a self operating car is all that nuts. A number of sensors can be employed creating redundancy, and solutions to problems can be programmed in which are used instantaneously, far quicker that the reaction time of a human being. If a child runs out onto the street I would trust a computer-guided car to avoid it more that the guy rushing home after a long day at work. When the technology is ready there is no reason to believe that the computer wouldn't do it better. Emergency situations benefit from the fast response times computers can deliver.

Even though roads already exist, traffic does impose a degree of wear and tear so the marginal car is not 'free'. Additionally, cars require parking, which gobbles up an astonishing amount of an urban area. If we could reduce parking demand by half it would make an enormous difference.

Autoshare combined with computer driven cars reduce parking the same amount. The maintenance costs of elevated structures are likely to outweigh what is actually required of a surface level street since most road maintenance after initial construction is simply smoothing out the surface.

I'd say PRT infrastructure makes sense if its fare revenue can cover the capital costs of construction in addition to its operating cost. Regardless of how you feel about how likely this is, I think we can agree that if it can achieve cost recovery, the infrastructure would make sense. It's more than you can say about any existing transportation infrastructure beyond toll highways, and perhaps the busiest subway lines.

If PRT is rolled out like roads are where cost recovery is not a goal then it likely wont achieve cost recovery. The only reason any mode of transport which does not pay for itself would continue to exist is that we have decided it is acceptable to be that way as a society. Mass transit will always be more efficient than PRT. PRT can be more efficient than cars if they are running on surfaces which have less resistance, and have designs which are more lightweight and aerodynamic... but the same could be said if cars and the roads they drive on were redesigned since after all PRT is simply elevated computer-driven auto-share.

 

[afransen]

I'm not sure we would require less arterials, just less lanes on the existing ones. Cars have the problem of operating on a plane but if used only to get to the railway or subway there isn't as far to go. I'm not sure that a self operating car is all that nuts. A number of sensors can be employed creating redundancy, and solutions to problems can be programmed in which are used instantaneously, far quicker that the reaction time of a human being. If a child runs out onto the street I would trust a computer-guided car to avoid it more that the guy rushing home after a long day at work. When the technology is ready there is no reason to believe that the computer wouldn't do it better. Emergency situations benefit from the fast response times computers can deliver.

It's still a liability nightmare. What driver would want to be liable for the actions of their robocar, and what robocar manufacturer would want to be liable for any accident their vehicle might be involved in? People demand very high reliability in automated systems and systems where the passenger is not in control. We tolerate tens of thousands of thousands of car-related deaths a year, but a single plane crash is totally unacceptable. No computer will ever achieve perfect reliability in terms of safe driving when operating on imperfect roads, in any weather condition and at high speeds in proximity to pedestrians, cyclists, etc. It really strikes me as a fantasy.

Autoshare combined with computer driven cars reduce parking the same amount. The maintenance costs of elevated structures are likely to outweigh what is actually required of a surface level street since most road maintenance after initial construction is simply smoothing out the surface.

Autoshare+computer guided vehicles is essentially an inferior PRT technology. It is less safe, more costly, and less efficient than other refinements that are possible. It's attempting to jam the existing car technology peg into the PRT hole.

If PRT is rolled out like roads are where cost recovery is not a goal then it likely wont achieve cost recovery. The only reason any mode of transport which does not pay for itself would continue to exist is that we have decided it is acceptable to be that way as a society. Mass transit will always be more efficient than PRT. PRT can be more efficient than cars if they are running on surfaces which have less resistance, and have designs which are more lightweight and aerodynamic... but the same could be said if cars and the roads they drive on were redesigned since after all PRT is simply elevated computer-driven auto-share.

PRT can be more light-weight because it doesn't have to be crashworthy, or at least not to the degree of cars. I don't think it would be acceptable for robocars to operate in mixed traffic with 40 ton trucks with the same crashworthiness a PRT vehicle could achieve. Secondly, PRT doesn't need to carry a powerplant, be it a battery pack or a internal combustion engine. It can be powered with third rail. Thirdly, there is the repeated stops/starts with cars operating on a plane than are a drag on efficiency.

'Transit' is only more efficient than PRT in terms of energy consumed per passenger kilometer where it has high enough utilization. A bus at 30% capacity wouldn't compare favourably. Never mind cost or travel time considerations.

You're right that robocar+carshare is essentially PRT. Now, why would we lock ourselves into a more expensive, less safe, less efficient PRT variant if another, superior variant is available?

 

[EnviroTO]

You're right that robocar+carshare is essentially PRT. Now, why would we lock ourselves into a more expensive, less safe, less efficient PRT variant if another, superior variant is available?

Because one is an upgrade, the other is rip and replace.

 

[afransen]

Even if the upgrade is more expensive?

 

[Mapleson]

It's still a liability nightmare. What driver would want to be liable for the actions of their robocar, and what robocar manufacturer would want to be liable for any accident their vehicle might be involved in? People demand very high reliability in automated systems and systems where the passenger is not in control. We tolerate tens of thousands of thousands of car-related deaths a year, but a single plane crash is totally unacceptable. No computer will ever achieve perfect reliability in terms of safe driving when operating on imperfect roads, in any weather condition and at high speeds in proximity to pedestrians, cyclists, etc. It really strikes me as a fantasy.

In 2007, the DARPA Urban Challenge had 6 cars complete a 96-km urban course within 6 hours without human intervention.

Next, look at some of the existing robotic 'reactions'.
[video=youtube;bfdHY26E2jc]http://www.youtube.com/watch?v=bfdHY26E2jc[/video]

Of course, a robot car has to be a better driver than a human, but it won't be hard to do within the next decade. Also, Ontario has no-fault liability laws, so even if your robocar ran into someone else, either you nor the manufacturer would be liable. If fact, I think insurance companies would probably give you a discount for having a robocar.

 

[jwill]

letting computer-guided vehicles operate at high speed near pedestrians in all weather conditions is positively nuts.

I'd say allowing human-controlled vehicles to operate at high speed near pedestrians in all weather conditions is nuts. Computers aren't in a rush to get somewhere, or talking on their phone.

 

[afransen]

Reliability of computer guided cars is not the only strike against a robocar/car-share paradigm. The vehicles, and infrastructure are inherently heavier and more expensive, and thus more energy and resource intensive. The fact that they operate on a plane in mixed traffic limits safe operating speed, increasing travel time. Operating in mixed traffic with imperfect traction will also lead to increased headways over human operators.