Personal Rapid Transit: How Innovation Can Make Transit
J. Edward Anderson
GroundSwell, September-October 2006]
The following presentation was given during a panel on "Practical
Solutions to Transit Problems" at the annual Council of
Georgist Organizations conference held July 20, 2006 at Des
Plaines, IL. Moderator was Chuck Metalitz, director of the Henry
George School of Chicago, IL. J. Edward Anderson is a Professor of
Engineering at the University of Minnesota
In spring 1989 I was informed that during a luncheon attended by a
Northeastern Illinois Regional Transportation Authority (RTA)
Chairman it was agreed that "We can't solve the problems of
transportation in the Chicago area with just more highways and more
conventional rail systems. There must be a rocket scientist out
there somewhere with a new idea!" The Illinois Legislative Act
that established the RTA had given the new agency an obligation to "encourage
experimentation in developing new public transportation technology."
The new idea they needed was and is High-Capacity Personal Rapid
Transit (HCPRT). A March 2006 European Union Report concludes: "The
overall assessment shows vast EU potential of the innovative PRT
In April 1990 the RTA issued a request for proposals for a pair of
$1.5 million Phase I PRT design studies. Two firms were selected and
after the studies were completed the RTA selected one of the designs
for a $40 million Phase II PRT design and test program.
Unfortunately, that program was not directly successful, not due to
any flaw in the basic concept of High-Capacity PRT, but to
institutional factors. There is more and more evidence that HCPRT is
an important answer to many urban problems.
Derivation of the New System
It will not be possible to reduce congestion, decrease travel
time, or reduce accidents by placing one more system on the streets
— the new system must be either elevated or underground.
Underground construction is extremely expensive, so the dominant
emphasis must be on elevation. This was understood over 100 years
ago in the construction of exclusive-guideway rail systems in
Boston, New York, Philadelphia, Cleveland, and Chicago. The problem
was the size and cost of the elevated structures. We have found that
if the units of capacity are distributed in many small units,
practical now with automatic control, rather than a few large ones,
and by taking advantage of light-weight construction practical
today, we can reduce the weight per foot by a factor of at least
With this finding in mind consider the cost of a fleet of transit
vehicles. The cost of the fleet is the cost per unit of capacity
multiplied by the capacity needed to move a given number or people
per unit of time. The major factor that determines the capacity
needed is the average speed. If the average speed could be doubled,
the number of vehicles required to move a given number of people
would be cut in half. The greatest increase in average speed without
increasing other costs is obtained by arranging the system so that
every trip is nonstop. The trips can be nonstop if all of the
stations are on bypass guideways off the main line.
Off-line Stations are the Key Breakthrough!
- As just mentioned, because of increased average speed, off-line
stations minimize the fleet size and hence the fleet cost.
- Off-line stations permit high throughput with small vehicles.
- Off-line stations make the use of small vehicles practical,
which permit small guideways, which minimize both guideway cost
and visual impact.
- Off-line stations permit nonstop trips which decrease trip time
and increase the comfort of the trip.
- Off-line stations permit the vehicles to wait at stations when
they are not in use instead of having to be in continuous motion
as is the case with conventional transit. Thus, it is not
necessary to stop operation at night — service will be
available at any time of day or night.
- There is no waiting at all in off-peak hours, and during the
busiest periods vehicles are automatically moved to stations of
need. Computer simulations show the average wait time will be less
than a minute.
- Stations can be placed closer together than is practical with
conventional rail. ... With off-line stations one has both speed
- Off-line stations can be sized to demand, whereas in
conventional rail all stations must be as long as the longest
- All of these benefits of off-line stations lead to lower cost
and higher ridership.
The Optimum Configuration
During the 1970s I accumulated a list of 28 criteria for design of
a PRT guideway. As chairman of three international conferences on
PRT, I was privileged to visit all automated transit work around the
world, talk to the developers, and observe over time both the good
and the bad features. ...
I compared hanging, side-mounted, and top-mounted vehicles and
found ten reasons to prefer top-mounted vehicles (which) ... can
accomodate a wheelchair, 3 adults plus fold down seats for small
people, riders plus a bicycle, 2 people with luggage, or 2 adults
with baby carriage.
Is High Capacity Possible with Small Vehicles?
...In 1973 Urban Mass Transportation Administrator Frank Herringer
told Congress that "a high-capacity PRT could carry as many
passengers as a rapid rail system for about a quarter the capital
cost." The effect of this pronouncement was to ridicule and
kill a budding federal HCPRT program. The best that can be said is
that PRT was thought to be too good to be true. But PRT was not an
idea that would die.
System Features needed to achieve Maximum Throughput Reliably and
The features needed are
1. All weather operation. ...
2. Fast reaction time. ...
3. Fast braking ...
4. Vehicle length ...
These features together result in safe operation at
fractional-second headways, and thus maximum throughput of at least
three freeway lanes, ie., 6000 vehicles per hour.
During the Phase I PRT Design Study for Chicago, extensive failure
modes and effects analysis, hazards analysis, fault-free analysis,
and evacuation-and-rescue analysis were done to assure the team that
operation of HCPRT would be safe and reliable. The resulting design
has a minimum of moving parts, a switch with no moving track parts,
and uses dual redundant computers. Combined with redundant power
sources, fault-tolerant software, and exclusive guideways, studies
show that there will be no more than about one person-hour of delay
in ten thousand hours of operation.
How does a Person Use a PRT System?
A patron arriving at a PRT station finds a map of the system in a
convenient location with a console below. The patron has purchased a
card similar to a long-distance telephone card, slides it into a
slot, and selects a destination either by touching the station on
the map or punching its number into the console. The memory of the
destination is then transferred to the prepaid card and the fare is
subtracted. To encourage group riding, we recommend that the fare be
charged per vehicle rather than per person. The patron (an
individual or a small group) then takes the card to a stanchion in
front of the forward-most empty vehicle and slides it into a slot or
waves it in front of an electronic reader. This action causes the
memory of the destination to be transferred to the chosen vehicle's
computer and opens the motor-driven door. Thus no turnstile is
needed. The individual or group then enter the vehicle, sit down,
and press a "Go" button. The vehicle is then on its way
nonstop to the selected destination. In addition to the "Go"
button, there will be a "Stop" button that will stop the
vehicle at the next station, and an "Emergency" button
that will alert a human operator to inquire. If, for example, the
person feels sick, the operator can reroute the vehicle to the
At the present time, mid 2006, all of the technology needed to
build HCPRT, including all the control hardware and software, has
been developed. All that is needed in the United States is the funds
(about $10 million) to build a full-scale test system. Such programs
are already underway overseas. HCPRT is a collection of components
proven in other industries. The only new thing is the system
arrangement. The system control software has been written and
excellent software tools are available for final design verification
and development of final drawings needed for construction. Because
there has been no US federal funding to support the development of
HCPRT during the past three decades, few people in the United States
have been able to continue to study and develop these systems. The
problem is likely the major factor that caused the collapse of the
Chicago RTA PRT program. However, thanks to continued efforts of
members of the Advanced Transit Association
(www.advancedtransit.org), there is a sufficient number of people
able to lead HCPRT development — it does not take many.
Three PRT systems are currently under development. ULTra
(www.atsltd.co.uk) is being developed at Bristol University in the
United Kingdom. The Vectus system is being developed by the Korean
steel company Posco (www.vectusprt.com). They announced last fall
they would built a test system in Uppsala, Sweden. Microrail
(www.megarail.com) is one of a family of automated guideway transit
systems under development by Megarail Corporation of Dallas, Texas.
Economics of PRT
The Minneapolis light rail system is called the "Hiawatha
Line." The newspapers announced that its capital cost was
$720,000,000 and that the ridership would be about 20,000 rides per
day. That works out to $36,000 per daily trip. Since the annual cost
for capital amortization and operation is about 10% of the capital
cost and the annual yearly ridership will be roughly 300 times the
daily ridership, the annual cost divided by the annual ridership
works out to $12 per trip. The average trip length is roughly 6
miles, so the cost per passenger-mile is about $2. This compares
with the total cost per mile of an automobile of around 40 to 60
We laid out and estimated the cost of a PRT system for downtown
Minneapolis. It is compared with the Hiawatha light-rail line. Our
estimate was about $100 million capital cost and a professional
ridership study showed about 73,000 trips per day. Because this
system has not yet been built, let's double its cost. Then on the
same basis the capital cost per daily trip would be $2740 and the
total cost for each trip would be $0.91. On this PRT system the
average trip would be about two miles so the cost per passenger-mile
or break-even fare would be about $0.46.
What would be the cost per passenger-mile on a built-out PRT
system? The cost per passenger-mile on a square-grid PRT system as a
function of population density and for values of the fraction of all
vehicle trips taken by PRT, called the modal split, is shown to be
from 0.1 to 0.7. Several studies suggest that an area-wide PRT
system with lines a half mile apart would attract at least 30% of
the trips. On this basis, one can estimate the population density
needed for a PRT system to break even. Revenue will be obtained not
only from passenger trips, but from goods movement and advertising
as well — roughly half is a reasonable estimate, meaning that
a passenger would have to pay only half the amount determined. For
example, if the population density is 6000 persons per square mile
(Chicago density is about 13,000 people per square mile) and the
mode split to PRT is 30%, the total cost per passenger-mile is about
40 cents, of which the passengers would pay about 20 cents.
In over 90% of the autos there is only one person, occasionally
two, and very occasionally three (on a freeway running at capacity
which is about 6000 autos per hour on a 3-lane freeway with the 4th
lane just an acceleration lane.) A typical freeway width from fence
line to fence line is about 300 feet. The two PRT lines in the
middle take up only 15 feet of width, giving a width reduction per
unit of capacity of 20:1 or 5% of the land area. But, land for a PRT
system is required only for posts and stations, which is 0.02% of
the land area. The auto requires about 30% of the land in
residential areas and roughly 50% to 70% of the land in downtown
areas. The enormous land savings permits development of safe,
low-pollution, energy-efficient, quiet, environmentally friendly,
Minimum energy use requires very light-weight vehicles; smooth,
stiff tires for low road resistance; streamlining for low air drag;
and efficient propulsion, all of which can be designed into a PRT
system if the designer wishes to do so. Moreover, unlike
conventional transit, in which the vehicles must run to provide
service whether anyone is riding or not, PRT vehicles need run only
when people wish to travel.
Comparing energy use per passenger-mile of eight modes of urban
transportation — heavy rail, light rail, trolley bus, motor
bus, van pool, dial-a-bus, auto and PRT — PRT would be more
than twice as energy efficient as the auto system, which in turn is
almost twice as energy efficient as the average light rail system.
Benefits for the Riding Public
* The system will be easy for everyone to use. No driver's license
* The vehicles wait for people, rather than people for vehicles.
* The trip cost will be competitive.
* The trip will be short, predictable, and nonstop.
* There will be minimum or no waiting.
* Everyone will have a seat.
* The system will always be available at any hour.
* The vehicles will be heated, ventilated, and air conditioned.
* There will be no crowding.
* There will be no vehicle-to-vehicle transfers within the system.
* The ride will be private and quiet.
* The chance of injury will be extremely remote.
* Personal security will be high.
* The ride will be comfortable.
* There will be space for luggage, a wheelchair, a baby carriage,
or a bicycle.
Benefits for the Community
- The energy use will be very low.
- PRT can use renewable energy.
- Deployment of PRT will reduce transit subsidies.
- PRT can augment and increase ridership on existing rail
- PRT will be attractive to many auto users, thus reducing
- Seniors, currently marooned, will have much needed mobility and
- The system does not directly pollute the air. Being more energy
efficient than the auto system and by using renewable energy,
total air pollution will be reduced substantially.
- By spreading the service among many lines and stations, there
will be no significant targets for terrorists.
- As to accidents, no one can say there will never be an
accident, but the rate per hundred-million miles of travel will be
less than one millionth of that experienced with autos.
- There will be huge land saving: 0.02% is required vs. 30-70%
for the auto system.
- PRT will permit development of more livable high-density
- The ride will be pleasant for commuting employees, thus
permitting them to arrive at work rested and relaxed.
- PRT will permit more people-attracting parks and gardens.
- PRT will permit safe, swift movement of mail, goods and waste.
- PRT will provide easier access to stores, clinics, offices and
- PRT will provide faster all-weather, inside-to-inside
- PRT will enable more efficient use of urban land.
- By making the inner city more attractive, urban sprawl will be