Light Rail Now WRONG about PRT

Other rebuttals

By J. Edward Anderson
By Steve Anderson
By Bill Wilde

Have you read the Light Rail Now PRT report?

Here's what they got wrong

One of the great challenges in educating the general public about Personal Rapid Transit (PRT) is monitoring the occasional criticisms. As is the case when any outsider attempts to influence the mainstream, a great deal of SoundPRT's time is devoted to responding to critiques of PRT that are just plain wrong.

SoundPRT understands that we have been living for so long with the problematic aspects of conventional transit technology (schedules, waiting, transfers, huge vehicles, etc.) that it is easy to assume that all transit must have such problems. Fortunately, when PRT is accurately described to them, most people immediately "get it."

But PRT is so unconventional that occasionally even smart people find it hard to comprehend. Such is the case with a new PRT critique issued by the group Light Rail Now (LRN). As in past examples of inaccurate analyses, this one (titled Personal Rapid Transit—Cyberspace Dream Keeps Colliding With Reality) appears to be a logical and reasonable examination of PRT.

However, an informed examination of LRN's report reveals it to be fundamentally flawed.

LRN understands some of PRT's basic characteristics, but their report demonstrates that they do not yet understand how PRT would work in actual practice. This misunderstanding is apparently influenced, at critical junctures, by the assumption that because a train operates a certain way, PRT must operate that way too. Another source of error is that "drawbacks of PRT" are raised, and left unanswered as though the supposed drawbacks are unsolvable—when in fact solutions have been found years ago. It is therefore inevitable that LRN reaches inaccurate conclusions.

       What did they get wrong? Read on.

LRN Claim REALITY
PRT is elitist
(Report Section: Theoretical Perfection)
Quote: "you don't have to ride with 'those people' "
PRT is the first transit system that treats everyone equally: no matter who you are or where you live, a city-wide PRT system provides non-stop transit, 24-7. How can it do this? Because there are 4-5 small stations per square mile, everyone is within walking distance of a station, and service is on-demand. Neighborhoods no longer have to compete for routes or service allocations, and service no longer ceases late at night.
LRN Claim REALITY
"barrage of visual obstruction"
(Report Section: "Visual barrage")

Quote: "The effect would be... a dense forest of concrete and steel support piers and aerial structures"

Quote: "PRT... guideways between a quarter and a half-mile apart, with stations about every 1/4 to 1/6 mile."
LRN presents a black & white illustration of Los Angeles City Hall (a copyrighted image from Fundamentals Of Personal Rapid Transit, D.C. Heath Co.) showing a combined merge and crossover intersection. This is not typical, but characteristic of the 1970s-era system envisioned by that designer. Most intersections in contemporary PRT designs would not have one guideway crossing over another.
The optimum spacing for PRT stations and guideway grid is actually 1/2 mile, or the length of nearly 9 football fields. The logic for this distance is that you never have to walk more than 1/4 mile to reach a PRT station. But LRN gets the math wrong—station spacing is not equal to walking distance. If you are standing halfway between two PRT stations that are 1/4-mile in either direction, how far apart are the stations? LRN apparently thinks 1/4 mile; or maybe 1/6 mile. The correct answer is 1/2 mile.
LRN's claim of "dense forest" of guideway structure is illusory. If one were to stand in the center of a PRT loop, stations and guideway would be 1/4 mile to the north, south, east and west.
Here is a photo of small elevated guideway about 1/8 mile distance from the camera—can you see it? Do you think it is overly intrusive (or was your attention first drawn to the picnic tables)? What if it were twice as far away, the preferred 1/4 mile? There is certainly no "forest" in this photo—although there are some pretty trees.
The only time PRT stations and guideway might be spaced more closely is if it is called for by existing urban densities—such as in a downtown, where little PRT guideways would be dominated by skyscrapers.

Quote: "...any outdoor, heavily used system is particularly vulnerable to mishaps [and] weather conditions... How would pasengers extricate themselves from their little cars and guideways?"
PRT has been designed to adhere to local codes for purposes of safety in collisions, wind and earthquakes.
One of the reasons planners from cold climates such as Sweden, Norway and Minnesota are interested in PRT is that it can be designed to be unaffected by snow and ice: running surfaces can be enclosed, and induction (magnetic) motors used for propulsion and braking (just like the Vancouver Skytrain) do not lose traction. What happens to light rail in heavy snow and ice? Well, in Portland the Max simply has to shut down.

Quote: "...evacuation walkways, wide enough for wheelchairs... would undoubtedly enlarge the profile of PRT guideway"
PRT guideway would not need walkways in most cases. If a vehicle were to unexpectedly slow down or stop in mid-journey, it is more dangerous for passengers to attempt to get out and negotiate a walkway (it might be night time, or slick with rain or snow). Instead, the next PRT vehicle would push the problem vehicle to the next station—the passengers safely inside, in intercom contact with the central control center. Unlike single-line systems like light rail or monorail, vehicles in a PRT system could simply be temporarily re-routed to other nearby guideways, keeping the system as a whole running and safe.
LRN Claim REALITY
"System droppings"
(Report Section: "Elevated Realities")

Quote: "Another problem facing [PRT] systems is..."system droppings"—...debris, lubricants, water, ice, or snow falling on pedestrians or vehicles below the elevated guideway... especially when the guideway is just a slender beam straddled by the PRT vehicles."
PRT is electric, with no tanks of fuel or oil, and no leaky transmissions. If the PRT design uses wheels instead of maglev, the hubs would have lubrication—inside sealed bearings. So 'droppings' would not come from PRT, they would come from rain, snow, birds, etc. Shall we remove everything in our cities from which such substances now drip? The Sierra Club would not be pleased. It would also be difficult for parts to fall from a PRT vehicle, especially Skyweb Express, in which the section with moving parts rides inside the guideway.
For the record, there are no current PRT designs in which the vehicle straddles the guideway like a monorail train. A straddle configuration would require slow monorail-style switches built into the guideway. PRT guideway has no moving parts, each vehicle steers itself, under supervision of the central control system.
LRN Claim REALITY
"[PRT] stations are yet another source of visual intrusion"
(Report Section: "Elevated Realities")
LRN shows a composite photo of an obsolete Raytheon "PRT2000" guideway inserted into a typical Minneapolis street scene. PRT activists have long decried the Raytheon guideway and supports as unnecessarily big and expensive. If one were to stand next to PRT guideway, it would actually look like this, or this. The shadows are no greater than a pedestrian footbridge.

Quote: "...stations are... visual intrusions... require substantial additional access guideway"
PRT sidings are the same size and construction as the main guideway. The amount of 'off-ramp/on-ramp' guideway for vehicles entering and leaving the station is not onerous, and the land underneath them is obstructed only by 1-2 ft diameter supports. Is this intrusive compared to an elevated light rail station?
Quote: "an actual PRT station would... form a "lid" over street"
There are no contemporary PRT designs that envision building stations over streets.
       Walnut Station is part of a people mover system in Morgantown, WV that, as LRN explains near the start of the report, was one that was intended to be PRT but instead grew into large-vehicle "Automated Guideway Transit". That is, Morgantown's system is called PRT, but it is really Group Rapid Transit. So using it to make claims about PRT is apples-to-oranges.
       Interesting note: Even Walnut Station is smaller than a light rail station, which must be as long as the longest train it plans to serve. SoundPRT believes it would be wrong to "lid" a street, just as it would be wrong to plop a train station in the middle of a major Seattle arterial.


Quote from caption illustration: "A conceptual map of a PRT system... [the] concept requires dense penetration of neighborhoods"
(Report Section: Theoretical Perfection)
This occurs near the beginning of the report; there are two inaccuracies. First, Light Rail Now has mistaken as PRT a conceptual diagram for a "Dual Mode" system proposed by a Danish company. Dual Mode, while it uses elevated guideways, is not PRT (otherwise it wouldn't be called Dual Mode). Second, the typical distance between the "dense penetration" guideways in the diagram is FIVE miles.
LRN Claim REALITY
ADA compliance requires big size
(Report Section: "Elevated Realities")
Quote: "simple logistics and requirements... dictate minimum size characteristics which are substantial."
PRT systems are not uniform on this, because they are being created by different companies with differing visions. Designs by some companies are less ADA compatible than others. This does not mean PRT cannot be ADA compliant. Skyweb, Urban Light Transport (ULTra) and MicroRail, for example, are compliant. SkyTran is not currently compliant, though the designers plan to create a wheelchair-capable version of their vehicle.
The small size of PRT means placement of guideway and stations is highly flexible; just because an illustration shows PRT in a certain location doesn't mean that has to be the actual site. The SkyTran example, in which a station is inserted into a scene of Seattle's Westlake Center, is just that—an example.
A PRT station seems to need large amounts of space only if planned using light rail assumptions. A 3-berth Skyweb PRT station would be 30 feet long, an ULTra station would be about the same. Stairs would be used, and an elevator would be provided to comply with access regulations. While the most expensive station item in a Skyweb station is the elevator, the type chosen would cost $70,000, hardly a budget buster. Escalators, while nice, would be too expensive for a PRT station and, as LRN gets right, would make the station site too big. New buildings with escalators could have PRT stations integrated into their 2nd or 3rd floors.
LRN Claim REALITY
PRT is not new
(Report Section: "Concept from the 1960s")
"Personal Rapid Transit is... far from new, and... 'can be traced back to at least 1953'... technical problems... seemed to be inherent in the concept."
And trains can be traced back to the late 18th century. Da Vinci drew helicopters and submarines, and the Greeks imagined airplanes. What point is LRN trying to make? If you try something and fail, you should give up? Although Susan B. Anthony began the effort in 1848, it took American women 75 more years before they won the right to vote.
       Like any complex technology or social reform, PRT has taken time to develop. In many ways it was ahead of 1950-70s technology, and had to wait for the times— such post-1970s innovations as microprocessors and composite materials— to catch up.
LRN Claim REALITY
Low capacity
(Report Section: "Constrained line capacity")
Quote: "... relatively miniscule capacity of the system, with its tiny cars... has proven to be PRT's Achilles heel, time after time."
Quote: "At a safe stopping rate of 5 mph per second... it would take 8 sec for a 40-mph vehicle to come to a complete stop..."
LRN does not list any times that low capacity has been PRT's downfall. Frankly, SoundPRT doesn't know of any either. The only "PRT" system ever deployed is in Morgantown, WV—and LRN tacitly acknowledges that it is Group Rapid Transit, not PRT.
       Under LRN's thinking cars and roads would be low capacity, since most cars only have one or two people in them, yet somehow currently over 90% of all travel is in cars on roads. Because PRT offers non-stop service near, or directly to, where people want to go, it is far more likely to reduce the number of cars on the road than other transit.
LRN tries to show that headways between PRT vehicles of 1 second or less are unsafe. Instead they say that 'response time' and weather requires a relatively long headway of 16 seconds, therefore reducing capacity.
       The hypothetical emergency stop situation would not be a problem for PRT. What LRN forgot is that in a PRT system all the vehicles are under central supervision and control—vehicles are never surprised by a so-called "brick wall stop."
       Here's an accurate scenario: PRT car B is 1 second behind PRT car A. If car A unexpectedly slows down, the system senses it and reacts, ordering car B to match the deceleration. What is the central system's response time? On the order of milliseconds (thousandths of a second). If car A continues to slow and then stops, car B stops. How long does this take? Longer than LRN believes: the act of deceleration actually lengthens stopping time.
       Could a PRT car stop if an obstacle suddenly falls in its path? It could, depending on the system design; anti-collision radars are now available in some newer automobiles.

Quote: 5 mph/sec over 8 sec stopping time, " response time... to detect an obstacle... [and] weather conditions impacting guideway adhesion, etc—at least doubling the "bare minimum" stopping time"
Poor "guideway adhesion" due to bad weather is not an issue, as already noted above in the discussion of 'system droppings', if the PRT system uses induction motors (magnetism; i.e., the wheels don't do the propulsion and braking).
Thus, the entire line of reasoning LRN uses to arrive at "minimum safe" headway of 16 seconds is erroneous.
LRN does not mention how light rail would perform in a comparable emergency-stop situation: a multi-ton light rail train definitely cannot stop in time. Under LRN's logic, light rail should be abolished.
Quote: Sub-second headways are "based entirely on speculation"
Theory is speculation, research-verified science is not—and PRT has been extensively researched. Anyone can read the material at Innovative Transportation Technologies.
Actual PRT testing at >= 1 sec headway bears out the science:
1970s: With technology of that decade, Cabintaxi operated safely with 2.5 second headway Video
1990s: Raytheon's PRT2000 successfully completed its testing program; they did not make any sales due to already-well-documented and preventable cost problems.
2002-03: The ULTra system was tested in Cardiff, Wales. APPROVED by the UK Government (Her Majesty's Rail Inspectorate) to carry members of the public.
No designers plan to take headways below 1 second until such intervals are verified by the same kind of testing.
LRN Claim REALITY
PRT can't handle crowds
(Report Section: "Congestion in PRT stations")
Quote: "disconnect with the realities of human group dynamics and transit passenger behavior"
LRN mistakenly thinks group dynamics and passenger behavior are the same! The assumptions in their imagined PRT station scenario are deeply flawed, reflecting a failure to appreciate the full implications and benefits of "on-demand" transit.
LRN gets right that PRT would almost totally eliminate waiting time, but then gets wrong the way this would be accomplished:

Quote: "it is not uncommon for 200-300 people to crowd onto transit platforms"
Quote: "...accommodating 200 people arriving almost simultaneously would require... about 50 of the little cars... which implies a 500-ft-long platform to hold all the little cars necessary to keep that peak crowd from waiting."
Large crowds are true of TRAIN stations. In a train station, people do not arrive simultaneously. They walk in at various times, but then are forced to wait for their scheduled train. It is the train timetable that is responsible for large crowds. In PRT, crowds would not accumulate because each person or small group would board a PRT car almost immediately.
LRN cannot seem to shake the idea of the timetable. They assume that if 200 people are coming to a station, 50 PRT vehicles have to be all queued up and waiting, and depart at the same time. But that's not how PRT works. PRT is demand-based, vehicles begin trips whenever someone enters the station, selects a destination, boards, and the vehicle gets under way. "75-100 tiny PRT cars" don't leave at the same time, but as dictated by individual —unscheduled—demand. Then the objective is to make sure that an empty vehicle takes its place in the station. This can trigger the system to simply send an empty vehicle that is nearby. But this is not needed if the system knows a vehicle is already about to deliver a rider(s) to the station; when it arrives, it becomes the waiting vehicle.

Quote: "it is unlikely that... boarding would be so smooth and simultaneous"
Exactly because PRT boarding does NOT need to be simultaneous, vehicle flow through the station is not seriously delayed if the occasional passenger holds a door open a little longer. Furthermore, systems like Skyweb, MicroRail and ULTra have been designed so that persons in wheelchairs can easily roll aboard.

Quote: "PRT cars unable to enter a station would... keep circling the great loops... taking perhaps 10-12 minutes to do so."
Since there are no crowds, and vehicles don't have to queue for riders, stations do not have to be the "750 to 1000 ft-long" that LRN thinks "obvious", nor would "small-vehicle traffic jams" materialize. On rare occasions when a station is briefly 'full', PRT systems allow for a "wave off": the arriving vehicle must make a circle of the local loop and reapproach the station. But this would not take "10-12 minutes". An optimum PRT network loop is 1/2 mile on a side, for a total of 2 miles; traveling this loop at 40 mph would take 1.3 minutes.
A factor that contributes to the number of people using each station is the distance between them. In a train system, each station might be responsible for rider demand across several square miles of a city. In a city like Seattle, with an average density of about 6300 people per square mile, it's then no wonder that light rail advocates worry about each train station being able to serve hundreds, if not thousands, of people per hour. Contrast this with PRT, where there could be 4-5 stations per square mile. Riders would go to their nearest PRT station, thereby distributing demand instead of concentrating it in a few big stations.
      Here is an example of the advantage of distributed demand. Again using Seattle's 6300 people/mi², if a square mile has 5 stations, each station serves 1260 residents. Even if ALL of those people use PRT twice a day, 2520 trips, that is only 105 trips per hour per station—1.75 per minute. LRN might object that such "low" demand makes a station uneconomical. It might, but only if a PRT station were as expensive as a train station; a standard Skyweb station would be about $280,000.
      And in peak commute periods? Even if the station attracted 15% of all daily trips in a single hour, that is 378 trips: only 2.1 trips per berth per minute in a standard 3-berth station.
With many stations and the ability to recycle vehicles, and freed from artificial demand surges caused by timetables, PRT can supply a station with as many vehicles as it can reasonably be expected to require.
LRN Claim REALITY
PRT is "energy wastefulness"
(Report Section: "Congestion in PRT stations")
Quote: "incessant flow of little PRT cars... virtually all the time... Because... the guideways themselves serve as storage areas"
Quote: "The energy wastage of this bizarre scenario of ever-moving pods"
These passages reflect a misunderstanding of PRT 'recycling' and 'storage'. As explained above, vehicles move only when carrying passengers, or to anticipate eventual demand by going to a station where there is an empty berth. Vehicles waiting in stations and not being used are only technically "in storage"—they are still on-call, instantly available to riders. And also as explained above, 'wave-offs' would add little time to a trip. Therefore, if there is movement of PRT vehicles it's to make sure service stays as close to on-demand as possible.
      Any "wasted" energy in PRT is probably hardly worth mentioning when compared with trains and buses, which waste significant amounts of energy due to frequent starts and stops. Trains and buses must still make their scheduled runs even with few passengers—and even if empty. "Energy-efficiency, anyone?"
LRN Claim REALITY
PRT is expensive
(Report Section: Cost realities of PRT)
Quote: "PRT has repeatedly seemd to offer far too little bang for the buck."
Here LRN begins by restating as a given that PRT is "extremely light-capacity", which they failed to prove above. But on what does LRN base the high cost contention?
First LRN states PRT cost projections are unrealistically low, citing a 1971 Popular Science estimate of "$4 million per two-way mile", and then brings in a 1973 estimate of $1.5 million/mile— in both cases omitting the inflation factor. SoundPRT invites the reader to enter the figures in the Federal Reserve Consumer Price Index Calculator to find out what they would be worth today. It turns out those estimates were quite reasonable for the early 1970s.
LRN discovers inflation in its next example, the cost of the Morgantown peoplemover system, which it gives as $319 million in 2004 dollars. But LRN admitted earlier that Morgantown's "PRT" is not true Personal Rapid Transit, so this example is irrelevant.
The Raytheon PRT2000 is the next case. LRN recounts the unrealized project to install a PRT system in the Chicago suburb of Rosemont. LRN mentions that Raytheon licensed the technology invented by J. Edward Anderson and Taxi 2000 Corporation.
Several passages from a 1996 article are cited, including:
"Headway is limited by switch maneuvers... requiring enormous reliability in sensing, computing, braking, and other controls"
The presence of this passage in a section about PRT cost implies that headway, automation and control issues contributed to PRT2000's high cost. In fact, operational issues had nothing to do with it, PRT2000 successfully completed its testing program. LRN omits a crucial part of the story: Raytheon did not stick to Anderson's original specifications. They instead created a larger, heavier vehicle and guideway that used more materials than necessary. Installation in Rosemont would have been more costly and time consuming. The result, predictably, was an economic failure.
LRN Claim REALITY
PRT has never worked
Quote: "every single attempt... to develop a PRT project has failed"
(Report Section: "PRT's long history of failure"
Having heretofore mischaracterized PRT and the reasons for its slow development, LRN now arrives at the wrong conclusion: that PRT has always failed because PRT can't work. Even so, LRN proposes that a college campus might be a good place for PRT. But the conditions given, "need for grade separation" and "meandering origin-destination" are true of a city too! LRN also wonders why Dr. Anderson's University of Minnesota has not constructed a PRT demonstration, as though that is the job of a university.
LRN next cites Dr. Vukan Vuchic, virtually the patron saint of light rail, to deliver the definitive argument against PRT: "the basic concept of PRT is inherently unsound." But LRN, as well as others who often cite him, neglect to mention that Vuchic's claims about PRT have all been debunked on the Innovative Transportation Technologies PRT Debate Page—by Anderson and others.
The most important point is that LRN is just plain wrong when it claims every PRT project has failed. There have been many technical successes, already mentioned above.
       Today, there are three programs past the design stage and constructing hardware: Skyweb Express, MicroRail and ULTra. Skyweb has attracted enough private investment to start a demonstration facility; a site selection process is underway. As noted, ULTra has been certified by the British government to carry passengers. ULTra's final obstacle? Political resistance from a traditional pro-transit coalition that wants only trains and buses.

Conclusion

In the final section of Personal Rapid Transit - Cyberspace Dream Keeps Colliding With Reality, the authors seem to ask Why does PRT keep getting support, when it fails? Why not just build more light rail, since PRT is "all in cyberspace"? The "no working prototype" claim is repeated here, again forgetting about ULTra's successful testing and safety certification. The term 'gadgetbahnen' is tossed in, though only the most dedicated transit wonk would know what it means (actually a reference to the title of a discredited 'non-analysis analysis' of PRT {L.W. Demery & M.D. Setty, 2002}). Well, if LRN wants to know why past PRT projects were underwhelming, their Publication Team could have stopped writing all the way back on page 1, paragraph 2:

"Early would-be PRT installations... eschewed any attempt to provide true PRT-style, small-vehicle, customized origin-destination service, and were implemented in effect as line-haul automated guideway transit." In other words, if the implementers ignore the designers, the result is something other than Personal Rapid Transit.

Look at a few of the authorities and opinion leaders who think PRT is worth studying— the European Union; Duluth, MN; Minneapolis; State of Washington; King County, WA; State of New Jersey; Norwegian telecom giant Telenor; Marina, CA; the developers of Syracuse's DestinyUSA; Finland's TechVilla; the President of the Congress for New Urbanism (p.4). Does LRN know better than all these?

Even Sound Transit, the agency implementing the controversial Seattle "Link" light rail project, features PRT in the Innovation section of its Sound Move strategic plan (p.20), and is currently engaged in a feasibility study of PRT as a collector-distributor system for the planned Tukwila, WA light rail station. The SoundPRT organization supports this study, as we see PRT not only capable of serving as a standalone transit system, but also as a complement to other modes including express bus, monorail and, yes, light rail. This position is drawn from the paper A PRT Deployment Strategy to Support Regional Land Use and Rail Transit Objectives by J.B. Schneider, and is enshrined on the Strategy page of this website.

PRT would help light rail work better, by providing a solution to the "last mile" problem. PRT will go a long way toward equalling or exceeding the service of the automobile, while offering the benefits of transit (capacity, efficiency, etc.). With improved mobility will come increased productivity, and we will all save time that can be spent on business, pleasure and with our families. Will LRN work with the international PRT community to improve the service experienced by transit patrons, as we are willing to work with Sound Transit? Or will LRN reflexively keep saying No to innovations in transit technology?

All the preceding true facts about PRT are readily available on the Internet for anyone to read, much of it at ITT, the vendors' websites, and education websites such as SoundPRT's Get There Fast. The information on the past and present of PRT is out there for LRN to find. But their report only included past failures, and left out the true story of PRT's successes—and its promise.

SoundPRT
Research & Inquiry Committee, March 2004

Other rebuttals

Dr. J. Edward Anderson

Steve Anderson

Bill Wilde

References

Innovative Transportation Technologies

ITT's PRT Debate Page

Skyweb Express, The System

Urban Light Transport (ULTra), Downloadable Files

MicroRail videos

Flyby PRT Consortium

Safe Design of Personal Rapid Transit Systems