Making Tracks: Will Taxi 2000 be the Rapid Transit of the Future?

This article was originally published in Bostonia magazine in the January/February, 1988 issue. It is posted here with permission from Bostonia (a Boston University publication) and the author.

by Jon Queijo

When you are riding the Green Line (in Boston) above ground this February, think of Frank J. Sprague. Exactly 100 years ago Sprague applied his overhead switching system to a streetcar in Richmond, Virginia. Prior to 1888, cars could only run in a straight line. But, with the introduction of the switch, streetcars were able to change tracks. Sprague's invention not only revolutionized the transportation industry, it became the mainstay of contemporary urban mobility.

Now consider J. Edward Anderson. He's one of a handful of people who has been working to improve Sprague's system. In fact, he's spent a good part of his 60 years trying to convince the government, academia, investors and the general public to consider his transportation idea for the future. He calls it Taxi 2000 . It is a form of personal rapid transportation (PRT). And Anderson believes that if it is implemented, you'll find his name in Encyclopedia Britannica alongside Sprague's.

Taxi 2000? Its evolution goes back some three decades to a time when Anderson, like other visionaries of the 1950s and 1960s, realized that if computer technology merged with transportation innovations, many urban traffic problems could be solved. In theory, PRT has the appeal of a well-crafted work of art; individual elements --economy, technology and service-are so closely related that the result is a masterpiece: of efficiency unheard of in conventional mass transit. Its concept is simple: If passengers are carried in small, three-or four-person vehicles, guide rails can be smaller and, hence, less expensive. Less costly guide rails not only mean more track can be built, but the track can be elevated to simplify construction. Finally, using automated cars and stations on bypass tracks allows for "demand-responsive'' service. Every trip is nonstop: Once passengers program their car for a station, the car travels directly to it, not stopping until it arrives at the station and switches onto its bypass track. In addition, computers send unused cars to stations to wait for passengers.

On paper PRT sounds great, and in the late 1960s and early 1970s, governments in the United States and abroad liked the idea enough to fund research. Unfortunately, Anderson and others suffered numerous technical and political setbacks. By the late 1970s these early failures, combined with cutbacks in federal transportation research, led to an impasse in PRT development. Recently, however, a growing number of transportation authorities have begun to reconsider PRT for three reasons: Traffic problems in many cities are getting worse; conventional mass transit is, by design, unable to solve the problem; and conventional mass transit is losing money.

"There's beginning to be a perceptible feeling of frustration, and sometimes when you're frustrated you're willing to open your mind to something else explains Jerry Kieffer, secretary/treasurer of the Advanced Transit Association (ATRA) in Washington, D. C., and a public policy consultant on problems of urban transportation.

Kieffer points out that today many U.S cities and suburbs are becoming strangled by their own traffic because of increasing "cross-traffic." That is, as jobs have moved to the suburbs, traffic in metro areas has changed from "corridors" in and out of the city, to traffic moving in all directions. Although ATRA does not officially advocate any one transit concept to solve this problem, it is now leaning toward PRT as a solution. "Once you gain acceptance for PRTs explains Kieffer, "they'll do the whole job"

Like many others in the field, Kieffer believes that if anyone can make PRT work, it is Anderson. By all accounts the Boston University engineering professor has been studying PRT systems longer than anyone in the world. His credentials include chairing three international conferences and remaining active in the field in the 1970s while others dropped out as government funding dwindled. It's probably no exaggeration to say that from 1968 to 1980 Anderson suffered more frustrations and setbacks than anyone; by the same token, his persistence has given him expertise in the technical, economic and political aspects of PRT

Thus in 1981, having spent 13 years studying the pros and cons of other people's PRTs, Anderson sat down to design his own system. After sifting through the best of past systems and adding a few innovations of his own, he came up with Taxi 2000. Far from being modest, Anderson is touting Taxi 2000's credentials when he states, "The system has built on the work of literally hundreds of inventors around the world.

No one doubts Anderson's word. "He's done more investigation with solid engineering companies behind him than anyone in the world:' says Kieffer. Martin Bernard, a transportation systems engineer at Argonne National Laboratory, agrees. Argonne is now requesting funds from the state of Illinois for an economic feasibility study of Taxi 2000, and he states, "If there is a PRT that's going to work, I think Taxi 2000 will."

Don't be misled by the crude models and "mere" paperwork that now comprise Taxi 2000. Anderson estimates that about $3.5 million worth of brainpower has gone into developing the system- $2 million spent in grants and research since 1968, and about $1.5 million invested since 1981. Much of that money has gone into studies to prove that the various components of Taxi 2000 can actually be built. Unlike earlier PRTs, which were built with inadequate forethought to economic and technical feasibility, Anderson's research is at a point where construction could begin tomorrow. "We're beyond the research and development stage he says confidently. "We can build it now"

The next step, however, is a big one. Anderson needs to build a full-scale demonstration -- an oval track about 700 feet by 240 feet, with eight vehicles and one off-line station -- to test the technology. But he needs about $6 million to build the track, and therein lies Taxi 2000's fate. Because of the drastically reduced federal funding pool, Anderson is trying to sell the system as a private businessman. So far, the business road has been rocky -- venture capitalists shun mass transportation because of its long-standing reputation as a money loser. But there are signs of promise.

Currently, Anderson is negotiating with a number of private investors and businesses across the country. He is working with the Raytheon Company, which has agreed to pursue studies of Taxi 2000 applications through company operating units. And with this newfound legitimacy, Anderson, who functions as president and chief executive officer of Taxi 2000 Corp., is negotiating with financiers.

Anderson believes business will take off once the demonstration track is built, and he poses a strong argument that its time has come. "There really hasn't been an innovation in ground transportation for 100 years he points out. "There was enormous activity from the 1880s to 1910 with the development of the electric streetcar and automobile, but then there was hardly anything until the 1950s. He adds that since Sprague introduced the overhead switching system most of us have taken mass transportation for granted. It is hard to believe, he adds, that the systematic loading of people onto large vehicles and toting them across the city did not appear in the United States until 1831, when John Stephenson, a well-known carriage builder, began charging New York City residents 12.5 cents to ride horse-drawn omnibuses along Broadway.

Stephenson spearheaded another period of change in 1832 by capitalizing on a 56-year-old innovation: the iron rail. Compared to cobblestone streets, the use of rail made it easier for horses and mules to pull coachloads of passengers. Stephenson designed two new coaches, each with a seating capacity of 30, for the New York and Harlem railroad line. When service began along the Bowery from Prince Street to 14th Street, New York -- a city of 200,000 -- could boast that its track was the first local passenger rail system in the world.

It took a while for other cities to catch onto the idea of horse-drawn mass transit, however. Massachusetts, one of the first, didn't open its first street railway, a Boston-to-Cambridge line, until 1856. Philadelphia, Cincinnati, Baltimore and Chicago followed a few years later. After that, the industry took off.

Meanwhile, of course, Sprague was perfecting his overhead system, which launched the next transportation revolution. Since its advent 100 years ago, there have been numerous refinements.

Nevertheless, the basic approach to urban transportation -- loading people onto large, scheduled trains that travel on limited routes--has not changed. Not that the technology for another revolution doesn't exist. One of the reasons it hasn't occurred, according to Anderson, is that engineers in the past 20 years have often applied the technology with too much enthusiasm and too little appreciation for the complexities involved.

In the giddy aftermath of the Apollo moon landings, Anderson explains, engineers -- many of whom had transferred from aerospace to ground transportation would brag "We can do the difficult today and the impossible by tomorrow." Some people believed they could design a new transit system in two years. The nine or so PRT systems that were designed but never got beyond the test track -- and numerous other automated transit system failures -- are ample testimony to this misconception.

The U.S. government's involvement in funding innovative transportation systems got a promising start in 1966 when the Urban Mass Transportation Administration (UMTA) funded 17 studies ( see review article ). The studies investigated new systems that "took into account the most advanced available technologies and materials." Although some promising ideas resulted, during the Nixon administration new UMTA leaders had little commitment to the program. UMTA then began taking on a role it would play through the 1970s and 1980s amid budget cutbacks -- ignoring new technologies in favor of funding existing bus and rail systems.

Still, several PRT systems were researched in the 1970s, and if few designs succeeded, many feel much of the blame belongs with UMTA (now called the Federal Transit Administration (FTA). Darwin Stuart, now manager of Strategic Planning for the Chicago Transit Authority, was a consultant who did research on UMTA-funded projects in the 1970s. He is not alone when he contends "UMTA has been lacking over the last 15 to 20 years in sponsoring research in this area. When they funded projects they bungled them -- provided inadequate funding, selected the wrong contractors and the wrong locations."

One common problem facing early PRT development was that as designs went from the drawing board to the test track, engineers or legislators succumbed to pressure from conventional streetcar interests. Three-passenger vehicles ballooned to six-, 12-passenger or even larger vehicles. This resulted in larger, more expensive guideways and defeated the purpose of PRTs. A classic example of inadequate planning, mismanagement by UMTA, cost overruns and after-the-fact engineering is seen in a system currently operating in Morgantown , Virginia. The vehicles were originally designed for six passengers, but were later expanded to accommodate 12. As a result the larger guide rails were not only more expensive, but accumulated more snow. This was "solved" by installing heated pipes in the track, which in turn led to energy costs per year twice what it costs to actually run the vehicles. For these and other reasons, the system is considered an economic disaster.

Anderson's involvement in transportation dates back to 1968 when as an associate professor at the University of Minnesota he became involved in a UMTA-funded study of new ground transportation technology. Then in 1979, two Indiana legislators--who had been impressed by Anderson's talk about PRTs a few years earlier -- got $300,000 from the state to plan an automated transit system for Indianapolis. They chose Anderson as lead consultant, and he and his associates set out to locate the best PRT design. After studying numerous systems, they chose the German-built Cabinentaxi -- a boxy three-person vehicle that travels below its elevated rail in one direction and on top of it in the other. Though it was the best PRT available in the late 1970s, Anderson was well aware of its technical shortcomings, including a wide track (five feet), an expensive switching system and an old-fashioned computer control system. In the end, however, it didn't matter. The Indiana contract ended, and again government did not come up with the money needed to proceed. Although Anderson was later hired by the German builder of Cabinentaxi to act as U.S. representative, the German government eventually abandoned PRT funding, and the company gave up on it. By the 1980s, virtually everyone had given up on PRT for two basic reasons: more technical research was obviously needed, and government was too saddled with keeping conventional systems running to fund innovative ideas.

"So I said in the summer of 1981, 'What am I going to do next?'" recalls Anderson. "I figured I might as well start designing a new system because by now we'd accumulated a lot of data about what did and didn't work. And even though something like $2 billion of work had gone into PRT, most of it was junk!" When Anderson sat down to design Taxi 2000, he began sorting through some 27 categories of engineering tradeoffs -- for example, the advantages of vehicles that travel on the track, versus those that hang down like gondolas; or of running vehicles on wheels, versus air cushions, versus magnetic fields. According to Anderson, in considering the numerous trade-offs, "I found something like 830 million ways of designing a new transit system" ( see trade-offs page for details ).

Fortunately, Anderson's 15 years of work simplified many decisions. Even so, various problems critical to the economics of PRT had never been solved, and Anderson focused his engineering ingenuity on designing three important elements: a narrow, but structurally solid guideway that was impervious to weather, easy to manufacture and assemble and easily adjusted to provide a smooth ride; a switch that was simple and inexpensive, yet could stand up to the constant switching required in a PRT network with off-line stations; and a computer control system that could safely merge and space cars.

Anderson achieved all of these goals and, with the help of a $100,000 patent grant from the University of Minnesota, received five patents for his efforts: two for the guideway, two for the switches and one for the computer control system. With his reams of diagrams and calculations complete, Anderson has since worked with electronics and construction companies to make sure his designs can be practically built. He has been assured that they can, and probably for less than his estimates.

It is when the system is up and operating, however, that Anderson believes Taxi 2000 will prove itself an economic gold mine. For one thing, the three-foot-wide by three-foot-deep guide rail weighs a mere 140 pounds per linear foot, compared to conventional railway weights of 2000 to 3000 pounds. This translates directly into numerous construction and labor savings; while conventional rail systems cost as much as $50 million per lane mile to install, he estimates Taxi 2000 can be built for about $7 million per mile.

Other economic savings will appear during actual operation. Thanks to off-line stations, Anderson estimates energy efficiency will triple by saving power normally expended in stopping and starting large vehicles. Another advantage of off-line stations is that they can be added into existing networks without slowing traffic flow. Anderson estimates that for the same cost as a conventional system, a Taxi 2000 network, with stations spaced at one-quarter of a mile, could cover 40 times as much urban area.

The features go on and on. Automated cars don't require salaries for drivers. Demand-responsive cars are used only when needed, avoiding the inefficiencies of scheduled trains that often run partly or mostly empty during off-peak hours. Since cars, unhindered by on-line stations, can run at an average of 25 miles per hour, Anderson estimates his three-foot guideway will have the capacity of a 300-foot-wide freeway.

Apart from the economic advantages, Anderson is most enthusiastic about Taxi 2000's service features. "The beauty of it is that you end up with what I call humanizing technology. It's a system with predictable, nonstop trips. Everyone has a seat and there are no transfers. You can travel with your companions, and it can be used by anyone, young or old".

If all that's not enough, one final element should make the system irresistible to venture capitalists. Thanks to the off-line stations, freight or mail can be carried on the guide rails without slowing passenger service; the additional revenue from such use could help Taxi 2000 provide one final element unheard of in mass transit: Profits.

This last feature convinces ATRA's Jerry Kieffer that investors cannot afford to ignore Taxi 2000. "Why venture capitalists haven't seen the profit angle puzzles me. The market is there. If the system ever gets built, I have no doubt that you're talking about a $5 - 10 billion industry within a couple decades.

Perhaps, but anyone familiar with PRT history knows the variety of political barriers Anderson is up against. Because of those barriers, former UMTA consultant Stuart disagrees with Anderson's approach on one point. He contends it would be wiser initially to sell a 10- or 12 person version of PRT "Even though they don't give the optimal solution that smaller cars do he explains, "smaller PRTs require a greater initial investment. With Taxi 2000 you have to build a large network to make the system work, and it's too big a decision for a locality to make." Stuart is probably right; Taxi 2000 will have to prove itself in smaller applications before anyone will risk investing in large networks. But Anderson is patient enough to take that approach: Initial applications, he says, will probably be in "small, politically simple" situations. Argonne's proposal, with the possibility of using Taxi 2000 to connect hospital buildings, is one example.

In fact, many business groups have shown a similar interest in using Taxi 2000 in small applications. Investors in Madison, Wisconsin; Denver; a suburb of Seattle; and even the Loon Mountain ski resort in New Hampshire have discussed funding the $6 million test track. Despite the promising negotiations Anderson says he is speaking to people in about two dozen cities --he will not be satisfied until the first deal is made and construction on the test track begins. "The problem:' he sighs, "is getting the thing organized. It's a staggeringly big business, totally new, and it goes against traditional transportation planning ideas.

That's not all Anderson is up against. Since 1981 he's had to restructure Taxi 2000 Corporation twice as business arrangements fell through. Already there have been mismanagement difficulties, a takeover attempt and a misunderstanding about Taxi 2000's technical requirements that could have led to the downfall of the system. He shrugs off these problems.

If he is less concerned about the vicissitudes of business, however, he is adamant that the integrity of Taxi 2000's engineering remains intact. "My biggest worry is that we give it to somebody who doesn't really understand it. It's not just money and amount of control, but being sure it's built right. If we put this into the hands of the wrong people, I'm afraid you can kiss the future of transportation good-bye."

Unusually strong words for Anderson. But Anderson is a level-headed person who insists on proceeding one step at a time. And that means knowing when to say "No." When some backers recently urged him to pitch his system to developers of Boston's Harbor Tunnel, for example, Anderson knew from past experience it was a bad idea. "It didn't make sense to attack a big existing project with lots of contractors lined up and government people who had invested a lot of time he explains. "In situations where there's bound to be a lot of opposition, I tell people the best attitude is to be a coward and run away."

Anderson is patient, but only because he is convinced that Taxi 2000's time is coming, and when it does come, things will happen quickly. He likes to point out that "The decade of the 1880s is full of examples of electric streetcars that were built but, because they couldn't switch, never went anywhere" Once Sprague solved the problem, however, "The transportation system took off." The same may prove true for Taxi 2000, given Anderson's recent innovations in guide rail, switching and control technology. Even with the technical problems licked, however, he faces one final barrier: 156 years of urban transit history--huge trains hauling crowds of people along limited routes -- and the cultural prejudice that it can't be done any other way.

Last modified: November 10, 2005