PRT... what's the deal?

[afransen TO]

I've done a bit of digging lately about 'personal rapid transit' and it seems to me that people tend to feel very strongly either one way or the other on it. I'm wondering what the consensus is among the saner minds around here. To me, cost benefit analysis seems like:

Pro:
-low track cost
-small guideway... can be suspended over streets without becoming oppressive
-low/no wait-times at stations
-24/7 service
-low labour cost
-high potential for network redundancy/robustness
-can be used for intra-city freight deliveries to large customers from large distributors
-provide better point-to-point service than conventional mass transit without transfers (approximates a car more closely) and without stopping at stations unnecessarily (provided off-line stations).
-can be operated with less capital/operating subsidy than a similar system with conventional modes
-can reduce long walks from stations


Con:
-limited capacity
-too expensive/low utilisation for suburbs, too low capacity for urban areas.
-safety concerns (not sure if these are valid)
-subways more practical for high volume transport between large traffic generators
-technology just doesn't work???


It seems to me that the pros outweigh the cons, which begs the question of why PRT isn't in use in a significant way. It means either my assessment is off, or I'm missing some considerations.

Either way, I'm interested in what you guys have to say, since my web searches haven't been that enlightening. The only one I found somewhat intruiging (though probably impractical) is:here

 

[EnviroTO]

Not exactly a typical balancing of pros and cons. It reminds me of an IBM ad where a project manager is all pumped up about some plan he hopes to implement and when he asks the question "is it possible" the answer is "no".

too expensive/low utilisation for suburbs, too low capacity for urban areas

Is one point for Con but

-low track cost
-low labour cost
-can be operated with less capital/operating subsidy than a similar system with conventional modes
-small guideway... can be suspended over streets without becoming oppressive

is four points for Pro?

We already have something PRT like... its called the road and automobile / truck.

- low track cost... slap down aphalt
- guideway fits in existing public ROWs
- low wait times at stations which are conveniently located right next to most homes and businesses
- 24/7 service
- low labour cost
- high potential for network redundancy/robustness
- can be used for intra-city freight deliveries to large customers from large distributors
- can be operated with less capital/operating subsidy than mass transit or bus
- can reduce long walks to/from stations

The goal of transit is to improve efficiency of transporting people. That efficiency can't be had if everyone has their own PRT module finding its way around the city. The efficiency of transit comes from people sharing the same vehicle, walking a distance to a stop, and not taking the direct route but rather taking a route which better serves the majority rather than the minority. The most inefficient vehicles are those hardly carrying any people.

A 747 loaded with passengers is more energy efficient per kilometer per person than a car driving one person around because you divide all those thousands of litres of fuel by a lot of people travelling a huge distance together. A Toyota Prius uses 4 litres / 100 km, a 747 gets approximately 1200 litres / 100 km... but if there is only one person in the car and 350 people in the 747 the figure becomes 4 litres / 100km / person versus 3.4 litres / 100km / person. So a 747, despite burning about 4 litres of fuel per second is more efficient than a Toyota Prius. The reason airlines use a hub and spoke system is due to efficiency, having people travel off course to a hub rather than directly to their destination saves the airline money... it costs less to fly you further due to the energy savings of putting more people in an aircraft versus more smaller aircraft flying direct.

It would make more sense to automate the driving of regular road vehicles and make those vehicles as fuel efficient as possible than it would to overlay a whole new system for PRT because the cost of the infrastructure, vehicles, etc wouldn't be able to easily pay off in efficiency gains.

 

[afransen TO]

I'm not sure you're actually giving a valid argument. Yes, cars are pretty much the ultimate mode of transportation, except they're expensive, they causes congestion (and new capacity is extremely expensive), inefficient from a capital point of view (everyone ties up $20,000 - $50,000 in their car).

Your argument against PRT can be applied to pretty much all transit, you can say that that people should drive to subway stations because buses are inefficient.

 

[EnviroTO]

If PRTs are carrying the same number of people at a given time as car then they are not going to be any more efficient than a car could be. If capital cost of a car is the concern then there is the option of carpooling, autoshare.com, etc. Efficiency in terms of fuel and cost will almost always favour the vehicle shared by more passengers.

Dealing with congestion it is speed and compactness of flow that determines the volume that can be transported at a given time. Efficiency as it relates to congestion is different than energy efficiency because it is all about volume and speed... an electric car spends no more energy to travel a distance in congestion than it does without congestion. Lets assume that a PRT can drive no faster and no closer together than a car (note: the reason I mentioned automating the driving process of regular vehicles in the previous post was that automation allows vehicles to run closer together at higher speeds). If a PRT vehicle is carrying the same number of people on average as the cars on the road, the congestion on the PRT system would be similar to that experienced on the road. If the PRT carries on average twice the number of people per vehicle as compared to the car it will require half the number of lanes of flow to have the same level of congestion... in the case of the 401, a PRT replacement conduit for the 401 would need 8 tracks to replace 16 lanes of congestion if the PRT vehicles held on average twice the number of passengers.

Your argument against PRT can be applied to pretty much all transit, you can say that that people should drive to subway stations because buses are inefficient.

No, not drive... walk. Taking a bus to the subway is more efficient than taking a car to the subway because it carries more passengers at a time. The more people you can get onto a vehicle the more efficient it becomes. A subway is more efficient than a bus so bus routes should be planned to quickly drop of people at a subway to combine the loads of a number of buses to fill a larger vehicle. If a PRT carries less people than a bus; then the only place it could efficiently fit into the transit landscape is below the bus as a "bus feeder". I don't think there is much use for bus feeders.

 

[Degnaw]

honestly, "prt" is pointless; its just public cars on guideways. And i don't really think the low cost thing applies, its basically just a monorail-type thing (and those aren't really cheap, are they?) If they build a prt guideway... put monorail cars on it.

 

[afransen TO]

"That efficiency can't be had if everyone has their own PRT module finding its way around the city. The efficiency of transit comes from people sharing the same vehicle, walking a distance to a stop, and not taking the direct route but rather taking a route which better serves the majority rather than the minority. The most inefficient vehicles are those hardly carrying any people."

If carrying lots of passengers in one vehicle makes transit efficient, why do we use anything other than those GO cattlecars? Efficiency is a matter of value of a trip over cost to provide it, say? I've seen some sites that make pretty good case for PRT systems that provide good value to passengers for reasonable cost. Value I suppose is speed, comfort and how close the endpoints of the trip match the departure and destinations of the user. PRT can more closely approximate the 'value' of a car than a bus could ever hope to, and costs don't seem totally unreasonable.

"Lets assume that a PRT can drive no faster and no closer together than a car"

That's a pretty unreasonable assumption for a host of reasons.

"in the case of the 401, a PRT replacement conduit for the 401 would need 8 tracks to replace 16 lanes of congestion if the PRT vehicles"

Well, I don't think any PRT system really suggests that it's a good alternative for highways. Also consider what you're saying. Why would you put 8 tracks side by side in one ROW when you can put one track on 8 parallel streets? You avoid the need for huge arteries like the 401.... and gain a network that allows people to be picked up/dropped off closer to their destination without transfers.

"If a PRT carries less people than a bus; then the only place it could efficiently fit into the transit landscape is below the bus as a "bus feeder""

Why would any feeding be involved, since PRT could take you to your destination without transfer? It seems to me that transferring to a faster mode of transit would only be advantageous if you were going fairly long distances.

 

[EnviroTO]

If carrying lots of passengers in one vehicle makes transit efficient, why do we use anything other than those GO cattlecars?

Because you can't fill up a GO train from all locations. You can't even fill up a GO train from most GO stations... most stations rely on being to fed from cars and buses.

That's a pretty unreasonable assumption for a host of reasons.

Why? I've been on the 400 on a holiday weekend and traffic was moving at 120km/h and was pretty much bumper to bumper. If the automated highways take effect this could be the norm rather than the exception. The subway has a maximum speed of 88km/h.

Why would you put 8 tracks side by side in one ROW when you can put one track on 8 parallel streets?

For the same reason you need the freeway despite having all those parallel streets.

You avoid the need for huge arteries like the 401.... and gain a network that allows people to be picked up/dropped off closer to their destination without transfers.

How? Where does the carrying capacity of these PRT guideways come from when all the freeways that exist can't get rid of the congestion experienced by cars despite all those cars going directly where they want to go?

Why would any feeding be involved, since PRT could take you to your destination without transfer? It seems to me that transferring to a faster mode of transit would only be advantageous if you were going fairly long distances.

Transfering isn't an advantage to the passenger, it is an advantage to efficiency in terms of cost and energy consumption.

There was a poster the TTC had a while ago showing how many cars where removed from the road by one streetcar. The streetcar taking up very little room on the street had 30 or probably more cars on the street behind it... something like four or five lanes wide and 6 or seven cars deep. That poster is what transit is all about... reducing the number of vehicles by increasing the number of people on each vehicle. Those cars in the poster could just have easily been a number of PRT vehicles... the point is the same, you need to find a way to handle a whole bunch of vehicles moving through the system (be it roadway, track, or otherwise) rather than one vehicle. Not only that but you need to maintain all those vehicles and power those vehicles, rather than looking after one.

 

[afransen TO]

"Why? I've been on the 400 on a holiday weekend and traffic was moving at 120km/h and was pretty much bumper to bumper. If the automated highways take effect this could be the norm rather than the exception. The subway has a maximum speed of 88km/h."

You weren't travelling with 0.5s headways. Cars are too unreliable to be automatically guided, road conditions certainly wouldn't always allow it, no good way to detect obstacles (like people), etc. Then you still have the problem of having to store the car at the destination.

"For the same reason you need the freeway despite having all those parallel streets."

Cars do it so that can travel faster than they would on the normal network. That wouldn't apply with PRT....

"How? Where does the carrying capacity of these PRT guideways come from when all the freeways that exist can't get rid of the congestion experienced by cars despite all those cars going directly where they want to go?"

Strictly speaking, were not going to get rid of cars altogether. A PRT system would be controlled by computer to distribute congestion. If any particular street required more capacity than PRT could provide, then higher modes could get involved (subway, etc.).

"Transfering isn't an advantage to the passenger, it is an advantage to efficiency in terms of cost and energy consumption."

It's a heck of a lot faster to take the bus to the subway and take the subway downtown rather than take the bus downtown. Buses are DAMN slow...


"There was a poster the TTC had a while ago showing how many cars where removed from the road by one streetcar. The streetcar taking up very little room on the street had 30 or probably more cars on the street behind it... something like four or five lanes wide and 6 or seven cars deep. That poster is what transit is all about... reducing the number of vehicles by increasing the number of people on each vehicle. Those cars in the poster could just have easily been a number of PRT vehicles... the point is the same, you need to find a way to handle a whole bunch of vehicles moving through the system (be it roadway, track, or otherwise) rather than one vehicle. Not only that but you need to maintain all those vehicles and power those vehicles, rather than looking after one."

What headway do streetcars operate? At least 2 minutes or so, lets say. PRT carrying as few as 1.5 passengers each can carry 180 people in that time, at capacity, which isn't too bad compared to a streetcar, and the travel time of passengers will be much better (where travel time in a streetcar suffers as passenger load increases).

And when we're talking about efficiency, let's not forget the TTC's #1 expense: labour. And you just can't automate things that operate at grade on the street. PRT can do what the streetcar can do faster and without the driver.

 

[Degnaw]

The TTC is NOT going to automate... *cough* SRT *cough*

And the prt is not going to reduce distances to stations, because you still need the infrastructure that IS still going to cost more than roads/brt.

I still don't see the point... you need to pay for the elevated guideway still, and even if a system is built and heavily used (like the 1 car/second you mentioned) extensive merging and stopping will be involved and eventually it'll just become another road system. Having automation will NOT prevent the fact that when 3 guideways merge into one, the prt vehicles will have to slow down to accomodate each other.

Either i've been reading too much of this, or you've been reading to little:
www.lightrailnow.org/facts/fa_prt001.htm

 

[afransen TO]

Thanks for that link. I hadn't been able to find a technical refutation of PRT claims. All the anti-PRT material I could find tended to be emotional rants rather than reasonable arguments. Although, I will say that I expect some of that paper's cricisms of PRT are a bit overstated.

I also enjoyed the link on that page regarding automation and whether it actually cuts operating costs. They argue it doesn't, which seems a bit hard to believe. That it hasn't in practice seems to me to indicate that the savings have allowsed inefficiencies to persist or grow in other areas. I find it hard to believe that automating the SRT, for instance, wouldn't at least yield moderate operating cost savings.

 

[rbtaylor]

I also enjoyed the link on that page regarding automation and whether it actually cuts operating costs. They argue it doesn't, which seems a bit hard to believe.

I would expect that any automation that can be achieved in a PRT system can also be achieved with minor alterations to your run of the mill car as well.

The big reason PRT is easier to apply it to is that it's elevated (fully separated) from pedestrian traffic. Highways also have that benefit. In fact, mostly automated truck trains (drive by wire) are used today.

The big thing against automation is psychological rather than technical. Once baby-boomers retire and the WWII generation is gone, expect these types of things to start appearing fairly quickly as devices to assist the driver (automated parallel parking, etc.).

After 10 years of that, once it has been demonstrated that the automated device has fewer accidents than the human, it quickly becomes required via law for the sake of the children. In 20 years the vast majority of the voting public will have grown up with heavy automation, and won't freak out.

 

[EnviroTO]

Cars do it so that can travel faster than they would on the normal network. That wouldn't apply with PRT....

Why not? On local systems there would be PRT vehicles slowing down, switches switching, and other obstacles which would necessitate a route which did not have those bottlenecks.

Strictly speaking, were not going to get rid of cars altogether. A PRT system would be controlled by computer to distribute congestion. If any particular street required more capacity than PRT could provide, then higher modes could get involved (subway, etc.).

Perhaps, but at a greater cost than a computerized traffic light system and GPS devices with traffic alerts.

It's a heck of a lot faster to take the bus to the subway and take the subway downtown rather than take the bus downtown. Buses are DAMN slow...

That isn't a measure of the bus speed versus the subway speed. A bus need not stop on the entire route all the way downtown. A bus is capable of driving non stop on freeways.

What headway do streetcars operate? At least 2 minutes or so, lets say. PRT carrying as few as 1.5 passengers each can carry 180 people in that time, at capacity, which isn't too bad compared to a streetcar, and the travel time of passengers will be much better (where travel time in a streetcar suffers as passenger load increases).

120 PRT vehicles in a period of 2 minutes means one per second. In a mechanical system with rail switches to switch, possible mechical failures, etc it seems incredibly unlikely that a system could operate at that level. We aren't talking about mechanical disk drive heads here... wer are talking about big vehilcles carrying people. In addition you would have 120 PRT vehicles to maintain versus one ALRT streetcar which can also carry 180 people. Also, having streetcars run at 30 second headways or far more achievable than vehicles with 1 second headways.

And when we're talking about efficiency, let's not forget the TTC's #1 expense: labour. And you just can't automate things that operate at grade on the street. PRT can do what the streetcar can do faster and without the driver.

But LRT and subways don't need to operate at grade and the labour costs are divided by the number of passengers served per hour. If a streetcar has a 30 minute route and carries 100 people during that trip the cost per passenger of that driver is 17 cents ($70000/50wk/40h/2trips/100people).