GRT vs PRT with Train Formation

[Mike C]

Ingmar Andreasson's raises the possibility of forming PRT trains in
his recent paper. I think PRT trains is a better option than GRT,
which some have suggested as a higher-capacity alternative to pure
PRT. I use the term "trains" as Ingmar does, because "platoon" seems
to imply dynamic assembly of the groupings at line speed, whereas
"train" implies low-speed assembly.

Trains of PRT vehicles offer capacities on par with larger GRT
vehicles, though GRT would still win out in pure capacity. But whereas
GRT vs traditional PRT is a huge capacity benefit (4-5 times the
maximum capacity for GRT), PRT trains close that gap considerably,
i.e. 1.5 times the capacity for GRT vs PRT with train formation.

PRT trains may require multiple "stops" to disassemble the train, but
this is no different than what GRT would require - whether there are
10 vehicles in a PRT train or 10 passenger groups in a GRT vehicle,
there will be as many as 10 stops en-route. In the PRT train case,
some of those stops may be averted by splitting the platoon at a
diverge point (as Ingmar describes), and if there are stations with
parallel berths, the train splitting can be done very quickly in the
stations themselves, because the remaining PRT vehicles don't have to
wait for the stopped vehicle to empty (because it turns off into a
parallel berth).

So from a "stops" perspective, PRT trains are no worse than an
equivalent GRT vehicle.

PRT trains also avoid the necessity of grouping passengers in the
station, as would be necessary for GRT vehicles. However, there is the
added complexity of forming a train of vehicles with similar
destinations. For example, when leaving a downtown stadium, vehicles
may be going several different directions (typically 4, I would think)
so they would need to be grouped by direction.

Once again, I think this can be solved within stations themselves,
with some combination of extra sidings and/or parallel berths. For
example, let's say there are 4 stations on a neighborhood loop
connecting to a main line. Each of those stations can serve as a train-
builder for different destination areas. Let's say the stations are
S1,S2,S3,S4 and the destination areas are D1,D2,D3,D4, and each
station is responsible for assembling a single destination-area train,
e.g. S1->D1, S2->D2, etc. Now, if someone enters S2 and requests D1,
the vehicle loops around to S1 and joins the train bound for D1. When
the D1 train is full, it passes through the station and connects to
the main line, where it joins other trains which have been assembled
at similar neighborhood loops. At the destination area, similar
disassembly occurs, with the added caveat that the train may have to
enter multiple different neighborhood loops in a given destination
area.

None of this extra looping or stopping happens off hours, when demand
is light. But during the peak rush, a little extra moving and stopping
would be tolerable if it improves overall throughput. And the
complexity is completely within the system itself, with none of the
human factors that a GRT system would need to account for. The added
delays of train formation would be akin to GRT station delays waiting
for a GRT vehicle to fill up.

For these reasons, I think PRT trains is a better overall solution
than pure-GRT or GRT/PRT hybrid, even though GRT still wins on pure
capacity.

Mike C

[Luca Guala]

Mike
Running vehicles of different size and dynamic performance is a
computational nightmare at the current point of development of most
softwares (and also those who claim they can, are surely happier to have all
identical vehicles). For this reason, I think that ANYTHING is better than
running large GRT mixed with PRT. I consider instead a GRT as a "shared
ride" PRT. The 2getthere vehicle is made to host 4 but since the seats are
benches, you can squeeze a bit and accommodate 6. For a short-is trip to
work with colleagues this is perfectly acceptable and better than standing
in a bus anyway.
This means an improvement of 5 times over "standard" PRT operation.
Accounting for a non complete filling of vehicles, you can still rely on
three times more persons transported per vehicle than with "standard" PRT.
Station design is essential but I think this is an easy way out of the
dreaded PRT capacity problem. If the "shared PRT's" can be routed in trains,
it would be even better
Regards
Luca



-----Messaggio originale-----
Da: transport-...@googlegroups.com
[mailto:transport-...@googlegroups.com] Per conto di Mike C
Inviato: lunedì 3 agosto 2009 18.17
A: transport-innovators
Oggetto: [t-i] GRT vs PRT with Train Formation

.......

[Ian Ford]

I'm not convinced GRT wins on capacity. In the 2-station linear case,
the larger the stations and vehicles, the more linear capacity there is,
but in the network case, it is not clear.

A large GRT network might be a grid of separate GRT corridors, with no
routing options (if anyone can think of an alternative to this, please
say so). In this case, a person might have to make 2-4 unwanted stops on
a typical trip including one transfer. Capacity is limited by station
size, and each person is wasting a bunch of station space/time on each
trip. To beat the equivalent of 2-3s PRT headway, GRT might have to have
vehicles spaced under 10s. With the longer load times, it seems like a
lot of berths would be needed to feed that many vehicles onto the track.
I don't know if it can be calculated without a simulation.

Mike C wrote:
> For these reasons, I think PRT trains is a better overall solution
> than pure-GRT or GRT/PRT hybrid, even though GRT still wins on pure
> capacity.
>

--
Ian Ford i...@ianford.com 505.246.8490

[Jerry Roane]

Mike

This is an important topic from a machine design and guideway design standpoint.  If the cars are coupled then won't the guideway need to be that multiple times as strong to carry the multiple car weight?  Guideway spans are determined by the maximum load each span will see plus some "safety factor" multiplied with that.  If we assume that spans are infinitely short and guideway strength is no cost factor then the forming of trains makes sense in the Z axis but what about the vector along the axis of the guideway?  If the cars are designed to be cars they will not have to have the internal beam strength to stop 50 cars on one side and 50 cars on the other.  I use 100 as a simple example not really realistic.  A regular train has such a capability.  You can string along 100 cars in a train and the beam strength of each and every train car is built to withstand the forces in both hill climb and stopping.  We already have trains and they are heavy, slow and do not compete with trucks for service and are poor at generating business for themselves.  I have had a UPS and FedEx person call my business and offer to help me move packages and give me preferred rates etc. for their services.  No train business has ever called me looking to sell me their service.  The train is slow because it is heavy it is heavy because of the thought that you can string 150 of them together.  For what trains do well which is carrying heavy objects they are great.  For moving people they are inferior on all counts.  Why would we suggest that a new paradigm in transportation take on the worst parts of the old system.  I could almost be talked into a poorly designed PRT car being two together but certainly not three and never 100.  The financial burden of making a guideway that can hold up the weight of 100 cars in the length of 100 cars would be out of the limits that the public can afford to pay.   It is possible that a two-car train would not break the bank but since almost all cars have one or two passengers aren't you just doing car pooling that can be done by agreement between two parties of two or less in this information age without doubling the cost of the guideway by making the guideway twice as strong as it has to be? 

Jerry Roane

On Mon, Aug 3, 2009 at 11:17 AM, Mike C <mwil...@gmail.com> wrote:

Ingmar Andreasson's raises the possibility of forming PRT trains in
his recent paper. I think PRT trains is a better option

 

snip

[Dennis Manning]

Even if GRT had some capacity advantage of which I also have doubts it
doesn't seem like the lower service level would be worth it. You wait longer
to board, you don't know where the stops will be, you don't know how long
the trip will take, etc. Assuming PRT and GRT have same max speed you can be
sure that the trip will take longer, and for the same system cost the number
of access points will be lower.

What is size of the "G"? How many passengers should a GRT vehicle
accommodate before the logic for it (if there is any logic to it) breaks
down and your back to the problems of "mass transit"? If there are any GRT
fans out there can you give us a number and why that's the opitimum.

Dennis

[Dan Homerick]

Just to be clear about definitions, when I'm referring to GRT I'm
referring to automated transit on dedicated ROW (or captive to a
guideway) that makes multiple pickups/dropoffs during the course of
its trip. Unlike PRT, a trip may not have a clear sense of beginning
and end, in that the vehicle may not become fully empty for long
stretches of time. I think of GRT as a description of the service
type, not of the vehicle size, thus it's possible to run PRT service
with large capacity vehicles, and to run GRT service with low capacity
vehicles. That said...

It's easy to see GRT as a compromise, after all, passenger travel time
will be higher than could be achieved with PRT due to the additional
stops.

But it's also easy to see that GRT could be a big step up over BRT
and other traditional mass transit. Available 24/7 and on-demand
rather than upon a fixed schedule, with no requirement for transfers
or their associated layovers, and being potentially much greener due
to electrification, and relatively few stops. What's more, during
off-peak hours, you get something close to PRT level of service, the
exact opposite of what happens with BRT or traditional transit, where
service quality typically declines off peak.

In cases where the capital outlay for a large number of PRT vehicles
would be too high, or in cases where the ROW for a PRT-like mesh
network is either too expensive or simply not available, I think GRT
can make a lot of sense. If, during early phases of buildout, you know
that you're going to have only a handful of stations (say, 5 to 10
stations, making just one loop) in a high-density area, then GRT may
be a better fit than PRT from a service quality perspective. While PRT
would have a lower travel time, GRT may offer a lower wait time for
the same amount of capital invested.

In another case, if you have a very low guideway cost and can afford a
large network into lower density-areas, a pure PRT service may not
fare as well as a GRT when faced with a 'flash crowd' that depletes
the small number of vehicles in the vicinity, and leaves a long wait
before more can arrive.

I think cellphones can play an important role in making for a
successful GRT. If you send a message that gives both your origin and
your desired destination some 10 minutes before you make a trip, it
will allow GRT trip planning to optimize which vehicle picks you up.
You would still be faced with more stops than PRT, but at least there
would be no risk of a large detour before you are dropped off.

I think PRT is ideal when you have a case of 'many to many'. Many
origins, and many destinations. But I also think that GRT has a role
to play when you have a case of 'many to few' -- for example, at an
airport, where a huge parking lot area is being fed to just a handful
of terminals.

It's tempting to say that GRT also has a role in the case of 'few to
few'. Places where traditional mass-transit could be a viable (or
almost viable, at any rate) -- few destinations, but high traffic
among them. Upon further reflection, I don't think it does, however.
Instead, I think non-stop service but with larger, shared vehicles is
a better choice in this case.

Cheers,
- Dan

[Dennis Manning]

I seems to me that you're making the case that GRT would be better in
smaller special case situations, or just fat PRT that can carry about 6-8
passengers and used in GRT mode part of the time? The vehicle size would be
determined by the specific application? Yes? No? Another question comes to
mind - what's the maximum number of stations suitable for a GRT network? It
seems like the larger the number of stations the more difficult it is to
write the rules of when to stop and pick up a passenger and what route to
take to deliver passengers going to multiple and changing destinations.

[Mike C]

Jerry,

I was thinking more along the lines of 5-10 vehicle trains, not 100.
But your point is still valid: tighter packing of vehicles means the
guideways need to be stronger. The question is, how much stronger? If
I recall correctly, PRT guideway already needs to be able to support
bumper-to-bumper traffic, because a failure might stack vehicles that
closely even if the normal distance is much larger. Perhaps the added
infrastructure strength required to run in trains is not significant
relative to the capacity benefit.

Mike

[Mike C]

One problem with this approach is what to do when the system expands
and many-to-few becomes many-to-many. Do you run hybrid PRT/GRT or
convert to pure PRT? Either option would be more difficult than
starting with PRT and using trains to get the capacity while the full
network is phased in.

Think of the typical light rail line proposal: they typically spec out
at around 3k-6k passengers per hour maximum, and run along a single
line. Now, instead of LRT, we might a loop of PRT covering the entire
segment of the LRT line. Just a single loop. Have stations on either
side of the street, as LRT does, but spaced more closely together. Use
stations which can accommodate 5-10 vehicle trains. This would have a
capacity in the range of 5k-6k passengers per hour, for a cost that is
likely less than the LRT, and providing unobtrusive 24x7 service.

The PRT would provide better service than the LRT, at lower cost, and
with greater potential for expansion.

Now, I'm not really suggesting this as a solution, but it's a way to
show how PRT can compete head-to-head with light rail, even fully on
LRT's terms. This could be the "snappy answer" to skeptical
transportation engineers expressing doubts about capacity. Once we
convince people that PRT can compete with LRT in capacity, then a
better network can be drawn up.

Others have suggested using GRT for this purpose -- i.e. getting a
starter system built that can handle capacities immediately while
providing a path to true PRT service. But PRT trains can do it too,
and the transition to larger networks later will be nearly seamless.

Mike

[Jack Slade]

Mike, you are correct that PRT guideway needs to be stressed so that it can support vechicles rhat stop end-to-end.  However, just supporting the weight between support pillars without breaking is different from stressing it to provide smooth travel for only one vehicle.   For example:  If one  moving vehicle causes vertical track movement of 1/2 inch, that might be a smooth ride,  but if there were 5 platooned vehicles between the same supports the track deflection moght be 1and1/2 inches....still below breaking point, but a bumpy ride.   This is what would cause the necessity to make the guideway stronger.   Jack Slade --- On Tue, 8/4/09, Mike C <mwil...@gmail.com> wrote:

From: Mike C <mwil...@gmail.com> Subject: [t-i] Re: GRT vs PRT with Train Formation

> > his recent paper. I think PRT trains is a better option > > snip

[Bruce Attah]

On Aug 4, 4:37 am, Dan Homerick <danhomer...@gmail.com> wrote:

> In cases where the capital outlay for a large number of PRT vehicles
> would be too high, or in cases where the ROW for a PRT-like mesh
> network is either too expensive or simply not available, I think GRT
> can make a lot of sense. If, during early phases of buildout, you know
> that you're going to have only a handful of stations (say, 5 to 10
> stations, making just one loop) in a high-density area, then GRT may
> be a better fit than PRT from a service quality perspective. While PRT
> would have a lower travel time, GRT may offer a lower wait time for
> the same amount of capital invested.

Has anybody done the math on this? What's the envisioned traffic
demand? How many PRT vehicles would be required in order to meet that
demand? How many GRT vehicles would replace them? What difference in
cost is envisioned between PRT vehicles and GRT vehicles, PRT
guideways and GRT guideways, PRT station berths and GRT station
berths?

[Bruce Attah]

I suspect that the technical problems and risks involved in "PRT
trains" are bigger than those involved in reducing headways to about 1
second, or even 0.5 seconds. Pair-coupling I can see as a valid
possibility, but any train longer than a pair would be frought with
problems, I think.

> > >> Ian Ford            i...@ianford.com           505.246.8490- Hide quoted text -
>
> - Show quoted text -

[Bruce Attah]

On Aug 4, 6:46 am, Mike C <mwill...@gmail.com> wrote:

> Think of the typical light rail line proposal: they typically spec out
> at around 3k-6k passengers per hour maximum, and run along a single
> line. Now, instead of LRT, we might a loop of PRT covering the entire
> segment of the LRT line. Just a single loop. Have stations on either
> side of the street, as LRT does, but spaced more closely together. Use
> stations which can accommodate 5-10 vehicle trains. This would have a
> capacity in the range of 5k-6k passengers per hour, for a cost that is
> likely less than the LRT, and providing unobtrusive 24x7 service.
>
> The PRT would provide better service than the LRT, at lower cost, and
> with greater potential for expansion.
>
> Now, I'm not really suggesting this as a solution, but it's a way to
> show how PRT can compete head-to-head with light rail, even fully on
> LRT's terms. This could be the "snappy answer" to skeptical
> transportation engineers expressing doubts about capacity. Once we
> convince people that PRT can compete with LRT in capacity, then a
> better network can be drawn up.
>
> Others have suggested using GRT for this purpose -- i.e. getting a
> starter system built that can handle capacities immediately while
> providing a path to true PRT service. But PRT trains can do it too,
> and the transition to larger networks later will be nearly seamless.
>
> Mike

I'd have thought that in a small network (with a few stations), a PRT
system with 3 second headways (1200 veh/hr) would be able to meet a 3k
passengers per hour peak demand merely by encouraging ridesharing.

[Carl Henderson (SD)]

Slight correction Jack.  The guideway would need to be made stiffer to smooth the ride not necessarily stronger.

 

Running groups of vehicles one span apart could also help stiffen the original structure.

 

~Carl

[Jerry Roane]

Mike

As Jack and Carl have talked about the beam material properties being more complex than a single number.  There are a lot of considerations that go into the beam design and the closest you can pack vehicles without twisting them sideways is nose to tail.  The beam sag is one parameter that goes into beam design and because these cars are moving at high speed in our case a small sag will offer a challenge to our active powered pre-knowledge suspension.  Our beam is extremely strong for ultimate breaking strength and it won't take a set (stay bent or kinked) till you take out a few poles in a row.  In the case of crashing into a row of poles the guideway needs to be recycled anyway to insure the integrity but the number of places where a ground traffic crash will take out three poles in a row at 60 foot spacing will be few and the recycling of the guideway with automated equipment will be no big deal.  It will only be no big deal if we can keep the size down.  Weight compounding is the problem for making trains out of bullet cars.  Everything goes up in price and weight on the high end you have trains like the 1800s.  I contend that if we keep weight and complexity under control then the guideways are low enough cost and visually small enough that we can just build enough to meet demand.  No need for car pooling or platooning or cell phone apps trying to figure out if your wife and preteen child is riding with a sex offender.  All that complexity drops off if we give up on the concept of car pooling and GRT.   It does mean we build more cars.  It does mean we have significantly more elevated guideway miles but if mobility is the goal then more is better.  If we are concerned about running out of raw materials to build cars and guideway we have a long way to go before we use up the recycled gasoline cars and extra highway steel after the transition to no traffic transportation. 

The other item on beams is how ductile it is.  Our beam will be able to sag completely to the ground before a break as you remove poles from under it.  Think of the reinforcing bars inside our guideway as the cables that hold up the Golden Gate bridge.  Those steel cables sag several hundred feet and then they add the weight of the bridge on top of that.  The guideway will hold up cars tip to tail but the beam sag would not be a comfortable ride at 180 mph all jammed together like that.  

On the GRT concept.  People are rarely in groups so why would you plan to transport them in groups?  If machines are the servant of man and not man the servant of the machine is seems to me that GRT is not the way to go.  What market need is asking for GRT?  I see it as a compromise based on an assumption that we don't have enough capability so just tough it out and live with it. To the customer what is the advantage of GRT? 

Jerry Roane

[Jerry Schneider]

At 10:08 AM 8/3/2009, Luca G. wrote:

>Mike
>Running vehicles of different size and dynamic performance is a
>computational nightmare at the current point of development of most
>softwares (and also those who claim they can, are surely happier to have all
>identical vehicles).

Didn't the German Cabintaxi system do this decades ago - without particular
difficulty so far as I know) using modular
designs? Most of the pictures I have
show only the smallest vehicles in operation but I think they may have built
and tested some larger ones too and used them in together on their test track.
I'll ask Marsden to comment.

- Jerry Schneider -
Innovative Transportation Technologies
http://faculty.washington.edu/jbs/itrans

[Marsden Burger]

Hello All,

 

Summoned up by Jerry to comment, I would refer this discussion back to the material that was sent to this board in September of last year. A good bit of discussion dealt with this topic of GRT/PRT.

 

To the issue of control technology being able to run different size vehicles, I would suggest that that has a great deal to do with the basic system concept that is being utilized.  Many "unknowns" are often hidden in a given system design that make the flexibility of systems less than an "outside" party would want.  Without being within a given system development team, it is often very hard, close to impossible, for someone on the outside to ever know "why/or if" a given development effort has any glitches, as it is the job of marketing to normally keep that type of information distracted from the potential customer - "Oh, that is not an issue for us, as we do this...." and in reality, it may not be an issues because one system approach has a whole different list of issues as another – a control problem for one system may not be an issue at all for another.

 

In the case of the Cabintaxi technology, there were no issues with running the different size vehicles that caused it to be something that we would not have been willing to have done. Again, Cabintaxi GRT and PRT vehicles were all the same cross section.

 

The video as seen on YouTube shows the two vehicle sizes operating over the same track and utilizing shared station platforms.  With the Cabintaxi system approach, its total system design approach, varying size Cabintaxi vehicles did not represent a problem.

 

In line with this topic, the material sent last September also showed the difference in simulation studies for the Hamburg application and dealt with the PRT/GRT differences.  One of the aspects of the studies was to determine if it was better to operate the GRT vehicles in married-pairs or as singles.  What was found from that work was that vehicles at three second separations were in-effect electronically coupled, and that the actual married-pairing “or training” gave little advantage over single vehicle operation where greater system flexibility was found to be more desirable.

 

Best wishes,

 

Marsden

[Jerry Schneider]

The Urban Mass Transit Administration and their consultants did a lot
of analytical
work on GRT back in the 70's and I suspect that the German Cabintaxi people did
as well. I don't have any ready on-line references but some digging
could probably
turn them up. There was a big battle between the feds and the pure
PRT people at
the time and feds had the money and gov. backing and it was GRT that got the
green light - resulting in Morgantown GRT and the Detroit, Miami and
Jacksonville
downtown people movers and a large number of airport systems that have since
been built.

[Jerry Schneider]

At 02:02 PM 8/4/2009, you wrote:
>Hello All,
>
>Summoned up by Jerry to comment, I would refer
>this discussion back to the material that was
>sent to this board in September of last year. A
>good bit of discussion dealt with this topic of GRT/PRT.

I think that this material comes from the
Cabintaxi studies done for the Hamburg project.
Links to them are provided at the Cabintaxi page:
http://faculty.washington.edu/jbs/itrans/cabin.htm
Also, I think I see three different vehicle sizes
on the video, the largest one is for 12 passengers, I think.

>
>To the issue of control technology being able to
>run different size vehicles, I would suggest
>that that has a great deal to do with the basic
>system concept that is being utilized. Many
>"unknowns" are often hidden in a given system
>design that make the flexibility of systems less
>than an "outside" party would want. Without
>being within a given system development team, it
>is often very hard, close to impossible, for
>someone on the outside to ever know "why/or if"
>a given development effort has any glitches, as
>it is the job of marketing to normally keep that
>type of information distracted from the
>potential customer - "Oh, that is not an issue
>for us, as we do this...." and in reality, it
>may not be an issues because one system approach
>has a whole different list of issues as another
>– a control problem for one system may not be an issue at all for another.
>
>In the case of the Cabintaxi technology, there
>were no issues with running the different size
>vehicles that caused it to be something that we
>would not have been willing to have done. Again,
>Cabintaxi GRT and PRT vehicles were all the same cross section.
>
>The

><http://www.youtube.com/watch?v=ERdF0FK-2io>video

[Jack Slade]

Thikk this over, Carl: If we have not already made the guideway as stiff as we can,  then we have wasted material, or not done the engineering properly.   Assuming we have done it right, there are only 2 options for stiffening. One is adding more support pylons, shorter spacing.  The other is build a bigger guideway. Niether is in compliance with my thinking.   We are trying to present people with a system that they find more acceptable than present systems.  All the advantages of PRT is aimed at accomplishing this. Now we are going to change it to a train of vehicles.   There goes: Non-stop travel                   Faster travel                   Non-scheduled service                   Personal service.... sharing vehicle with strangers                   No wait service...trains don't fill up immediately, by magic                   More obtrusive guideway.   Have I missed any?  Too many people are trying to improve something that has been thought through pretty well . In fact,  I would say our concept is damn near perfect.  Stop trying to fix something that isn't broke, or at least wait till it gets built.  Then you can see if improvements make sense.   Here are my thoughts on brickwall stops.  Baloney!!  The world needs better, transport.  If it is possible to provide it with 1/2 second hearways,then the world will damn well accept it or do without it. People pay 5 bucks for a ride on a roller-coaster. Heard any complaints...too fast, too close together, etc?  The only reason we are getting such crap is from people with overactive imagination, who are not desparate yet because we still have some oil left. So, do nothing, and just wait.   Doesn't anybody, especially the Vuchics, realize that I am talking about the capability of 3 lanes of traffic on just one little, out-of-everybodys way guideway.  7200 vehicles per hour, at constant speeds. Just try that cheaper with trains, roads,  or anything else in use. I guess I am copying Jerry's "Rant for today". Thats it.   Jack Slade --- On Tue, 8/4/09, Carl Henderson (SD) <carl.he...@silvertipdesign.com> wrote:

[eph]

Been away for a while...

PRT trains do have advantages (as has been pointed out).
- All the vehicles can be the same and so stations can be shared
- Vehicles are used more efficiently - less PRT available at rush
hours, but frequent trains are available.
- Perhaps 4 lines can be squeezed into a single LRT/BRT sized ROW.
- Single vehicle to build and test
- Guideway simplified since only one vehicle type.

Disadvantages:
- Likely no standing room (unless we have narrow, tall trains perhaps
2 lines to LRT/BRT ROW)
- Size limited to a single platform size, so likely limited DM
possibility.
- Extra switches and switching needed.
- An automatic coupler or virtual coupler is needed
- Routing complexities.
- Availability - someone mentioned Vectus working on it, any other
viable candidates?

F.

[Jack Slade]

No Standing Room is a Disadvantage? It is one of PRT Advantages or at least you would think so if you had to ride TTC in Toronto every working day.   Jack Slade --- On Tue, 8/4/09, eph <rhaps...@yahoo.com> wrote:

From: eph <rhaps...@yahoo.com> Subject: [t-i] Re: GRT vs PRT with Train Formation

[eph]

Standing Room is a question of space and efficiency. Yes it's at the
expense of passenger comfort, but cities don't seem to care about that
much. It's a minor point I'm willing to concede. Even if you
postulate that PRT would be as cheap or cheaper, a similar standing-
room/GRT system would likely be even cheaper.

F.

On Aug 4, 7:25 pm, Jack Slade <skytrek_...@rogers.com> wrote:
> No Standing Room is a Disadvantage? It is one of PRT Advantages
> or at least you would think so if you had to ride TTC in Toronto every working day.
>  
> Jack Slade
>

> --- On Tue, 8/4/09, eph <rhapsodi...@yahoo.com> wrote:

[Jerry Schneider]

At 04:16 PM 8/4/2009, eph wrote:

>Been away for a while...
>
>PRT trains do have advantages (as has been pointed out).
>- All the vehicles can be the same and so stations can be shared
>- Vehicles are used more efficiently - less PRT available at rush
>hours, but frequent trains are available.
>- Perhaps 4 lines can be squeezed into a single LRT/BRT sized ROW.
>- Single vehicle to build and test
>- Guideway simplified since only one vehicle type.

I think Cabintaxi has shown that all vehicles do not necessarily have
to have the same
capacity, but do need to be able to use a common guideway. All that is
needed is a modular design.

[Bruce Attah]

Say you have a train made up of 10 PRT vehicles, and it's in the
station, and it's full of passengers, and it's ready to go. The
network is very busy, because it's peak time. How long will the train
have to wait until there is a gap big enough in the main line for the
train to safely merge? While it is waiting, will it be blocking
vehicles behind? I think it's very likely that all the hoped-for
increase in capacity will be cancelled out by issues like this. It
might actually turn out that capacity is lower with PRT trains than
without. Even if the effect on capacity is not negative, I doubt that
it will be hugely positive. It might work in a very high speed
network, where the length of the vehicle does not have much effect on
the headway, but I doubt that it would be worth the extra technical
complexity, not to mention the huge reduction in quality of service to
the passengers.

I see the ghost of Aramis floating through this thread.

May I suggest that instead of PRT trains, we think about this: if the
vehicles can reliably brake at 1 g, and the brick wall rule is obeyed,
a headway of about 1 second is possible at 25 mph/40 kph, and 2
seconds at 80 mph/125 kph. This gives a line capacity of 1800 to 3600
vph (or, assuming 2 passengers per veh avge at peak times, 3600 to
7200 pphpd), higher than most services demand (leaving aside metro
rail in large cities). PRT at any speed will generally give trip times
that match those offered by a train or tram running at a nominal
operating speed 50% higher, as a result of PRT's non-stop operation
(tortoise and hare, you could say). A PRT system with the above
characteristics, which could be described as technically fairly
unambitious, could go head to head with both light rail and intercity
rail, and win eight times out of ten.

> > > Mike C- Hide quoted text -

[Bruce Attah]

On Aug 5, 12:50 am, eph <rhapsodi...@yahoo.com> wrote:

> Standing Room is a question of space and efficiency.  Yes it's at the
> expense of passenger comfort, but cities don't seem to care about that
> much.  It's a minor point I'm willing to concede.  Even if you
> postulate that PRT would be as cheap or cheaper, a similar standing-
> room/GRT system would likely be even cheaper.
>
> F.

I don't think you will make the system cheaper by having standing
room. Standing room means more weight per unit length of guideway --
maybe twice as much -- which is likely to mean a heavier, more
expensive guideway. The guideway might be only a little bit heavier,
but it might cancel out any savings you thought you were going to make
on the number of vehicles required. Standing room also means you have
to break and accelerate gently, and this will increase the length of
on- and off-ramps.

[eph]

Aramis was very interesting.

PRT train:
4 m vehicles
80 m platforms (equivalent to 3-car LRT)
20 vehicles, 4 passengers per vehicle (6 max?),
80 people per train,
90 second headways
3200 pphpd but 4 lines per ROW = ~12,800 (optimistically)
Shorter PRT vehicles (3 m) 4,000 pphpd or ~16,000

LRT:
80 m platform
3 car train
~15,000 (2 min. headways)

Urbanaut:
98 seated + 114 standing per 40 m station (doubled for 80 m stations)
90 sec headways
~ 17,000

With space for regular PRT in between and I would assume stations
would be off-line.

Hmmm.

F.

[Jerry Roane]

Bruce

This list is about advanced transportation.  1 G deceleration is what a present non-advanced car does on pavement.  I think we can and should bump up the braking performance if we allow ourselves to bump up the speeds.  Following your thought with better than 1 G braking the throughput would be even higher.  A planned trip across town using a central routing computer will certainly beat trains with park-n-ride parking lots or worse bus circulators that have to make a loop regardless of who is on the bus and which direction around the loop would be quickest for that bus load. 

A combination of Jack's suggestion that brick wall is baloney and more aggressive braking, comparisons with big box leaves big box the looser on every trip in the city.   

Jerry Roane

[Michael Weidler]

I have found that the shape of the seat makes a big difference in how much g force a passenger can experience before needing to be strapped down. I have mentioned before that I routinely made 25mph 90 degree turns in my Intrepid with no difficulty. I have made - or rather attempted to make - the same turns in the Escort I am now driving. The seats are both "bucket" designs, but I tend to "fly around" more in the Escort. --- On Tue, 8/4/09, Jerry Roane <jerry...@gmail.com> wrote:

[Kirston Henderson]

on 8/4/09 10:58 PM, Michael Weidler at pstr...@yahoo.com wrote:

I have found that the shape of the seat makes a big difference in how much g force a passenger can experience before needing to be strapped down. I have mentioned before that I routinely made 25mph 90 degree turns in my Intrepid with no difficulty. I have made - or rather attempted to make - the same turns in the Escort I am now driving. The seats are both "bucket" designs, but I tend to "fly around" more in the Escort.

   Your Intrepid suspension system tended to remain very flat in turns, something that most cars do not do as well.   I have owned several Chrysler cars of that same basic design and suspension and, in fact, we are still driving a couple of them on a daily basis.  Those cars had much longer wheelbase and track width than your Ford Escort, something that leads to greater stability and much better handling.  We once bought a small Dodge with short wheelbase and narrow track of that same vintage for one of our sons to drive to College.  I drove the thing only once and when I got back home, I remember telling my wife that I felt that I had just driven a riding lawnmower.  (I never drove the thing again!)  Our youngest son inherited the car when he started driving to high school.  After a time, I found that he made a habit of taking back streets to high school rather than I-30.  The reason was that he felt unsafe driving the thing.  After it finally fell apart, we replaced it with a Chrysler Concorde of the type that my wife and I were driving and he then felt comfortable driving it on freeways.

   Wheelbase length and track width make a very large difference in handling.  For example, the current Chrysler 300 series of cars have a wheelbase of a very long 120-inches are really handle nice on curves.  It's a shame that the Germans who controlled the company when that car was brought out didn't understand that down here in the Texas heat, a lot of us will drive nothing but white cars and that is the reason that I continue to drive my 2004 Concorde Limited.  When it finally wears out, I may have to walk because the very thought of driving some tiny Obomamobile scares the daylights out of me!

Kirston

[Luca Guala]

Standing passengers mean lowering significantly accelerations and
decelerations and hence having longer headways. Not something that benefits
PRT/GRT operation
Regards
Luca

-----Messaggio originale-----
Da: transport-...@googlegroups.com
[mailto:transport-...@googlegroups.com] Per conto di Bruce Attah
Inviato: mercoledì 5 agosto 2009 3.16
A: transport-innovators
Oggetto: [t-i] Re: GRT vs PRT with Train Formation

[Jerry Roane]

Michael

The seat does make a big difference.  I am toying around with the concept of a padded lap bar that rotates down to hold your waist from a pivot near the floor.  It is an adaptation of a roller coaster restraint system but will need to be more adjustable than one for the general public.  At high G stops it will be needed for ultimate safety but at regular speeds it will make you feel more secure and slip around less on the seat upholstery.  I am positive I want a padded lap bar.  How it rotates into position I am still tweaking in my mind.  Because the triangular guideway goes right up the middle of the car it is most likely going to pivot along the two planes of the triangle.  The center triangular tube that runs the length of the car is a very strong feature so mounting the pivot mechanism on that structure is where my thoughts are right now.  Since I do not have a steering wheel in the way with drive by joy stick the driver seat and passenger seat restraint will be mirror images of each other. 

The trick will be to make this look like it belongs in the interior design.  The styling and industrial design of this bar will be crucial to its acceptance by the public.  They are not going to give up the seat belt and lop sided shoulder strap easily after all these years of "education" to get you to wear them.  Naturally we will offer ordinary seat belts with the one sided shoulder harness as a very high priced option.  We will also offer a traditional steering wheel as an even more expensive option for those who just can't give up the steering wheel.  Change is hard and it will take time to move away from the mechanically linked steering Pittman arm device. 

I got great news from my guy in China on the composite body shell price.  Outsourcing this part we can go full graphite honeycomb and still be low cost.  That may cut a few more pounds off this car. 

Jerry Roane  

[Robbert Lohmann]

@ Jerry: outsourcing to low cost countries has great benefits; be
careful about quality though - we learned the hard way.

@ Luca: running different vehicles under one control system is no
issue; we have done this 10+ years ago in industrial environments
where there were different vehicles to carry different types of loads.
The logic for a passenger transit system would not be any different.

In general: I do believe in GRT. It will be more suited than PRT in
certain applications because it will help keep the capital costs
lower. In some cases this will be at the expense of the service level
that is being offered, but in other cases this will not make a
difference (a good example being the Rivium application). Please note
that GRT in my definition is vehicles accommodating anywhere from 6 to
25 passengers. Anything above this figure is MRT (Mass Rapid Transit).

I also believe in applications featuring both PRT and GRT. PRT would
offer on-demand service, direct origin-to-destination trips,
comparable to a taxi (requiring passengers to pay for use of the whole
vehicle, whether they are alone or in a group). GRT would operate on-
schedule, stopping at all stations (that are indicated as destination
or where transport is requested, skipping other stations) comparable
to a bus service (and thus allowing passengers to pay for their seat
only). GRT should be cheaper to ride (as there are less advantages)
and would be attractive to the masses, while PRT would serve the
passengers willing to pay more.

The control system will accommodate the set-up, the infrastructure is
another story. Any section of the track that needs to be accessible to
the GRT system (which is wider), would need to be sized for the GRT
system. Hence the routing and set-up of the infrastructure is probably
the most vital part to allow such a system to be realized.

You don't have to agree with me on the vision, but believe me when I
say we are in the position to realize it from a technical perspective
today.

Robbert

> On Tue, Aug 4, 2009 at 10:58 PM, Michael Weidler <pstran...@yahoo.com>wrote:
>
> > I have found that the shape of the seat makes a big difference in how much
> > g force a passenger can experience before needing to be strapped down. I
> > have mentioned before that I routinely made 25mph 90 degree turns in my
> > Intrepid with no difficulty. I have made - or rather attempted to make - the
> > same turns in the Escort I am now driving. The seats are both "bucket"
> > designs, but I tend to "fly around" more in the Escort.
>

> > --- On *Tue, 8/4/09, Jerry Roane <jerry.ro...@gmail.com>* wrote:

>
> > From: Jerry Roane <jerry.ro...@gmail.com>
> > Subject: [t-i] Re: GRT vs PRT with Train Formation
> > To: transport-...@googlegroups.com
> > Date: Tuesday, August 4, 2009, 7:27 PM
>
> > Bruce
>
> > This list is about advanced transportation.  1 G deceleration is what a
> > present non-advanced car does on pavement.  I think we can and should bump
> > up the braking performance if we allow ourselves to bump up the speeds.
> > Following your thought with better than 1 G braking the throughput would be
> > even higher.  A planned trip across town using a central routing computer
> > will certainly beat trains with park-n-ride parking lots or worse bus
> > circulators that have to make a loop regardless of who is on the bus and
> > which direction around the loop would be quickest for that bus load.
>
> > A combination of Jack's suggestion that brick wall is baloney and more
> > aggressive braking, comparisons with big box leaves big box the looser on
> > every trip in the city.
>
> > Jerry Roane
>

> > On Tue, Aug 4, 2009 at 8:01 PM, Bruce Attah <bruce.at...@googlemail.com<http://mc/compose?to=bruce.at...@googlemail.com>

> >> On Aug 5, 12:50 am, eph <rhapsodi...@yahoo.com<http://mc/compose?to=rhapsodi...@yahoo.com>>

> >> wrote:
> >> > Standing Room is a question of space and efficiency.  Yes it's at the
> >> > expense of passenger comfort, but cities don't seem to care about that
> >> > much.  It's a minor point I'm willing to concede.  Even if you
> >> > postulate that PRT would be as cheap or cheaper, a similar standing-
> >> > room/GRT system would likely be even cheaper.
>
> >> > F.
>

> >> > On Aug 4, 7:25 pm, Jack Slade <skytrek_...@rogers.com<http://mc/compose?to=skytrek_...@rogers.com>>

> >> wrote:
>
> >> > > No Standing Room is a Disadvantage? It is one of PRT Advantages
> >> > > or at least you would think so if you had to ride TTC in Toronto every
> >> working day.
>
> >> > > Jack Slade
>

> >> > > --- On Tue, 8/4/09, eph <rhapsodi...@yahoo.com<http://mc/compose?to=rhapsodi...@yahoo.com>>
> >> wrote:

>
> >> > > From: eph <rhapsodi...@yahoo.com<http://mc/compose?to=rhapsodi...@yahoo.com>
>
> >> > > Subject: [t-i] Re: GRT vs PRT with Train Formation

> >> > > To: "transport-innovators" <transport-...@googlegroups.com<http://mc/compose?to=transport-...@googlegroups.com>

>
> >> > > Received: Tuesday, August 4, 2009, 11:16 PM
>
> >> > > Been away for a while...
>
> >> > > PRT trains do have advantages (as has been pointed out).
> >> > > - All the vehicles can be the same and so stations can be shared
> >> > > - Vehicles are used more efficiently - less PRT available at rush
> >> > > hours, but frequent trains are available.
> >> > > - Perhaps 4 lines can be squeezed into a single LRT/BRT sized ROW.
> >> > > - Single vehicle to build and test
> >> > > - Guideway simplified since only one vehicle type.
>
> >> > > Disadvantages:
> >> > > - Likely no standing room (unless we have narrow, tall trains perhaps
> >> > > 2 lines to  LRT/BRT ROW)
> >> > > - Size limited to a single platform size, so likely limited DM
> >> > > possibility.
> >> > > - Extra switches and switching needed.
> >> > > - An automatic coupler or virtual coupler is needed
> >> > > - Routing complexities.
> >> > > - Availability - someone mentioned Vectus working on it, any other
> >> > > viable candidates?
>
> >> > > F.
>

> >> > > On Aug 3, 12:17 pm, Mike C <mwill...@gmail.com<http://mc/compose?to=mwill...@gmail.com>>

> ...
>
> read more »

[Mike C]

It's not about feasibility, it's about what regulators will approve.
Regulators will not (currently) allow short headways, but they
certainly do allow trains.

Mike C.

[Mike C]

On Aug 4, 9:01 pm, Bruce Attah <bruce.at...@googlemail.com> wrote:

> Say you have a train made up of 10 PRT vehicles, and it's in the
> station, and it's full of passengers, and it's ready to go. The
> network is very busy, because it's peak time. How long will the train
> have to wait until there is a gap big enough in the main line for the
> train to safely merge?

When the system knows in advance that it needs a 200-ft gap, it will
adjust vehicles accordingly to provide that gap. It's not like the 10-
pod train comes out of nowhere; the system knows it is coming.
Furthermore, in the time that passes as the train is assembling,
vehicles on the line are not merging those 10 vehicles, each of which
would require its own gap.

At 3s headway and 25mph, the per-single-vehicle gap would itself be
over 100 feet, and current systems manage that merge without
congestion. So the train approach would eliminate the need to create
10 gaps of 100 feet, and in its place a single gap of maybe 200-250
ft. That sounds like a win to me.

> Even if the effect on capacity is not negative, I doubt that
> it will be hugely positive. It might work in a very high speed
> network, where the length of the vehicle does not have much effect on
> the headway, but I doubt that it would be worth the extra technical
> complexity, not to mention the huge reduction in quality of service to
> the passengers.
>

First, train formation would only occur when and where it is required,
i.e. when the system senses that traffic is approaching the limits of
single-vehicle capacity.

Second, regarding the speed question, I've calculated the capacity of
trains at various train lengths, using typical assumptions (brick wall
stopping distances between trains, 0.4g emergency decel, 1.2 pax/veh,
etc) and at 5-vehicle trains, the max capacity is 6400 passengers at
25mph line speed. For single vehicles at the same assumptions, max
capacity is only 2700 and occurs at 12.5mph. In fact, at the more
typical 25mph of current systems, single-vehicle PRT capacity is only
2100. So 5 vehicle trains can triple capacity over single vehicles,
even at 25mph.

Third, regarding quality of service: if trains are forming, that means
the system is nearly saturated, meaning that vehicles would be waiting
longer to leave stations and passengers will be waiting in line for a
vehicle anyways. Trains would not only alleviate the capacity crunch,
reducing overall wait times, but the short waits that do occur are
inside the vehicles where the passengers are settled in, not standing
in a line.

> I see the ghost of Aramis floating through this thread.
>

Aramis was decades ago, and I think they proposed fully-dynamic train
formation, which is not what we're proposing here.

> May I suggest that instead of PRT trains, we think about this: if the
> vehicles can reliably brake at 1 g, and the brick wall rule is obeyed,
> a headway of about 1 second is possible at 25 mph/40 kph, and 2
> seconds at 80 mph/125 kph. This gives a line capacity of 1800 to 3600
> vph (or, assuming 2 passengers per veh avge at peak times, 3600 to
> 7200 pphpd), higher than most services demand (leaving aside metro
> rail in large cities). PRT at any speed will generally give trip times
> that match those offered by a train or tram running at a nominal
> operating speed 50% higher, as a result of PRT's non-stop operation
> (tortoise and hare, you could say). A PRT system with the above
> characteristics, which could be described as technically fairly
> unambitious, could go head to head with both light rail and intercity
> rail, and win eight times out of ten.
>

I agree in principle, but we're just starting to see 3-second systems
approved, and regulations have not budged on the brick wall
requirement, so PRT trains would be an attempt to achieve higher
capacities in today's regulatory environment, with a minimal long term
impact on design. If short headways are approved, then train formation
becomes unused code in the control program and nothing is lost (except
for some added guideway strength required for the trains, which may or
may not be significant). In the meantime, we will have a much more
compelling capacity argument.

Even at 0.5-second headways (if we ever get there), PRT pair formation
might be a win, providing up to maybe 25% capacity boost, so the added
complexity would not be completely wasted even in the long term.

Mike C

[eph]

True, so this means that the trains running every 90 seconds will have
longer headways. The rest of the vehicles can have shorter headways.
I get a <7 second gap (tail to nose) with 0.13g deceleration (same as
LRT) and a jerk factor of 12 from 60 km/h. This also means station
sidings of ~100 m or forcing any PRT vehicles to slow down a bit at
stations if a train is ahead of it.

F.

[Bruce Attah]

Mike,

I have looked at the UK regulations (spelled out in "The Railways and
Other Guided Transport Systems (Safety) Regulations 2006"), and the
guidelines published by Her Majesty's Rail Inspectorate, in particular
"Safe Movement of Trains", and have come to the conclusion that there
is no rigid brick-wall stop rule that applies to all transit systems
(including novel ones) and is enforced by the regulator in this
country. (For trams and for conventional trains under "permissive
operation", there is a principle in the guidelines that says trains
must be "operated in such a way and at such a speed to enable drivers
to stop within the distance they can be seen to be clear ahead" --
"driver" can mean whoever is in charge of a system that controls the
train remotely, and there's no assumption that the emergency brakes
have failed, which some people have said is a required assumption.)
For new systems, there is instead a risk assessment approach. Each
aspect and component of a new system must be thoroughly risk assessed
by independent experts to demonstrate that it is safe, and HMRI will
study the evidence collected, and, if satisfied, certify the system. I
do not know about other countries.

Anyway, I pointed out that quite high capacity can be achieved
*without* breaking the brick wall stop rule, if one merely installs 1g
brakes. This hardly requires a technical breakthrough, so it is not a
matter of waiting years for the technology to be developed. Depending
on the design of vehicle and track, 2g or even 3g braking in
emergencies is feasible using technology that is already available and
tested. If the system offers such hard braking, the cabin will need to
be designed to protect passengers from bruising themselves against its
interior and each other in the event of those brakes being applied,
but this again does not involve any novel or untested technology. It
seems to me that regardless of whether one has PRT trains or not, it
is a good thing to have good brakes, and if one has good brakes, one's
system will probably have more than adequate capacity anyway, so
moving the vehicles as trains may never be needed.

On the matter of safety, if there are no trains, then it is very
unlikely that an accident will affect more than one vehicle at a time,
and extremely unlikely that it will affect more than two, whereas if
there are trains, the chance that an accident will affect n vehicles,
where n is the length of a train, is nearly as great as the chance
that an accident will affect one vehicle. That's not good. The system
of coupling and decoupling will also have to be very solidly proved,
or it will be a source of danger to passengers. If the coupling is
merely electronic, then "train" is nothing but a euphemism for "very
small headway".

> > > - Show quoted text -- Hide quoted text -

[Jerry Roane]

Robert

"comparable
to a bus service (and thus allowing passengers to pay for their seat
only). GRT should be cheaper to ride (as there are less advantages)
and would be attractive to the masses, while PRT would serve the
passengers willing to pay more."

snip

This is a common thought in the transit world but it is untrue in general.  Show me where a bus seat costs less than 5 cents a mile TOTAL energy budget to ride.  I disagree with this common perception as we can come in under the price of big box transit as it exists in reality land.  Now if we talk about non-reality land where every seat if full on every route mile including the daily trip to the bus barn then the bus still looses.  The bus is so ineffective and energy inefficient and NOx polluting that it can't beat advanced electrified elevated guideway on the energy front and thus the cost front.  The ONLY reason the bus seat is cheap is you are not paying your own way.  I quote the Boy Scout Law number 9:

Thrifty_ A scout is thrifty.  He saves his money so that he may pay his own way.  The bus does not conform to this basic tenant of the Boy Scout law number 9 and it makes little physical sense either. 

There is a lot of energy represented in a bus seat as there is as much administration behind the bus, driver, mechanic, scheduler, diesel engine manufacturer lobbyist and CEO of the bus organization.  Dual mode/PRT streamlines this operating overhead that has to be accounted for by charging against the bus seat in energy and cost.  

It is possible to have GRT that has less bus-like properties if it is elevated (grade separated) and electrified. 

Jerry Roane

[Jerry Schneider]

At 01:24 AM 8/5/2009, Robbert L. wrote:
>snip -------------------------

>The control system will accommodate the set-up, the infrastructure is
>another story. Any section of the track that needs to be accessible to
>the GRT system (which is wider), would need to be sized for the GRT
>system. Hence the routing and set-up of the infrastructure is probably
>the most vital part to allow such a system to be realized.

Would the guideway have to be wider for a PRT/GRT system or could
the GRT vehicles be a little wider and longer. I'm thinking of the Cabintaxi
system which appears to use only one guideway width.

>You don't have to agree with me on the vision, but believe me when I
>say we are in the position to realize it from a technical perspective
>today.

It certainly needs to be considered and analyzed, IMHO.

[Jerry Schneider]

I note that the RUF system is a GRT/PRT concept, with maxiRUF vehicles
and smaller RUF vehicles using the same guideway, so far as I know.
And, in addition,
it offers both SM and DM services. Better than 3 lanes of SM in a
corridor and/or
PRT trains? But, I don't know what the capacity calculations show for
various combinations
of maxiRUFs and RUFs - but that would be interesting to see.

[Kirston Henderson]

on 8/5/09 10:59 AM, Jerry Schneider at j...@peak.org wrote:

> Would the guideway have to be wider for a PRT/GRT system or could
> the GRT vehicles be a little wider and longer. I'm thinking of the Cabintaxi
> system which appears to use only one guideway width.

The answer is no with regard to our MicroRail system. The same guideway
can accommodate PRT, GRT, CarFerries for small automobiles, and mass
transit trains on the same guideway.

Kirston Henderson
MegaRail®

[eph]

I agree, I can see PRT fetching a per km fee like a taxi, maybe $1 per
km (and cover it's costs) whereas GRT/PRT trains would fetch maybe $2
per trip (averaging perhaps 9 km each, still losing money because of
the bus feeder system). Taxi fare is $1.75/km plus a $3.30 initial
charge and bus fare is $2.30 (with tickets) in Ottawa.

F.

> ...
>
> read more »

[Michael Weidler]

Perhaps it would be a good idea to sue the regulators? If 2 seconds is accepted on the freeways, why would it not be acceptable for PRT? --- On Wed, 8/5/09, Mike C <mwil...@gmail.com> wrote:

[eph]

Seat belts and air-bags mean ~1g deceleration.

I wonder if sub-second headways or high-g stops could be used for
empty vehicles in a PRT application?

F.

On Aug 5, 10:09 pm, Michael Weidler <pstran...@yahoo.com> wrote:
> Perhaps it would be a good idea to sue the regulators? If 2 seconds is accepted on the freeways, why would it not be acceptable for PRT?
>

> --- On Wed, 8/5/09, Mike C <mwill...@gmail.com> wrote:

[Robbert Lohmann]

@ Jerry R.
Note I am comparing our PRT system with our GRT System (operating
similar to a bus). The comparison with regard to emissions thus is
more complex as both are electrically powered and in the end it will
depend on how the energy is generated that is supplied to the
vehicles. And note that you are talking about costs, while I am
referring to the charge administered to passengers! Not necessarily
related: you can actually elect to install a lower fare to stimulate
the usage of the GRT system to ensure the required fleet size (and
capital investment) are minimized.

@ Jerry S.
In our case a wider infrastructure is required as we are a supported
system operating on a basic tarmac infrastructure. The ParkShuttle
(GRT) is 2,1 meters wide, while the CyberCab (PRT) is 1,4 meters wide.
Both need some residual space on either side of the vehicle related to
our obstacle detection system and evacuation space. Other systems
might not need this additional space; depends on the technology.

You could also make the vehicle the same width and (much) longer, but
this will generate problems as the turning radius would increase
significantly; creating problems at curves, but especially at the
entry of off-line stations...

Robbert

[Mike C]

On Aug 5, 11:22 pm, eph <rhapsodi...@yahoo.com> wrote:
> Seat belts and air-bags mean ~1g deceleration.

I've said this before: 1g deceleration over a short duration (less
than 1-2 seconds) is very tolerable for seated passengers, as long as
it's a very rare occurrence (e.g. a catastrophic fault). I think it
can be done without restraints, but the key here is that it only
occurs in catastrophic faults.

>
> I wonder if sub-second headways or high-g stops could be used for
> empty vehicles in a PRT application?

Igmar discusses that possibility in his paper, along with train
formation and ride sharing.

>
> F.
>


Mike C

[Mike C]

On Aug 6, 4:40 am, Robbert Lohmann <robb...@2getthere.eu> wrote:
> @ Jerry S.
> In our case a wider infrastructure is required as we are a supported
> system operating on a basic tarmac infrastructure. The ParkShuttle
> (GRT) is 2,1 meters wide, while the CyberCab (PRT) is 1,4 meters wide.
> Both need some residual space on either side of the vehicle related to
> our obstacle detection system and evacuation space. Other systems
> might not need this additional space; depends on the technology.
>
> You could also make the vehicle the same width and (much) longer, but
> this will generate problems as the turning radius would increase
> significantly; creating problems at curves, but especially at the
> entry of off-line stations...
>

Robbert, these are the reasons why I think PRT trains would be
preferable to GRT/PRT hybrid systems. The infrastructure modifications
required for PRT trains would likely be minimal (with a nod to Jerry
R.'s concerns about guideway strength) compared to those required for
GRT operation. So, strengthen up the guideway a bit, add some extra
siding to a few key areas, and modify the control software to handle
trains, and you have a brick-wall-proof doubling or tripling (or more)
of capacity.

But infrastructural needs aside, I think the biggest factor is
probably the need for GRT systems to aggregate passengers, which
involves coordinating human beings (unpredictable) as opposed to
coordinating vehicles (predictable) as PRT trains would do.

Mike C.

[Mike C]

On Aug 5, 10:21 am, Bruce Attah <bruce.at...@googlemail.com> wrote:
> Mike,
>
> I have looked at the UK regulations (spelled out in "The Railways and
> Other Guided Transport Systems (Safety) Regulations 2006"), and the
> guidelines published by Her Majesty's Rail Inspectorate, in particular
> "Safe Movement of Trains", and have come to the conclusion that there
> is no rigid brick-wall stop rule that applies to all transit systems
> (including novel ones) and is enforced by the regulator in this
> country. (For trams and for conventional trains under "permissive
> operation", there is a principle in the guidelines that says trains
> must be "operated in such a way and at such a speed to enable drivers
> to stop within the distance they can be seen to be clear ahead" --
> "driver" can mean whoever is in charge of a system that controls the
> train remotely, and there's no assumption that the emergency brakes
> have failed, which some people have said is a required assumption.)
> For new systems, there is instead a risk assessment approach. Each
> aspect and component of a new system must be thoroughly risk assessed
> by independent experts to demonstrate that it is safe, and HMRI will
> study the evidence collected, and, if satisfied, certify the system. I
> do not know about other countries.

Bruce, I am only going by the empirical evidence I've seen. The people
who are building these systems today (and dealing directly with
regulators) seem to be focusing on no less than 3 seconds. This
includes Masdar, Heathrow, and probably some Vectus-targetted efforts
in Sweden. Now, admittedly, they may be working behind the scenes on
reducing that 3 seconds down to 1 second, but I've seen no evidence of
that, and if it's happening secretly, that does little good for
someone trying to sell PRT elsewhere, where they are asking hard
questions about capacity.

PRT trains are (I think) achievable today even in a brick-wall-
constrained setting, and can help boost capacity to levels which will
appeal to candidate cities considering PRT. Maybe PRT trains are the
"snappy answer" to capacity questions.

>
> Anyway, I pointed out that quite high capacity can be achieved
> *without* breaking the brick wall stop rule, if one merely installs 1g
> brakes. This hardly requires a technical breakthrough, so it is not a
> matter of waiting years for the technology to be developed. Depending
> on the design of vehicle and track, 2g or even 3g braking in
> emergencies is feasible using technology that is already available and
> tested. If the system offers such hard braking, the cabin will need to
> be designed to protect passengers from bruising themselves against its
> interior and each other in the event of those brakes being applied,
> but this again does not involve any novel or untested technology. It
> seems to me that regardless of whether one has PRT trains or not, it
> is a good thing to have good brakes, and if one has good brakes, one's
> system will probably have more than adequate capacity anyway, so
> moving the vehicles as trains may never be needed.

Right, no disagreement here in principle, but the need for restraints
would likely raise some regulator eyebrows, and it may create more
problems than it solves.

>
> On the matter of safety, if there are no trains, then it is very
> unlikely that an accident will affect more than one vehicle at a time,
> and extremely unlikely that it will affect more than two, whereas if
> there are trains, the chance that an accident will affect n vehicles,
> where n is the length of a train, is nearly as great as the chance
> that an accident will affect one vehicle. That's not good.

Of course it's not good. :-) But what's good for regulators doesn't
always coincide with what's good for passengers. A well-designed PRT
system would be much safer than a train running on streets, even if
the PRT runs at subsecond headways. But that's not what the current
regulations say, so we have to live with it.

Again, I'm going by the current operational evidence we have: both PRT
systems currently under construction are targetting 3 seconds, so the
acceptance of 1-second or sub-second headways are still an unanswered
question.

> The system
> of coupling and decoupling will also have to be very solidly proved,
> or it will be a source of danger to passengers. If the coupling is
> merely electronic, then "train" is nothing but a euphemism for "very
> small headway".

Right, but as someone else (Luca?) pointed out in an earlier thread,
vehicles operating very closely together will have low energy of
impact during a fault. So a fault in the leader will cause the
follower to almost immediately bump into it and they become one unit
during most of the fault deceleration. The only chance for a high
impact collision is if the leader itself is in a high impact collision
("brick wall") in which case all vehicles in the train are subject to
fatal impact. Yes, I know it would be preferable from an absolute
safety standpoint for the vehicles to be running at short headways
than in a train during a brick-wall stop, because fewer vehicles would
suffer fatal decelerations. But that's the way the regulations are
written, as far as I know, so a train is more acceptable from a
regulatory standpoint.

As for coupling, if it is required, perhaps a magnetic coupling would
be sufficient? But my guess is that today's control systems could
operate with no coupling at all. i agree that such functionality would
need to be demonstrated to regulators.

Mike C

[eph]

I'm not sure unrestrained 1g stops hold for infants and seniors.
Anyway, different discussion.

I think sub-BWS headways can lead to whiplash whereas platoons (PRT
trains) would not.

My numbers for GRT trains give me about 4,500 pphpd per line and about
1,000 pphpd for PRT trains, with about 80% of the guideway left for
PRT. So a 4-line ROW would be required to get LRT equivalent
numbers. I think there would be significant expense to a 4-line ROW
compared to a 1 or 2 line ROW that can be placed at grade (though
still grade separated). Also, a 4-line ROW means strobing from pylons
used to hold up the upper deck.

F.

[Bruce Attah]

Mike,

I don't share your pessimism about the prospects for shorter headways.
Even if regulators are responsible for the current large headways, I
don't think that means they're going to insist on those headways
staying large for long. I'm pretty sure that there was a statement
from ATS at the time ULTra got certified saying that the regulators
expressed an openness to seeing shorter headways after the system has
been proved for a while in commercial operation. Once systems have
been proved useful and safe in practice, large headways will disappear
like the man with the red flag disappeared from in fron of early cars.

On harnesses, etc., and hard braking: the seatbelts and airbags we
find in cars are designed to protect passengers in the event of
decelerations of tens of g, whereas we're only anticipating a need to
protect passengers from the deceleration due to emergency brakes of
two or three g, so the protection system might be less obtrusive.

On PRT trains: I'm skeptical that any regulatory body that was
reluctant to permit headways shorter than 3 seconds would be quick to
approve of PRT trains such as you propose.

On bumping of platooning vehicles, the low energy thereof: indeed such
slight bumping would pose no danger to passengers (though it might
diminish the quality of the ride if it occurs), but that's not the
issue. What's at issue is the outcome if a vehicle comes to a sudden
halt for some reason (such as collision with an obstacle). In that
event, all the following vehicles in the train will come to a halt at
the same rate, or nearly the same rate, putting all their passengers
in danger. Without trains, this hazard does not exist.

> Mike C- Hide quoted text -

[Bruce Attah]

GRT + PRT gives 5500 pphpd? If that were PRT alone, it would be
achieved with 1 second headways, and 1.528 passengers per vehicle.
Doable with 1 g braking.

As for four lanes of guideway going in one direction, there are
options: they could all go down the same ROW, or they could go down
two or more parallel ROWs as part of a network - an option not open to
LRT.

> > Mike C- Hide quoted text -

[eph]

5,000 pphpd for GRT train alone (80 m long, equivalent to a 3-car LRT
platform, 5 pass. per vehicle, 3 m vehicle)
80% still available for pure PRT which adds another 1,000 pphpd with
conservative 0.3g braking, 1.1 passenger per vehicle (assuming
commuters). Could double that number with two-car PRT platoons, still
a little shy of LRT numbers for 2 lane per ROW. Might be good in some
circumstances (< 15,000 pphpd). Requires some sort of dynamic
coupling.

I was interested in a "drop-in" replacement for LRT so same ROW use.

F.

[eph]

Sorry, that was 4,500 for GRT trains and so good for <13,000 pphpd for
the whole thing.

F.

On Aug 6, 2:52 pm, eph <rhapsodi...@yahoo.com> wrote:
> 5,000 pphpd for GRT train alone (80 m long, equivalent to a 3-car LRT
> platform, 5 pass. per vehicle, 3 m vehicle)
> 80% still available for pure PRT which adds another 1,000 pphpd with
> conservative 0.3g braking, 1.1 passenger per vehicle (assuming
> commuters).  Could double that number with two-car PRT platoons, still
> a little shy of LRT numbers for 2 lane per ROW.  Might be good in some

> circumstances (< 13,000 pphpd).  Requires some sort of dynamic

[eph]

I don't see how PRT trains could be denied when regular trains are
allowed.

In the event of a collision, not all the vehicles would decelerate at
the same rate if there is a crumple zone built into the cars or the
cars jumped the track as they do in railway crashes. The trailing
cars decelerate more slowly. http://www.swedetrack.com/flyweave.htm#2

I think shorter headways are a good argument for future capacity
increases with initial systems proving how safe they are with BWS
headways.

F.

[eph]

Disregard the Ottawa numbers, they are for afternoon peak hours, so
not representative of yearly modal split.

F.

On Aug 6, 3:22 pm, eph <rhapsodi...@yahoo.com> wrote:
> I don't see how PRT trains could be denied when regular trains are
> allowed.
>
> In the event of a collision, not all the vehicles would decelerate at
> the same rate if there is a crumple zone built into the cars or the
> cars jumped the track as they do in railway crashes.  The trailing

> cars decelerate more slowly.http://www.swedetrack.com/flyweave.htm#2

[Walter Brewer]

My numbers are about like San Diego. Peak hour transit share there is about
3.5%
Remember walking/biking produce many less pass-miles. I assumed 1/10 that
for transit.
Other assumption is equal rideshare buses and LRT. Again about true in San
Diego.
Mode ocupancies are from USDOT "Public Transportation's Role in Responding
to Climate Change". FTA January 2009.

Walt Brewer

----- Original Message -----
From: "eph" <rhaps...@yahoo.com>
To: "transport-innovators" <transport-...@googlegroups.com>
Sent: Thursday, August 06, 2009 3:36 PM
Subject: [t-i] Re: GRT vs PRT with Train Formation

[eph]

Thanks, somehow I posted this to the wrong thread (should have been
"VMT Reductions: Apparent Meager GHG Gains "). having one of those
days I guess...

F.

[Jerry Schneider]

At 12:22 PM 8/6/2009, you wrote:

>I don't see how PRT trains could be denied when regular trains are
>allowed.

Depends on the prevalence of logical thinking on the part of the "deciders".

>In the event of a collision, not all the vehicles would decelerate at
>the same rate if there is a crumple zone built into the cars or the
>cars jumped the track as they do in railway crashes. The trailing
>cars decelerate more slowly. http://www.swedetrack.com/flyweave.htm#2

How many collisions would it take to kill off the PRT concept? Remember that
during the initial start-up of Morgantown, which had a few "non-fatal
glitches", their
was some talk of abandoning and dynamiting the whole system.

>I think shorter headways are a good argument for future capacity
>increases with initial systems proving how safe they are with BWS
>headways.

How much operating experience does it take, with zero incidents, to
"prove" that the system is "safe"?
What is the definition of "safe" likely to be for a PRT system?
Theoretical and/or experiential?

[eph]

Good points.

I think zero collisions are better of course and any collision would
damage the early reputation. I also think that if the current rules
are broken and a collision happens, that will be the end. If the same
rules are used, we can point to numerous rail crashes and remind the
"deciders" that we followed all the same rules.

It's not so much "prove" as convince the public and politicians that
it is safe. Years of incident-free operation goes a long way to
convince the public that headways can safely be reduced to sub-BWS.

If PRT trains can be used as a gateway to PRT, why not go that route?

F.

On Aug 6, 5:59 pm, Jerry Schneider <j...@peak.org> wrote:
> At 12:22 PM 8/6/2009, you wrote:
>
> >I don't see how PRT trains could be denied when regular trains are
> >allowed.
>
> Depends on the prevalence of logical thinking on the part of the "deciders".
>
> >In the event of a collision, not all the vehicles would decelerate at
> >the same rate if there is a crumple zone built into the cars or the
> >cars jumped the track as they do in railway crashes.  The trailing

> >cars decelerate more slowly.http://www.swedetrack.com/flyweave.htm#2

[Jerry Schneider]

At 03:27 PM 8/6/2009, you wrote:

>Good points.
>
>I think zero collisions are better of course and any collision would
>damage the early reputation. I also think that if the current rules
>are broken and a collision happens, that will be the end. If the same
>rules are used, we can point to numerous rail crashes and remind the
>"deciders" that we followed all the same rules.

My impression is that most rail crashes are derailments - not a lot of
collisions - but I'm just guessing.

>It's not so much "prove" as convince the public and politicians that
>it is safe. Years of incident-free operation goes a long way to
>convince the public that headways can safely be reduced to sub-BWS.

Is there a difference between "prove" and "convince" ?

>If PRT trains can be used as a gateway to PRT, why not go that route?

I'd rather see initial PRT implementations in places where its
capacity limits are
not likely to be a problem. And, then I'd like to see headways
reduced. And then
more small applications. After a couple of years of error-free
operation, then some
efforts with trains and/or GRT/PRT service where the capacity needs
for pulse times
require assistance. Not the best path to becoming rich, but a path that might
eventually be productive and profitable and make you famous
contributor to real societal
needs. And, then some dualmode applications or mixed SM & DM
applications with some
tube transit here and there.

[eph]

For me, the race isn't to money, it's to low GHG emissions. It's just
not happening fast enough.

F.

[Mike C]

I agree with the phased approach -- probably the worst thing that can
happen is someone tries to do too much up front and failing miserably,
putting a taint on the entire technology (Aramis, Raytheon).

But having said that, even if initial systems are small, there needs
to be a path to higher capacity down the line or some places won't
even consider it. So far, the solutions to higher capacity have been
more guideway (raising concerns about aesthetics and cost) , GRT modes
(which has usability concerns) and shorter headways (so far rejected
by regulatory authorities). PRT trains could provide a regulation-
friendly path to higher capacity PRT, without the need for a lot of
extra guideway or the complexities of GRT coordination.

So now when some city says "we need a system that can scale up to 7000
pphpd", PRT trains can be presented as a future option. The future
higher capacity capability can help sell the lower capacity starter
system.

Mike C

[Dennis Manning]

Mike C wrote earlier today:

>So now when some city says "we need a system that can scale up to 7000
pphpd", PRT trains can be presented as a future option. The future
higher capacity capability can help sell the lower capacity starter
system.>

Only the most heavily used metro rail systems ever achieve 7000 pphpd
volumes. I'd challenge a city on why they need that much capacity and give
them a few examples of actual ridership on various systems. No LRT system in
nation has that much peak hour ridership. Another way to focus on what
capacity is actually needed is to ask what daily ridership estimates they
are using. The peak hour percentage of daily ridership doesn't vary too much
from one system to another (probably about 10% - maybe someone has a better
number). The request for maximum capacity would no doubt be way beyond what
is actually needed.

A good ridership estimate would make it obvious, but I'll bet most studies
don't report the estimated peak hour number or hide it as foot note.

I always like the Denver example. They asked for 14,000 pphpd capacity. They
run perhaps 3500 peak hour, and they now bragging at how they are exceeding
ridership estimates.

Dennis

----- Original Message -----
From: "Mike C" <mwil...@gmail.com>
To: "transport-innovators" <transport-...@googlegroups.com>
Sent: Thursday, August 06, 2009 7:41 PM
Subject: [t-i] Re: GRT vs PRT with Train Formation

[Kirston Henderson]

on 8/6/09 4:59 PM, Jerry Schneider at j...@peak.org wrote:

> How much operating experience does it take, with zero incidents, to
> "prove" that the system is "safe"?
> What is the definition of "safe" likely to be for a PRT system?
> Theoretical and/or experiential?
>

Jerry,

I am convinced that it will first require extensive failure and effects
analysis of every element of the control system, (probably using computer
operational simulations) followed by perhaps thousands of hours of
full-scale systems testing on actual guideway with actual vehicles without
passengers with deliberate induction of all possible failures to observe
that the system can heal itself or, at least, perform an orderly safety
shutdown. That is the reason that we say that we do not expect to offer any
fully automated vehicle operation for, at least, four years. We will then
only be able to do it by operating vehicles in the automated mode during off
hours wherein passengers are excluded from the system. That is one of the
key reasons that we are anxious to get a small, manually-controlled system
operating in revenue service carrying passengers during most of the 24-hour
day and making it available for extensive automated testing at other times.

I realize that the above approach is very unpopular with a lot of those
who participate in this discussion group, but we see no other practical
means to the end of achieving fully automated operation with assurance of
passenger safety and that is precisely where we intend to go.

I really don't think that the limited testing that one can do with small
test tracks is likely to to sufficient and we do not intend to attempt that
route. Perhaps, it is because we have extensive experience with development
of fault, fly-by-wire and fail-safe weapons control systems for nuclear
weapons. One can dream of accomplishing such tasks quickly, but I don't
know any way to really short-cut the process without endangering people.

Kirston Henderson
MegaRail®

[Kirston Henderson]

on 8/6/09 5:27 PM, eph at rhaps...@yahoo.com wrote:

> If PRT trains can be used as a gateway to PRT, why not go that route?

I believe that you are precisely on course with the above statement.

Kirston Henderson
MegaRail®

[gu...@systematica.net]

Dennis
As I noted before, the most heavily used metro trains achieve nearly 10 times the pphpd figure that you're quoting. And if you don't like Moscow there are several metro lines in the world that far exceed 10'000 pphpd any Monday to Friday. This includes new York for those who are short sighted.
Peak hour traffic is a lot more than 10% of daily traffic. If it were 10% then, considering that night time traffic is usually irrelevant, you'd get an almost flat distribution in the remaining 16 hours.
In the very common case of an unbalanced situation like that of a CBD with twice as many workplaces as residents it is not unusual for one direction peak hour traffic to be higher than 40% of daily two direction traffic. 30% is a very common and comfortable figure.
Cheers
Luca
Le mail ti raggiungono ovunque con BlackBerry® from Vodafone!

-----Original Message-----
From: "Dennis Manning" <john.m...@comcast.net>

Date: Thu, 6 Aug 2009 21:56:19
To: <transport-...@googlegroups.com>

[Michael Weidler]

I am wondering how it would be possible to make a control system that is worse than what we currently use? Just taking the unreliable human operator out of the picture drastically increases safety. Binding the vehicle to a guideway adds even more safety. Making the guideway exclusive adds more safety. Just how "safe" does the the system need to be??? --- On Thu, 8/6/09, Kirston Henderson <kirston....@megarail.com> wrote:

From: Kirston Henderson <kirston....@megarail.com> Subject: [t-i] Re: GRT vs PRT with Train Formation

[Jack Slade]

Mike, I don't think this problem will ever come up, unless somebody builds PRT with routes spaced too far apart.  Existing systems only get such usage because they collect people from a couple of miles each side of their route.   The problem for PRT is more  likely to be "show me places where I can be sure of averaging 3000 passengers per hour".  That is what I have calculated to be the figure necessary to provide the expansion profits that I need.   Jack Slade

--- On Fri, 8/7/09, Mike C <mwil...@gmail.com> wrote:

From: Mike C <mwil...@gmail.com> Subject: [t-i] Re: GRT vs PRT with Train Formation

[Walter Brewer]

Not disputing the numbers, and even remembering Manhattan, (different from
NYC). But the basis for comparison is single vehicle doorstep to doorstep vs
multivehicle mass transit.
Why do those bikes/scooters/autos/etc show up in mass transit saturated
cities?

Walt Brewer

[eph]

Ottawa claims over 10,000 pphpd, quoted several times. The article
below tries to deny this, but people in Ottawa cycle and use other
methods to get to the city core in the summer (yes, people don't like
mass transit enough to pay even 50% of O&M cost - less students in the
summer too). The link below includes references to quotes of 10,000
pphpd.
http://www.publictransit.us/ptlibrary/specialreports/sr8.1.OttawaTransitPart2.htm

Projected volumes for 2031, the design date, pphpd is even higher.
They plan on building platforms for 6 car LRT which doubling 3-car LRT
numbers allows almost 30,000 pphpd!

7,500 pphpd won't even register for Ottawa transit planners.

F.


On Aug 7, 12:56 am, "Dennis Manning" <john.manni...@comcast.net>
wrote:

[Dennis Manning]

Luca:

Perhaps the problem is with the daily ridership number. Any system may show
say 50,000 riders per day but it doesn't mean 50,000 pass the maximum use
point. Maybe the most used section gets 30,000. Also daily ridership numbers
are usually for both directions, not per direction. At any rate you are
speaking mostly of heavy metro systems in Europe. I was speaking mostly of
LRT in the US. I'll stand by my assertion no LRT system in the US has more
than 7000 pphpd. I could be low with the 10% figure. Anybody have better
figure for US LRT systems? Overall a big problem is that the pphpd figues
are very hard to come by. If one googles to find it good luck. In contrast
there are 100s of references to maximum theoretical capacity which is a
relatively useless number by comparison. Ridership numbers aren't too hard
to find, but again they don't express the maximum pphpd.

[Dennis Manning]

Speaking of Manhattan the max pphpd that I recall is 50,000 in a tunnel
where two major lines come together. Otherwise the max is 30,000 (the 30k
number is from an Ed Anderson paper so may be conservative).

[Walter Brewer]

And people trips usually mean boardings, not pass-miles, etc.
Doorstep to doorstep, mass transit typpically needs about 3 boardings, vs 1
for PRT.

Walt Brewer

----- Original Message -----
From: "Dennis Manning" <john.m...@comcast.net>
To: <transport-...@googlegroups.com>

[Dennis Manning]

Yes, that's another problem. When pphpd figures are available the
methodology is obscure. The producer of the numbers is not impartial.

The FTA ought to order any federally funded system or those applying for for
Federal funds to produce accurate peak pphpd numbers.

[eph]

A study done for Ottawa gave 2.7 for peak hours, 2 times and 85% for
the rest of the time.
so 627 bus run at peak hours
*2.7 = 1693 bus hours
*2 peaks = 3,386
85% for the rest of the time = 2,879
Total operating hours = 6,265

So peak hours equal 5.4 hours
The rest of the time (18? - 5.4 = 12.6 hours?)
12.6 hours = 5.4 * .85
So off-peak hour bus needs averages about 36% of peak hour bus needs
in this consultant's model.

F.

On Aug 7, 10:35 am, "Dennis Manning" <john.manni...@comcast.net>
wrote:

[Mike C]

On Aug 7, 9:52 am, eph <rhapsodi...@yahoo.com> wrote:
> Ottawa claims over 10,000 pphpd, quoted several times.  The article
> below tries to deny this, but people in Ottawa cycle and use other
> methods to get to the city core in the summer (yes, people don't like
> mass transit enough to pay even 50% of O&M cost - less students in the
> summer too).  The link below includes references to quotes of 10,000

> pphpd.http://www.publictransit.us/ptlibrary/specialreports/sr8.1.OttawaTran...

>
> Projected volumes for 2031, the design date, pphpd is even higher.
> They plan on building platforms for 6 car LRT which doubling 3-car LRT
> numbers allows almost 30,000 pphpd!
>
> 7,500 pphpd won't even register for Ottawa transit planners.

See, that's where the blurred line between light rail and metro rail
is beneficial to rail proponents. There is a HUGE difference between
metros and light rail, especially street-level light rail, but people
use them interchangeably.

My answer would be: how much did you pay to get to 10,000, and how
much MORE will you pay to get to 30,000? My guess is that it will
likely exceed $200M per mile to get to 30k (at least in North
America), and for that money you could build 4 PRT lines at 7500 each,
using PRT trains and/or ridesharing and/or reduced headways.

In any case, the fact that gets lost in all the rail muddiness is that
PRT is *competitive* in construction cost and *superior* in both
service and operational cost. Whether it's 5000 pphpd light rail at
$60M/mile or 30,000 pphpd metro rail at $200M/mile, PRT at 5000 pphpd
and $30M/mile is competitive, even on a strictly line-based metric
(not even accounting for PRT's superior network benefits). But as long
as people start quoting light rail costs coupled with metro rail
capacity, PRT will always look too expensive.

>
> F.

Mike C

[Luca Guala]

Walt,

> Not disputing the numbers, and even remembering Manhattan, (different from

NYC). But the basis for comparison is single vehicle doorstep to doorstep vs

multivehicle mass transit.

When I dispute the numbers on mass transit that now and then show up on this
list, it is not because I think mass transit is superior to PRT (I wouldn't
be contributing to this list) but only because I don't want to fool
ourselves with nice numbers. Sometimes I read messages that say "LRT can't
do this" and "metros don't do that" while I have knowledge and numbers to
show that they do.
If you want to promote a new transport system (or a new recipe for pizza,
for that matter) remember that you are doing it against a real world, not
against a rosy world that someone has fabricated to please you.
Not recognizing the potential of competition is the sure way to failure.

> Why do those bikes/scooters/autos/etc show up in mass transit saturated
cities?

If at peak hour mass transit carries tens of thousands of passengers per
hour, it is not because for some strange collective folly, between 7.30 and
8.30 everybody chooses to travel by mass transit and leave the roads free,
but only because there are hundreds of thousands of people travelling to
certain destinations, typically, in a city, it is work and school. Mass
transit always carries only a fraction of the total of travellers
Regards
Luca

Walt Brewer
----- Original Message -----
From: <gu...@systematica.net>
To: <transport-...@googlegroups.com>
Sent: Friday, August 07, 2009 1:25 AM
Subject: R: [t-i] Re: GRT vs PRT with Train Formation


> Dennis
> As I noted before, the most heavily used metro trains achieve nearly 10
> times the pphpd figure that you're quoting. And if you don't like Moscow
> there are several metro lines in the world that far exceed 10'000 pphpd
> any Monday to Friday. This includes new York for those who are short
> sighted.
> Peak hour traffic is a lot more than 10% of daily traffic. If it were 10%
> then, considering that night time traffic is usually irrelevant, you'd get

> an almost flat distribution in the remaining 16 hours.
> In the very common case of an unbalanced situation like that of a CBD with

> twice as many workplaces as residents it is not unusual for one direction
> peak hour traffic to be higher than 40% of daily two direction traffic.
> 30% is a very common and comfortable figure.
> Cheers
> Luca

> Le mail ti raggiungono ovunque con BlackBerryR from Vodafone!

[Luca Guala]

Dennis,

you wrote:
"Only the most heavily used metro rail systems ever achieve 7000 pphpd
volumes"

If by this you meant:

"no LRT system in the US has more than 7000 pphpd"

Then I can't but agree. When similar misunderstandings take place, I usually
blame myself and the fact that I am not a native English speaker.
In this specific case though, allow me to doubt that the cause of the
misunderstanding accrues entirely to my limited knowledge of English as a
foreign language.
Regards
Luca


-----Messaggio originale-----
Da: transport-...@googlegroups.com
[mailto:transport-...@googlegroups.com] Per conto di Dennis Manning
Inviato: venerdì 7 agosto 2009 16.35
A: transport-...@googlegroups.com
Oggetto: [t-i] Re: GRT vs PRT with Train Formation

[Jerry Schneider]

At 07:35 AM 8/7/2009, Dennis M. wrote:

>Luca:
>
>Perhaps the problem is with the daily ridership number. Any system may show
>say 50,000 riders per day but it doesn't mean 50,000 pass the maximum use
>point. Maybe the most used section gets 30,000. Also daily ridership numbers
>are usually for both directions, not per direction. At any rate you are
>speaking mostly of heavy metro systems in Europe. I was speaking mostly of
>LRT in the US. I'll stand by my assertion no LRT system in the US has more
>than 7000 pphpd. I could be low with the 10% figure. Anybody have better
>figure for US LRT systems? Overall a big problem is that the pphpd figues
>are very hard to come by. If one googles to find it good luck. In contrast
>there are 100s of references to maximum theoretical capacity which is a
>relatively useless number by comparison. Ridership numbers aren't too hard
>to find, but again they don't express the maximum pphpd.

Some systems are supposed to be using automatic
boarding/deboarding counters on their LRT trains.
If they are accurate and reliable, then the
transit system should have pphpd data, by time of day, for such
trains. Getting it from them might be difficult
to impossible and I've heard that the accuracy
and reliability is not so great - but that is
just hearsay. I do know that some people in Portland tried to
get visual, manual, passenger counts at several
LRT stations but were apprehended and forced to
stop doing it by TriMet security people. Must be
a sensitive issue! Why would FTA require such embarrassing
data to be collected and made public?

[Dennis Manning]

Luca:

 

With such a large discrepancy between your 30% and my 10% for pphpd as percentage of daily ridership I've found a very good web site. It points to a US average of 13%. I'll email more about this web site. It's quite extensive in analyzing capacities (actual and theoretical), ridership, and ridership forecasts. I'm still trying digest a lot of it. 

 

I'll eat crow. It shows one US LRT system with pphpd of 10,000, but it's the only one above 7,000. Average closer to 2000.  More later...........

 

Excerpts:

"Peak traffic share" (PTS) is the ratio of one-way traffic during the busiest hour (in the busier

direction, at the maximum-load point) to two-way, all-day ridership.

 

The "benchmark" PTS value of 13 percent for U.S. cities is supported by a wide variety of data6.

Individual corridors may have a lower PTS, reflecting relatively high levels of off-peak and

reverse-peak traffic -- or capacity constraints leading to "peak spreading." PTS values below

Ridership Forecast Analysis - Los Angeles Red Line May 1, 2005

 

6 Demery, Leroy W., Jr. 1994. “Supply-Side Analysis and Verification of Ridership Forecasts for Mass Transit Capital

Projects.” Journal of the American Planning Association 60, 3: 355-371.

Dennis

 

[Dennis Manning]

http://www.publictransit.us/ptlibrary/specialreports/sr12.LosAngelesSubway.pdf

 

To all who have been following this thread. I highly recommend studying this report. It has loads of data on capacities, pphpd, and ridership on US transit systems, goes a long to explain why initial estimates for LA Red Line missed by a factor of about 2.5, and much that is highly relevant to all that we have been discussing. Walt touched on aspect why of why transit ridership estimates can be so far off. It has to do with not accounting for length of trips vs boardings and  actual service that is projected to be provided.

 

Enjoy.

 

Dennis

[Dennis Manning]

Is this system considered metro rail or LRT?

[eph]

The city core section will have a 3 km tunnel and will effectively be
metro/subway. The rest of the system should be LRT. The price tag is
climbing to $5-$7B for the whole system and there will be no new lines
and the suburbs won't get rail until after 2031. It's completely
ridiculous but there is no alternative presented save a bus tunnel
which would work but isn't sexy (and not GHG friendly) and the price
tag comes in at about $2B less. The price tags came out about even
when the consultant presented the options.

We are preaching to the choir.

F.

[Jerry Roane]

Luca

I rode the Metro a few times in Paris and I never rode one that was even close to full.  I was working for a French parent company at the time and I was going off peak I am sure on the subway because the work hours I was doing a reverse commute from inside Paris out to the suburbs where the credit card factory was.  One two week trip I took my wife because the airlines were offering a free companion ticket just after they blew up the tourist shops back in the 1980s.  The airlines were giving away companion tickets trying to get people to go visit France again after that spat of terrorist activity.  The Metro was not allowed to be filmed as my wife found out by accident as the police put their hand onto the camera lens.  The subways were linked like we are discussing here but they were linked poor man's train coach and upper class coach.  Apparently you can pay more money to ride the upper crust train sort of like congestion pricing I guess in subway style.  I see no way that the Paris subway trains I rode could carry the number of people per hour discussed here.  They did not carry any more than a bus could carry but they were grade separated which gave them the advantage they had.  The disadvantages were the subways only served a subset of the city so if you were trying to go somewhere specific there was still a taxi or rental car to find.  If you were just knocking around town looking at subway station sign boards it was great.  You would think that Paris would have a high transit flow but I did not see it in my travels there.  Maybe Paris is not dense enough for mass transit with trains to be profitable. ;-)

Jerry Roane

[eph]

Is the "phd" chart measured as an average of peaks for the year? By
this measure Ottawa would likely average out lower than 10,000 phd,
but there are times (winter) when this capacity is needed.

F.

On Aug 7, 1:35 pm, "Dennis Manning" <john.manni...@comcast.net> wrote:
> http://www.publictransit.us/ptlibrary/specialreports/sr12.LosAngelesS...

[eph]

According to wikipedia, it would qualify as light-metro. LRT would
have grade crossings (with barriers) ALRT (automated LRT) must have a
segregated ROW, so capacity is higher. It seems like anything that
uses an LRV is considered LRT by many, but then price per km and
capacity vary wildly.

http://en.wikipedia.org/wiki/Light_rail#Comparison_to_other_rail_transit_modes

F.

On Aug 7, 2:11 pm, "Dennis Manning" <john.manni...@comcast.net> wrote:
> Is this system considered metro rail or LRT?
>
> Dennis
>
>
>
>
>
> ----- Original Message -----
> From: "eph" <rhapsodi...@yahoo.com>
> To: "transport-innovators" <transport-...@googlegroups.com>
> Sent: Friday, August 07, 2009 6:52 AM
> Subject: [t-i] Re: GRT vs PRT with Train Formation
>
> Ottawa claims over 10,000 pphpd, quoted several times.  The article
> below tries to deny this, but people in Ottawa cycle and use other
> methods to get to the city core in the summer (yes, people don't like
> mass transit enough to pay even 50% of O&M cost - less students in the
> summer too).  The link below includes references to quotes of 10,000

> pphpd.http://www.publictransit.us/ptlibrary/specialreports/sr8.1.OttawaTran...

[Dennis Manning]

The FTA won't make such a requirement of course, but I think the data would
be valuable.

[Dennis Manning]

Luca:

Not a problem. Your English is excellent, and it's easy for anyone to
misinterpret the swiftly written emails of this list as we see it happen
rather frequently.

Dennis

----- Original Message -----

From: "Luca Guala" <gu...@systematica.net>
To: <transport-...@googlegroups.com>

[Dennis Manning]

I don't know.

----- Original Message -----
From: "eph" <rhaps...@yahoo.com>
To: "transport-innovators" <transport-...@googlegroups.com>
Sent: Friday, August 07, 2009 11:25 AM
Subject: [t-i] Re: GRT vs PRT with Train Formation

[Jerry Roane]

Dennis

Getting this data would be a mixed bag.  It would indicate how many people went out of their way to go down a corridor, time synchronized with other citizens going that part of the trip that day.  To get useful data they would need to collect the door-to-door-to-door trips made that day per person.  So you ride a bus/train/bus to commute but then you drive in your black smoke belching diesel pickup truck to Costco to save on fresh fruit flown in from Hawaii the night before by transcontinental jet.  I suggest that knowing data that the FTA knows but does not have on their opening page of their web site will not be useful if it only enumerates the corridor/circulator portion.  Unless we escalate the discussion past simplistic terms we will not get to a solid comprehensive solution.  This number would do almost nothing because it is far too incomplete to have value. 

It would be nice if the FTA was more up front though.  The whole idea of "ride" blows me away.  It is important how things are measured and it is often that the measurement technique gives results you may not want.  This concept is taught in corporate self-help seminars and stripping away the hype it is true often.  If you want a behaviour, tell your employees on what measure their salary rests.  All other measures may suffer so be careful what you ask the employees for.  Example Ford--- "Quality is job one"  Measured quality may go up while viability goes down etc.  An employee needs to be productive and by the way the parts need to have quality as viewed by the ultimate payer.  You know --- balance.  Transportation data would be useful even if incomplete as long as it is truthful out in the real world.  

Jerry Roane      

[Jerry Roane]

F.

I tried to count heads and passengers in each of the photographs in the wikipedia article.  Only the Irish one had customers showing of any consequence.  I saw no way that those systems could move 10,000 passengers per hour per D.  One station had 8 riders/bystanders hanging around.  To move that many people these train-like things would need to be immersed in human flesh with legs, arms and feet hanging out of all cracks. 

If the photo of your system shows it being useless then how are we to accept the wild unsupported claims of miles of useful travel from these devices?

Every time I see a bus here I always count the heads.  There is no way this will meet the need that cars fulfill. 

Jerry Roane

[Dennis Manning]

Jerry:

 

This thread began when Mike C said PRT is at a disadvantage when transit agencies demand a capacity of 7000 pphpd. I suggested he make them justify why or if that capacity was needed. Where I see the value is in making it clearer just what the actual usage and needed capacity is likely to be, and some value in making ridership estimates more realistic.

[eph]

Here's a video that shows the flow of buses along one of the busiest
sections. Note the frequency of buses and also that this stop (Rideau
Centre) would be one of the first CDB stop (after the University)
filling up along the CBD to head west.
http://www.youtube.com/watch?v=cLOGuPoftAI

I might find better examples if I look. I don't think 10,000 phd is
exaggerated, especially during winter months.

F.

On Aug 7, 4:04 pm, Jerry Roane <jerry.ro...@gmail.com> wrote:
> F.
>
> I tried to count heads and passengers in each of the photographs in the
> wikipedia article.  Only the Irish one had customers showing of any
> consequence.  I saw no way that those systems could move 10,000 passengers
> per hour per D.  One station had 8 riders/bystanders hanging around.  To
> move that many people these train-like things would need to be immersed in
> human flesh with legs, arms and feet hanging out of all cracks.
>
> If the photo of your system shows it being useless then how are we to accept
> the wild unsupported claims of miles of useful travel from these devices?
>
> Every time I see a bus here I always count the heads.  There is no way this
> will meet the need that cars fulfill.
>
> Jerry Roane
>

> On Fri, Aug 7, 2009 at 1:33 PM, eph <rhapsodi...@yahoo.com> wrote:
>
> > According to wikipedia, it would qualify as light-metro.  LRT would
> > have grade crossings (with barriers) ALRT (automated LRT) must have a
> > segregated ROW, so capacity is higher.  It seems like anything that
> > uses an LRV is considered LRT by many, but then price per km and
> > capacity vary wildly.
>

> >http://en.wikipedia.org/wiki/Light_rail#Comparison_to_other_rail_tran...

[Gérard Massip]

Fait partie de GRT vs PRT with Train Formation

http://groups.google.com/group/transport-innovators/browse_thread/thread/bb628cfd8863db07/58f3e6bdaa6b95b7?hl=fr&lnk=gst&q=pphpd#58f3e6bdaa6b95b7

[Jerry Roane]

Dennis

Agree.  Getting them to produce real data would be step one.  Their measure of rides pushes them to maximize rides so the measure they use internally is flawed but it is probably the best they can tell us. 

I was amused by the flow of mostly empty bus seats in the last video.  The tall skinny girl after 7 minutes still had not found a bus she needed and the camera moved before she boarded so we don't know if her bus ever came.  A single boarding up the steps of the bus is about the rate we load 4 passengers at a time onto guideway.  If you notice a group boarding the first three jump up the steps but the fourth gets a people traffic jam and has to wait for the first three in the video.  Everyone in all those buses has to stop multiple times at this "station" curbside.  Getting real data from the video is not possible because the camera can't look into the tinted glass and the foreground people block the view of boarding people in the background.  To design a good efficient system it is necessary to get solid true data.  Someone somewhere has that data surely!

Jerry Roane

[Jerry Schneider]

At 01:13 PM 8/7/2009, you wrote:
>Jerry:
>
>This thread began when Mike C said PRT is at a disadvantage when
>transit agencies demand a capacity of 7000 pphpd. I suggested he
>make them justify why or if that capacity was needed. Where I see
>the value is in making it clearer just what the actual usage and
>needed capacity is likely to be, and some value in making ridership
>estimates more realistic.

I suspect that the justification would be derived from "forecasts"
for a proposed system. Actual data would have to come from
operating systems and the transit agency involved. Given the American
interest in and belief in "growth", the forecasts are likely to be
pretty large, compared to the "actual".

[Dennis Manning]

I find hiding the actual use numbers sort of odd given that the highway system spends so much counting traffic all the time all day long.

[Jerry Roane]

Dennis

That is because roads are insufficient and empty transit seats are abundant.  Why count empty transit seats?  There is no money or power in it.  In the name of the poor and neglected they keep all those bus and train seats moving.  What is sad is the poor and neglected will thrive with dual mode, energy efficient transportation we are selling.  Why subsist when you could thrive?  The difference is first believing that something better can be. 

Jerry Roane

[Jack Slade]

I had a TTC Engineer tell me that the Queen St streetcar line does 10,000 PPH. However, he did not know if that figure was for both directions.   Also, the line is 10 miles long, with transfer ponits to and from many other lines, so there is no way of knowing how many if these trips are very short....get on the streetcar, go 10 blocks, transfer to subway or bus line, etc. The trip counts twice ....one streetcar trip, and one subway trip. One seat can be filled by several different people on a line this long, but almost none of the trips are full line length.   Jack Slade   This makes it even easier for PRT to provide better service.  --- On Fri, 8/7/09, eph <rhaps...@yahoo.com> wrote:

From: eph <rhaps...@yahoo.com> Subject: [t-i] Re: GRT vs PRT with Train Formation

To: "transport-innovators" <transport-...@googlegroups.com> Received: Friday, August 7, 2009, 1:52 PM

Ottawa claims over 10,000 pphpd, quoted several times.  The article below tries to deny this, but people in Ottawa cycle and use other methods to get to the city core in the summer (yes, people don't like mass transit enough to pay even 50% of O&M cost - less students in the summer too).  The link below includes references to quotes of 10,000 pphpd.

http://www.publictransit.us/ptlibrary/specialreports/sr8.1.OttawaTransitPart2.htm

Projected volumes for 2031, the design date, pphpd is even higher. They plan on building platforms for 6 car LRT which doubling 3-car LRT numbers allows almost 30,000 pphpd! 7,500 pphpd won't even register for Ottawa transit planners. F. On Aug 7, 12:56 am, "Dennis Manning" <john.manni...@comcast.net> wrote:

[Luca Guala]

Dennis

Partly my mistake: the 30% (or 40%) data I hinted to is the likely share of commuter travellers that use a transit system at peak, not that of ALL travellers. The share of daily traffic that takes place at peak is obviously lower and depends on what share of total ridership is represented by commuters. Example: a business district where most of the trips are made by workers who live elsewhere may have a peak hour share of 25-30% of daily traffic. In a mixed use city centre with shops, houses and businesses peak hour share will be a lot lower.

This kind of analysis is commonplace when planning a transit system and should be applied to PRT too

Regards

Luca

 

 

---

Da: transport-...@googlegroups.com [mailto:transport-...@googlegroups.com] Per conto di Dennis Manning
Inviato: venerdì 7 agosto 2009 19.16
A: transport-...@googlegroups.com
Oggetto: [t-i] Re: GRT vs PRT with Train Formation