The Economy of Light-Beam Traffic

A fool and his money are soon partying.

nterested people frequently ask what it would cost to build a light-beam railway. The question cannot be answered in exact figures today, as the technology is partially new. As far as the hardware goes, the usual rules apply: - Production of prototypes and short series are usually awfully expensive. - Production of a longer series, when prototypes have been thoroughly tested, is cheaper. On the whole, this is where we stand today, regarding beamcars and beams. In most respects, prototypes already exist. An exception to this, however, is the far-reaching design ideas advocated by SwedeTrack.

- Mass production of standardized equipment means that it would pay to streamline production facilities as Henry Ford did during the 1920-ies with streamlined production of standardized motorcars. He allegedly told customers that The Hillbilly model from 1918. Henry Ford created the first inexpensive mass-produced automobile -- the Model T -- and revolutionized American industry by developing and refining assembly line manufacturing. Ford began his working life as a machinist, then became an engineer with the Edison Illuminating Company. (He and Thomas Edison remained close friends for decades.) In his spare time Ford tinkered with creating a motorized vehicle, and in 1896 introduced the Quadricycle, a four-wheeled cart with a gasoline engine. Ford has often been credited with inventing the automobile, though historians now agree he was only one of many people who built motorized cars. In 1903 the Ford Motor Company was founded, and in 1908 Ford introduced the Model T. By 1924 10 million Model T cars had been sold and Detroit had become the auto-making capital of America. Ford remained one of the country's most famous and influential businessmen until his death in 1947. "Sure, you can get any color you choose, as long as you choose black!" It is clear that mass production would drastically reduce prices on beam traffic hardware.

- When competition finally enters this market, prices will really go down. The main key to low prices on components is, thus, to get production going in reasonably big quantities. This can only happen if city planners start production on a system that will grow reasonably quickly, thus generating new orders for parts continuously. Cost estimates Analysis of cost factors Requirements to achieve low costs How large-scale production affects the price

1. Cost Estimates

Figure 1:2

Figure 1:3

About the Histograms in figures 1:2 and 1:3 ne can see from figure 1:2 above that manufacture and construction of the guideway beams and manufacture of vehicles are the big costs. This assumes a densely populated urban area, where the average number of vehicles per kilometer guideway are very high. In rural areas, the cost of guideway would tend to dominate all other costs, because of the longer distances relative to the potential traffic demand. There are of course innovative suggestions on how to streamline guideway production. The People Pod website suggests manufacturing beams on the building site, using a huge robot. The more traditional way would be to transport the beams in sections, to be assembled on the construction site. The People Pod website also has suggestions for manufacture and pricing of vehicles. It should be noted, though, that both manufacture of parts and design and construction of guideways and beams would be considerably lower in developing countries than in the Western World, due mainly to differences in labor costs. Figure 1:3 shows how much various parts of the system cost, as percentages of total costs. There are, of course, many factors that influence these estimates, such as size of the vehicle fleet, total length of guideways, where and in how long production series they are manufactured, etc. These two figures (1:2 and 1:3) can thus only give rough ideas of what costs to expect. A Self-Financing System SwedeTrack has a separate page on this website that deals with the introductory phase of a beam network in a city. The financing part is always of interest. One method would of course be to let the taxpayers pay for this, wholly or partly. The generally preferred method of financing would probably be a proper balance of: Tax money to pay for land-leases and preparatory work in the city (such as providing electrical power supply, etc.). Loans to finance development and testing, and also the introductory traffic phase. Fares from the users to pay for operation, maintenance and purchase of more beam vehicles. Traffic tolls levied on motor vehicles using streets that are serviced by beam traffic, to pay for the further extensions of the beam network. After a break-even point, fares will increase more rapidly than needed to pay the items in point 3. Time to pay back the loans.

The statement in point 5 is of course due to the fact that: As the network grows more extensive, it will become more usefull for the users. Users will gradually accept the beamcars as viable transport alternatives, both because peoples will discover how quick and comfortable a ride can be (contrary to manys expectations?), and because more opportunities open up as the network grows. This effect can be compared to Internet; the more people it reaches, the more useful it becomes for various uses, thus becoming more valuable, thus being desireable to even more people, and so on. Increasing revenues over time will permit an ever faster rate of extension of the network, and also provide money for paying back loans. Profitability and competing technologies The most likely scenario is that public money will finance at least 50% of the building of a new beam traffic system, at least until it turns a profit. No private investors can likely collect (without financial guarantees from state or city) enough money for such a project. And there has to be political decisions behind such a venture, since it involves changing to a new traffic infrastructure. The risk is also big that conntracts will be offered to several competitors with various not quite developed prototypes, having inherited defects which might give the infrastructure as such a bad name. These companies will have as their main purpose to attract some money to cover accumulated losses, and to further develop their systems, at the risk of squandering public money to no avail. Moreover, there are a lot of competing systems for this public money: Fuel cells, ITS, battery cars, maglev, monorail, LRT, peoplemovers - GRT, etc. Experience from many other fields shows that it is usually the most efficient lobbying efforts that wins the public money. Nothing of a technical or economic nature forces the authorities to choose just PRT. Summing Up When considering both building costs and operating costs, the beam traffic system will, in the long run, be considerably cheaper than motor vehicle traffic. This would be true even if one did not consider the costs to the environment and social consequences (such as accidents) caused by today´s conventional traffic systems. This is because the beam-carried traffic per person- and ton-kilometer requires considerably less resources (land, material, labor and energy) than today´s traffic systems. Today, the costs of constructing the infrastructure of the beam system is on the same level as that of building a streetcar system. This is because the today's beam traffic system is manufactured in short series, manually, in workshops. This should be contrasted to the streamlined production facilities owned by today's motorcar manufacturers.

E.F. Schumacher (1911-1977), was a German-born British economist and businessman who advocated replacing large organizations with small working units and communal ownership of "alternative" technologies. In his 1973 book "Small Is Beautiful: A Study of Economics As If People Mattered," Schumacher said that modern economists are "used to measuring the 'standard of living' by the amount of annual consumption, assuming all the time that a man who consumes more is 'better off' than a man who consumes less. A Buddhist economist would consider this approach excessively irrational: since consumption is merely a means to human well-being, the aim should be to obtain the maximum of well-being with the minimum of consumption. The less toil there is, the more time and strength is left for artistic creativity. Modern economics, on the other hand, considers consumption to be the sole end and purpose of all economic activity." Prophetically, he further noted: “A civilization built on renewable resources, such as the products of forestry and agriculture, is by this fact alone superior to one built on non-renewable resources, such as oil, coal, metal, etc. This is because the former can last, while the latter cannot last. The former cooperates with nature, while the latter robs nature. The former bears the sign of life, while the latter bears the sign of death.” Later, in his most famous essay, he advocated a Buddhist form of economics based on “Right Livelihood” as part of the Buddha’s Noble Eightfold Path. Fundamental to such an economics would be simplicity and nonviolence, the importance of community, and the necessity and dignity of work. Schumacher was convinced that a sustainable form of economics must be found that would be appropriate as a path for the developing world, “a middle way between materialist heedlessness and traditionalist immobility.” He spent the rest of his life seeking and advocating that path. Schumacher was equally foresighted in his analysis of the industrial world. In 1958, before the founding of OPEC and to the disbelief of his colleagues, he warned that Western Europe would attain “a position of maximum dependence on the oil of the Middle East. The political implications of such a situation are too obvious to require discussion.” Even greater than his concern about the conflicts that would ensue was his fear of the possibility of a nuclear exchange. He became adamantly opposed to the use of nuclear energy. The accumulation of large amounts of toxic substances, he claimed, “is a transgression against life itself, a transgression infinitely more serious than any crime ever perpetrated.” Echoing the Gandhian philosophy of nonviolence he also wrote: “A way of life that ever more rapidly depletes the power of the Earth to sustain it and piles up ever more insoluble problems for each succeeding generation can only be called violent."

Figure 1:5

Revenues and Expenditures The graphical depiction of financing a new beam network shown in figure 1:5 can be illustrated along a time axis, as shown in figure 1:6. New prototype development will of course be rather expensive in the beginnig. But if we assume that this is not the first network of its kind, original costs will be lower and consist mainly of planning, providing power and the like (the dashed red line). At time T1 trial operation begin, at reduced fares, to wean the public with the new travelling mode. At time T2 regular operations begin, and revenues from passengers and freight increase as the network grows. These revenues will increase linearly, but they will increase faster than the rate of expansion of the network, because the farther the network reaches, the more people will be able to use it. The guideways will gradually be more and more efficiently used. The line for expansion of the network and purchase of more vehicles could of course be horizontal, by using some of the revenues for other purposes. Or the rate of expansion could increase over time, as shown in the illustration, if one prefers to use all the net revenues for expansion of the network. As stated above, streamlined production facilities makes purchase of large quantities of beam elements and vehicles less costly per item than purchase of smaller quantities. The costs for operation and maintenance will of course be roughly proportional to the size of the network and to the number of vehicles. But large networks can save money on this, too, compared to small networks, since large networks can benefit from streamlined maintenace procedures. Thus, looking at figure 1:7, one can see that, when the network grows, traffic grows even faster, as the beams are used more, and this makes revenues grow faster than the beam network expands. Operating and maintenace costs, on the other hand, do not grow at quite the same rate as the network grows, meaning that the profit margin will grow over time. In the figure, this margin is a lot bigger at time T2 than at time T1.

Figure 1:6 Figure 1:7

In the long run however (say 20 - 30 years), if we were to compare: an urban area based on a beam-traffic infrastructure to an urban area based on today's road traffic infrastructure, the costs of real estate (including bridges and tunnels) ought to be 2 - 4 times lower, the cost of material 3 - 7 times lower, the cost of labor 8 - 12 times lower, and the cost of energy for running the system 5 - 8 times lower in the case of the beam-traffic infrastructure as compared to the roadtraffic system, if we consider the same amount of transport work.These are rough estimates, but they illustrate the combined effects of: a) Mounting costs for all these factors in order to to maintain an adequate roadtraffic system when the number of cars are steadily increasing b) Increasingly streamlined production methods for producing the hardware to the beam system, coupled to a more rational running of the system as one gains experience and learns how to handle it. Regarding the figures above; they depend very much on the size and structure of the urban area, as well as the influence points a) and b) above have in the individual case. Thus, the more rational the production facilities for beam system components are, and the more expensive it becomes to maintain a working system of roads, the bigger the cost difference. In any case, the beam traffic system is a hell of a good investment!!

Design of PRT Guideways and their influence on Cost J. Edward Anderson, Ph.D., P.E. has this to say in his paper "The Design of Guideways for PRT Systems", of July, 1997: PRT guideways have been analyzed in this paper based on a result of dynamic analysis of vehicles moving across flexible spans by SWR, in which it was found that the shorter the headway between vehicles, the smaller is the difference between dynamic and static deflections, and that in the limit of zero nose-to-tail spacing between vehicles, the static and dynamic deflections are the same. This result occurs because in the limit case the load is practically uniform and non-time varying. The analysis is applicable to systems in which small vehicles operated at a full-range of speeds including speeds applicable to inter-city travel. The following results were found: The load-supporting weight of the guideway is directly proportional to additional loads applied, called the payload. If a guideway is stressed to its design limit, the ratio of guideway weight to payload is more than 50% higher with simply-supported spans than if the guideway is clamped to the support posts. The guideway width needed to resist wind or earthquake loads is less than the depth. The limit speed for acceptable ride comfort increases with span length, because longer spans mean lower crossing frequencies, and lower frequencies usually imply higher tolerable amplitudes of vertical acceleration. For given design stress, beam depth and span, the limit speed is higher for simply-supported guideways than for clamped guideways, but at a heavy price in guideway weight.

To minimize guideway cost, the vehicles must cruise at a crossing frequency higher than the fundamental natural frequency of the fully loaded guideway, but generally lower than the fundamental natural frequency of the empty guideway. This practice will be satisfactory if vehicles operating at less than maximum throughput are spaced randomly, as is the case if the control system is asynchronous. By use of active control, crossing frequencies that may be a problem can be avoided. This practice is new for guideway transit, but has been common for decades in the operation of large turbines, where the phenomenon is the same, and substantial cost savings result. Corresponding large savings in guideway cost can be expected. Final proof of the design must come from dynamic analysis of moving vehicles across flexible spans on straight and curved guideways. Some comments: "Depth" is apparently the same as "height" here. It could hardly mean anything else. The strict consequences of point 1 above would be that when the payload is zero, the weight of the guideway is zero. As the payload increases, the beam weight increases proportionately. Mathematically, this is possible. If there is no load, a weightless beam (non-existent) will provide adequate support for the non-existent load. If the load is 1 kilogram, perhaps a beam with steel 1 millimeter thick will do it. Given a very light beam, just sufficiently strong to support its own weight; as a payload of increasing weight is added, the redesigned beam weight correspondingly increases. These increases in beam weight are directly proportional to the added payload. This could perhaps mean that the total beam weight is proportional to the applied payload.

The motor vehicle traffic in urban areas in the developed world is expected to double during the 25-year period covering the years 1995 - 2020, according to statistics from OECD), if nothing happens that would alter today's trend. Thus, the car explosion is faster than the population explosion in these countries. This means that an equal amount of urban roads have to be build during these 25 years as are already in existence!. Since the demands for higher standards also increases (in the shape of more bridges and tunnels, as available land runs out), the demands for payment for these constructions increases even faster. The rough estimates above are also based on today's trends for urban living. If we assume that a considerable number of people move back to the country (fed up with urban life) or to smaller cities.		Then, the cost differences for the two scenarios above would not be so glaring. Let's assume that the city whose main commuting is based on road traffic (such as Los Angeles) can get by with investing only twice as much as the city whose commuting based on beam traffic, for the corresponding amount of transport work. Then, all roads and all road traffic, both the existing and the additional during this 25-year period, could be replaced by a sufficiently large beam traffic infrastructure for the same amount of money as would have been required to extend the road infrastructure as demand grows. On top of this, there would be the huge savings for coming generations in terms of labor. Ordinary motor cars don't last very long. They are not meant to last! If the carpark has to be renewed during the next 20 years, and they have to be supplied by tires and other spare parts during that time, this alone represents a huge expenditure that could be devoted to better things.

2. Analysis of Cost Factors

by dr. of Technology Sten Staxler evelopment costs for a new system such as FlyWay are of course high. They are also difficult to calculate. The Siemens SIPEM in Dortmund, Germany (which is a GRT system) required 35 million dollars between 1972 och 1984 for development and construction, when the H-Bahn (as the Germans prefer to call it) in Dortmund was developed and built. The Raytheon PRT 2000 required 50 million dollars between 1991 and 1997, when the test track trials were finished. At that point, the project was abandoned. The cost estimates get a bit higher when they are converted to monetary values of the year 2000 (because of inflation). Still, these development costs are of a similar size as for instance the construction costs for half a kilometer of a city highway with three lanes in each direction in Sweden. It should be noted, then, that there are no "development costs" involved; the technicalities of road construction are well known. The investment costs for a new system such as FlyWay are totally dominated by the infrastructure. The Swedish Communication Research Board (KFB, short for "Kommunikationsforskningsberedningen") has studied three PRT projects with a network length of about 30 kilometers. The infrastructure (tracks, stations, information centres, parking, maintenance shops, recycling centres) takes 87 % of the investment cost, and the beam vehicles themselves just 13 %. For roads of repeated type, the investment cost is the same for every new kilometer built. In Sweden the investment costs for city highways are about 80 million dollars per km. at the present (year 2003). This is an average figure, where there are some parts on concrete roads in the air (i.e. bridges), some parts on the ground, and some parts in tunnels. But with beam-carried GRTs and PRTs, the total investment cost is reduced for every new kilometer of runway, because larger production volumes reduce the cost of each beam section. KFB (mentioned above) has also studied 15 GRT systems in Europe, Japan and USA, with a network length from about 1 km, for 6 doublings of network lengths (1, 2, 4, 8 etc), up to and above 64 km. The longest GRT network length studied is SkyTrain in Vancouver, Canada, with about 70 km in length. The GRT investment cost per kilometer was found to be dropping from about 40 million dollars per kilometer for about 1 km length, down to about 8,5 million dollars per kilometer at 64 km. This indicates a 23 % decrease for each doubling of beam network length. An economic rule-of-thumb says that the decrease should be 15 - 20 % per doubling of the length of the series. The SIPEM network at Dortmund included in the study is only about 1 km. in length (none of the systems in the study are really "networks"). But the total investment cost was only 14 million dollars per kilometer, compared to the 40 millions for the mean value of the 15 GRT systems studied. This corresponds to only about 35 % of the mean value. Our interpretation at SwedeTrack is that is caused by the fact that the Dortmund system uses vehicles that are suspended below the steel beam, which thus can be more narrow. The other 15 systems in the study were mostly using heavy vehicles on top of a concrete guideway, with the same width as the vehicles. As noted above, a Swedish city highway costs 80 million dollars per kilometer. An average road investment in the Stockholm county, according to the Regional Planning and Traffic Authority (RTK), costs 14.5 million dollars.

Imagine that a FlyWay beam bus total investment cost for a first demonstration track of 1 kilometer would be 24 million dollar per kilometer, instead of the 14 million dollars above. Assume the same total investment cost for the first kilometer with larger beamcars to be 15 million dollars, and the same cost for smaller beamcars to be 9 million dollars. After 10 doublings of the network length up to 1024 km. with 20 % decrease in each doubling, these three costs should be about 7 % per kilometer, which comes to about 1.8 millions, 1 million and 0.65 million per km. respectively This is about 12 %, 7 % och 4 % respectively of an average road investment. And note that the road investmen costs do not include parking areas, road cars, traffic control centers, maintenance shops etc., which the total investment cost for the beam traffic does. These figures can be compared with the introductory factor of 4 (25 %), which up to the year 2050 should lead to 500 billions of dollars in savings. The operating costs for a road car can be estimated to about 3 dollars per 10 km, depending on how large and old it is. The petrol cost today can be about 0.7 dollars of that sum per 10 km. The cost of the driver, when driving 10 km at a speed of 30 km/h during 20 minutes in the course of the rush hour in the congested streets of a large city, with a human time value of say 10.5 dollars/hour, gives a cost for the driver of 3.5 dollars per 10 km. The total operating cost, including the driver, can then roughly be summed up to 6.5 dollars per 10 km. With 1.3 persons in the car during the rush hour, the road car operating cost will be around 0.5 dollars per person-kilometer. KFB (mentioned above) has also looked at the operating costs per person-kilometer for GRT systems (12 systems), Underground Rail (11 systems), and Light Rail (12 systems). The result comes to 0.098 dollars per person-kilometer for GRT systems, 0.126 dollars for Underground Rail, and 0.175 dollars for Light Rail. This means that the operating costs for GRT systems are about 5 times lower than for a road car, compared to Underground Rail about 4 times lower, and compared to Light Rail about 3 times lower. The differences seems mainly to be caused by the absence of a driver in the GRT systems. Also, the operating costs per person-kilometer for the GRT systems decrease rapidly with the number of person-kilometers, as the systems are used for a year. The costs decrease from about 0.8 dollars per person-kilometer down to about 0.08 dollars. These are figures from the Vancouver SkyTrain. This corresponds to a decrease with a factor of ten. Is it a good idea to cram many passengers into one beamcar? The calculated figures from the report show also, that the number of passengers per car in GRT systems, Underground Rail, and Light Rail in average is rather similar. GRT system cars contain about 16 passengers/cabin, Underground Rail about 20 passengers, and Light Rail about 22 passengers. The demand for capacity clearly varies in growing cities, from large in the city center, to rather high in the so called half central band, to modest in the suburbs (with city blocks consisting of flats), to low in the residential suburbs (where people live in their own houses). Travel demand also varies very much during the day, from the morning rush hour, to the middle of the night. The maximum traffic capacities of a 32-seat beambus, and a 16-seat beam-bus, clearly are 32 times and 16 times respectively higher than with a 1 seat beamcar, if the vehicles pass by a given point with the same time interval (say 1 second). But they cannot quite do that, even with a good automatic control system, because the vehicles have different lengths. And, at the close time-spacing between vehicles that can be achieved with computer-controlled traffic, these vehicle-lengths really do make a difference. If a 1-seat beamcar can pass a given point on the beam every 0.5 second, the 16-seat beambuses need to be interspersed about 1 second, and the 32-seat beambuses about 1.5 seconds (this depends on their speed, of course; you will find calculated tables on other web-pages).

This means that the maximum capacity of the 16-seat beambus is about 8 times larger, and the 32 seat beam bus about 11 times larger. In reality the vehicle velocity has a more complicated influence on the capacity result, as can be seen from our tables. The varying demand for capacity in different parts of the city during the rush hour on one hand, and the heavy part of the infrastructure in the investment costs on the other hand, makes it natural to adapt the infrastructure costs to the capacity demands. SwedeTrack has done this, with small beam cars, larger beam cars, and beam buses on different beams. Still, the smaller vehicles can use the more costly beams, especially outside the rush hours. The different dimensions and weights for the vehicles also gives different infrastructure costs to stations, poles, crossbars, pole feet, parking beams etc. But they can all be handled by the same industrial robots in production factories, maintenance workshops and recycling centres, or by the same installation machinery (which is used to erect the beams on site). Considering this, it is not a good idea to divide the GRT and PRT production on two different system producers and/or produce separate GRT and PRT systems. Outside rush hours, it is good to take the bigger vehicles out of traffic, to reduce operating costs. This is also an argument to use the same technology throughout the whole network, and not use two incompatible nets. We have mentioned that it is the difference between the operating income and the operating costs which preferably should pay all the fixed costs. Therefore it is important to increase the operating income, and this can be done in two ways. Firstly by designing the system so that it is able to handle the capacity demand even during eush hour. This means vehicle flexibility. Secondly to use the dominating infrastructure costs effectively, by also transporting goods, cars etc, outside the rush hours, and outside the working days, as SwedeTrack plans to do. We at SwedeTrack do not believe that the beamcars should carry standing passengers. In for instance SIPEM (in Dortmund), you have 16 seats in each cabin, with room for about 40 passengers in all. That means, that it is possible to increase the amount of passengers with a factor of 2.5 during high traffic times. But if a beam system carries only seated passengers, and these passengers also use seat bealts, you can increase the retardation from about 0.1 g to about 0.5 g. This brings four advantages that are readilly noticed: It means that the vehicles can pass approximately a factor of 5 more frequently along the line. In other words; they can travel closer together for a given speed, since they can stop faster in case of emergency, without risking injury to the passengers. In round figures you gain in line capacity a factor of 2. At the same time one would increase the comfort for the passengers, and this will in turn generate an increase in demand for transportation by the GRT. This would decrease the required station length and cost. This would also enable us to standardize all seat modules, the surrounding cabins, and the landing surfaces at the stations. It is thus a fallacy to believe that one could increase traffic capacity on a beam network with fast cars by allowing for standing passengers. It would work for slow cars (such as H-bahn in Dortmund), but why would anybody want to travel long distances in a slow vehicle? Only tourists might appreciate that. Apart from this, it is clear that the weight of the loaded vehicle will be larger the more people that are crammed into a car, which means that vehicles, beams and supports will have to be sturdier, and automatically more expensive to manufacture. It is perhaps difficult for a seasoned traffic planner to rid his mind of the idea that cramming people together in big vehicles will increase traffic capacity. But public transportation systems consisting of driverless automatic vehicles require another kind of thinking than do vehicles with human drivers. For more information about costs, read the article "Comparison of Costs between Bus, PRT, LRT and Metro/rail" written by Goran Tegner of the Swedish consulting firm Transek.

3. REQUIREMENTS TO ACHIEVE LOW COSTS

	1. THE WHOLE SYSTEM, THE WHOLE LIFETIME. Lowest use of resources, in terms of land, material, labor and energy, for a given amount of transport work. Attractive for motorists, aiming at creating largest possible demand. Negligible demands on the environment and negligible social consequences (such as accidents). Immunity against snow, ice, rain, wind, sand and sabotage. All parts flexibly pre-manufactured, having open interfaces and are easy to put together. Rational maintenance and low operating costs. All components are easy to move, re-use and reclaim.
	2. THE MAIN TRAFFIC ROUTES (This is the cost dominating factor). Cheapest possible pre-fabricated traffic conduits for vehicles with varying weights, in three size-groups (weighing from 1 to 11 tons). Traffic conduits jointly used by different kinds of vehicles, as long as they don't exceed the maximum allowed weight for the conduit in question. Lowest cost for construction over land and water, under land and water, and indoors. Every single new traffic route should pay its own way in revenues, with shortest possible travel time at an economical "break even". Quick installation and dismantling of the conduits. Extra incomes from such services as information transmission between vehicles and the surrounding world, that the travelers would be willing to pay for.
	3. THE VEHICLES. Mass production of vehicles and vehicle components, flexibly and in long series. Keeping the number of empty seats at a minimum in order to reduce the number of vehicles in use, and thus reduce operating costs. Shortest possible waiting time between vehicles that travel according to schedules. A sufficient range of different kinds of vehicles, which could carry both people, freight and ordinary motorcars. Quick transports, without crowding, congestion, noise, pollution and accidents, will contribute to the financing of the system. Free rides and vandalism are prevented with the aid of information technology, such as surveilance cameras and continuous position information.
	4. THE STATIONS. The stops placed at the side of the conduits, in order to maintain a high traffic flow. Distributed stops at short distances in between, in order to achieve shorter walking distances for travelers. No expensive and hard-to-reach stations above ground level. Outdoor stations should, as far as possible, be on the street level. No unnecessary costs for stops outdoors, apart from the sidings and for the call terminal in the pole (the terminal being used for calling a vehicle, like a taxicab). Several stops in parallel, should the need arise, to handle big passenger flows (or much freight). Parking for motor vehicles underneath the conduit, which will save on land and parking costs.

4. How Large-scale Production Affects the Price

Figure 4:1

Figure 4:1 shows estimated investment costs (in millions of $US) for beams, supports and other fixed equipment. Massproduction reduces cost per length in the manufacturing and also in the assembly process, out in the field. As a rule of thumb, costs are reduced between 15% and 20% for each doubling of the serial length. We have calculated for these two percentages, and compared costs between 3 sizes of beams. The starting points are costs for short networks for demonstrations purposes, and for short runs, such as between a city and an airport. As can be expected, the lightest beams come cheapest, but the difference between all these costs gets progressively smaller for larger production series. The reason is, of course, that total costs can be divided into material costs and production costs. While production costs per unit decreases for long manufacturing series, material cost per unit stays about the same. For long series, material costs thus become a progressively larger part of the unit cost. Percentage-wise, the extra material that goes into sturdier beams and supports is small, and therefore the curves for the three beam-categories tend to approach each other for longer production series.