vehicle velocities. One method which is particularly important to investigate is an adaptive system in which the characteristics of the suspension are varied in accordance with the type of disturbance encountered.

As speed increases the demands on suspensions of wheeled vehicles become severe. Research is needed into methods of aerodynamically supporting the vehicle as a means of reducing the demands on the suspensions. One approach which can be studied is the use of the aerodynamic lift provided by ram air at speeds above 200 m.p.h. Another approach which has an almost infinite number of variations is an air cushion vehicle where levitation is achieved by means of a layer of air under pressure beneath the vehicle. The layer of air can range from several feet in thickness to a few thousandths of an inch, depending upon the system used. Means of controlling the layer and pressure include plenum chambers, skirts, perforated plate, flexible diaphragms, and accurately machined ports on accurately machined guide surfaces. The research effort needed to study the feasibility of all these concepts and to provide information as to which if any provide the most suitable answers to the HSGT system requirements is an immense one.

The operation of a HSGT system at speeds of 200 m.p.h. or more implies a well protected, alined and maintained guideway and infra-structure with the result that the guideway is potentially the most expensive single subsystem of virtually any HSGT system. A considerable portion of this expense is inversely proportional to the allowable guideway alinement. A part of the research effort on any suspension or levitation system must provide answers on these guideway requirements.

The power requirements placed on the propulsion subsystem also must be determined in research on levitation methods or adaptive suspensions. Tradeoff studies between levitation, suspensions, propulsion, and guideway design are needed before selections of the overall HSGT system can be made.

5. Guideway structures.-The guideway will inevitably undergo displacements from supporting structure deflections and, also, movements in the rock or earth supporting the foundations. A knowledge is required of the guideway movements, arising from such causes as live load, wind, and temperature. Both the minimum movement which may be achieved and the cost of approaching it are Important factors.

Research is required to support economic analysis of the various methods of supporting the guideway-hard rock tunnel, soft earth tunnel, earthfill under a surface structure, deep foundation under elevated structures. The research must provide information to balance the cost of construction of the guideway structure against the cost of designing, building, and maintaining a vehicle and a vehicle guideway interface which can tolerate the movements expected for the guideway structure. Tradeoffs between the vehicle and the guideway must be studied thoroughly. Since the cost of the guideway is many times as much as the total cost of all the vehicles, a vehicle which can operate on a less sophisticated, cheaper guideway will result in major investment and operating cost reductions, even though the cost of individual vehicles may be increased significantly.

6. Human factors.-The translation of human needs and constraints into technological specifications has precedent in previous engineering of aircraft and air terminals and to a lesser extent in other vehicular systems. Human factors are construed to include (a) comfort and safety; and (b) the roles of operating personnel.

Constraints on system component designs relative to acceleration, vibration, acoustic noise, illumination, and air pollution need to be compiled.

Terminal facilities and procedures for entry (reservations, ticketing, baggage, etc.) and exit need to be designed.

7. Computer control.-The objective of research in control by central computers is to determine feasible schemes for sensing, processing, and evaluating all information needed to control traffic automatically in the network and to sustain the required level of performance established for the network.

Decisions on scheduling (based, for example, on predicted or predetermined passenger demand), dispatching, and vehicle speed control and regulation can be made either at a central computer facility serving the entire network or at several regional facilities. It is possible that the final transport system will contain both central and local information processing and decisionmaking centers. In this case, it is necessary to evaluate cost and reliability factors associated with the degree of central computer control responsibility.

8. Dynamics and control of vehicle groups. The research outlined is concerned with the determination of several optimal control schemes for the velocity control of each individual vehicle within a string of high-speed vehicles. This research is necessary in order to establish

(a) The optimum spacing of the vehicles as a function of the desired average velocity;

(b) The transient decleration and acceleration of each vehicle whenever switching and injection of vehicles takes place;

(c) The speed control associated with the rendezvous process of a through vehicle with a local one;

(d) Bunching problems that may arise as a result of system disturbances and their effect upon the system operation; and

(e) Sensitivity characteristics of the system whenever unpredictable disturbances occur.

9. Communications.-Research on the role of communications in a HSGT system has two principal aspects:

(a) The determination of the nature and amount of information content required in the system in order to achieve the operating performance characteristics specified for the overall system; and

(b) The determination of the form and organization of communication links required among the various elements within the transport network and between the network and the users it serves.

The control of traffic in the HSGT system requires the sensing of information concerning the state of the system, transmission of this information to appropriate points, and the processing of the information for presentation to the controlling and decisionmaking devices. The quantity of information to be handled, and its form, are determined to a great extent by the degree of optimality required in system performance.

Purpose

II. DEMONSTRATION PROGRAM

DEMONSTRATIONS

An essential part of the program to explore the potential of fixed-path/highspeed ground transportation utilizing special-purpose route facilities is to test the market as inexpensively as possible, before large sums are committed for new or improved facilities.

By measuring public response to varying levels and combinations of speed, cost, comfort, and convenience, obtainable, without delay, at low cost, the demonstrations will help to indicate the economic prospects of long-term investment in improved ground transportation.

Since the only guided surface transportation using special-purpose fixed facilities now operative on a commercial basis is railroads, the project will utilize existing rail routes in the northeast corridor for the controlled experiment.

Providing improved passenger service utilizing the best of present technology, will (1) test public reaction to measurable improvement in present techniques and (2) provide a basis for statistical projection of probable response to still further improvements.

It is desirable to make projections before commitments are made to build systems whose costs may range between $750 million and $3 billion in the northeast corridor. For a total expenditure by the Federal Government and by interested private groups of less than $50 million, rail service in terms of comfort and convenience can be greatly improved, and elapsed time in route reduced considerably. This should give a good indication of the results of more farreaching improvements.

Although the demonstration will use selected rail routes along the northeast corridor, its results are expected to be useful to many other parts of the Nation such as: (a) Seattle-Tacoma-Portland; (b) the east coast of Florida; (c) Milwaukee-Chicago-South Bend-Cleveland; and (d) San Francisco-Los Angeles. It must be emphasized that the demonstrations are tests. They are not intended to be long-term commitments of the Federal Government to provide intercity rail passenger service.

Federal funds would be employed exclusively to defray those costs of equipment required to attain the standards of service prescribed by the Government as essential for a valid test of public response. The railroads would receive no assistance in sustaining the present level of passenger service or standards of

plant and equipment. The demonstration would have no effect on their present eircumstances with regard to tax accruals, access to capital, public obligations, etc. Increased revenue resulting from the improvements incorporated in the demonstration would be used to reduce the Government's contribution to increased expenses.

transportation is implied.

The $8 million now sought as the public contribution to the incremental cost of the demonstration decidedly is a proper responsibility of the Federal Government. The probe is designed to produce answers which Government must have for longterm transportation planning-not only with the corridor, but also elsewhere in the Nation. In any case, no commitment to a future development of railroad Investigations by the Department of Commerce indicate that improved railroad service in the corridor shows promise of potential usefulness. At the same time, alternative forms of improved passenger transportation are in prospect. Before committing any party in interest to an estimated investment of perhaps as much as $2 billion in order to provide even an attractive rail-type service, using facilities, it is essential that public reaction to demonstrations of selected improvements be obtained. Only in this way can we assess the probable public patronage of any form of improved ground transportation-including a highspeed railroad itself.

Secondly, the economic cost of varying levels of service standards applied to railroad transportation must be compared with alternative forms of surface transportation measured against a similar yardstick.

The cost of the demonstration projects to the Government will not exceed $8 million in the first year of the 3-year experiment. Of this sum, about $200,000 represents the expense of developing and implementing wholly new techniques of measuring and analyzing trabel demand response factors and trends on a particular travel route.

The remainder, $7.8 million, would constitute the national contribution to increased investment required to provide appropriate test conditions (ie., improvements in railroad passenger service quality) along the corridor rail routes which have been selected for this purpose.

In drawing plans for improvements prescribed for the demonstration, the Department of Commerce has exaluated carefully all relevant developments in railroad passenger service since the end of World War II. Included was a survey of the limited number of demonstration projects involving railroad commuter service which were sponsored by the Housing and Home Finance Agency under the authority of the Housing Act of 1961.

The Department has also reviewed experiments by the railroads—and their suppliers with innovations in equipment, schedules and fares, most of which occurred in the immediate postwar years, in an effort to arrest the decline in passenger revenues and reduce their deficits.

The new Tokaido Line in Japan and the Trans-Europ Expresses and other modern European trains have received attention. Such foreign innovations as appear appropriate to American conditions and feasible in the limited demonstration project will be given serious consideration.

Washington-New York Line

The key items in the demonstration improvement, funded jointly by the Federal Government and the railroad, are expected to be:

Acquisition of a number of new high-speed, stainless steel, self-propelled electric "multiple-unit" passenger cars operable in trains without locomotives;

Upgrading of electric current catenary structure as required;

Revision of signal system where necessary to permit high-speed operation; More intensive improvement of a selected stretch of track between Trenton, N.J., and New Brunswick, to permit tests of still higher speeds.

In addition to these items, it is hoped that other improvements could be included in the demonstration program. Some of the new improvements which have been suggested as desirable are:

New suburban stations with ample parking and convenient highway access;

Improved baggage handling;

Car level platforms at Wilmington, Baltimore, and Washington;
Automatic protection at all grade crossings;

Improved ticketing procedures.

More extensive changes, curve reductions for example, are not contemplated.

With intensive utilization at the higher speeds which engineering studies show to be feasible, the new equipment could add a substantial number of runs to the present frequency of through Washington-New York trains, and without disturbing the present through service between Washington and Boston and between New York and the South and West. Such a large number of schedules would provide frequent, regular service to both terminal and intermediate points. The new cars would be scheduled, where possible, to provide runs of less than 3 hours elapsed time between the two cities (compared with present best schedule of 3 hours, 35 minutes) at hours of peak demand for through travel. The added service, of course, increases peak period capacity.

Contemplated also is the adjustment of schedules provided by existing locomotive-hauled passenger trains on the New York-Washington route for optimum blend with the additional runs to be performed by the new cars. Included in the plan is operation of a train of conventional equipment on a new fast schedule between New York and Philadelphia at the peak business travel period, morning and late afternoon.

Specifications for the new cars now being prepared will call generally for faster acceleration and braking than is currently available in railroad train operation, as well as for a sustained running speed of 150 miles per hour. Among the advantages of self-propelled cars, compared with locomtive operations, are: 1. Quick turnaround and flexibility of train makeup without terminal switching;

2. Lower rolling weight stress on bridges;

3. Faster, smoother acceleration and braking.

Design of these cars is basically similar to that of equipment recently acqiured for suburban railroad service in the Philadelphia area. Demonstration specifications, however, call for more spacious seating and refinement in seating, lighting, decor, baggage accommodation, springing, etc., adapting the units to longer haul, intercity operations. Facilities for a light food service will be provided, obviating the need for food vendors. There will be parlor car as well as coach accommodations. It is hoped that the new cars will be at least as comfortable at speeds over 100 miles per hour as the best present cars at 80 miles per hour. In addition to improvements in overall comfort and decor, attention will be paid to such details as air-operated, easy-to-open doors and ashtrays at each seat in smoking sections of the cars.

New York-Boston Line

In view of the serious physical obstacles to higher sustained speeds along much of the so-called shore line between New York and Boston, the initial demonstration project is expected to confine physical improvements to the relatively straight and well-graded stretch of the New Haven's main line between Boston, Mass., and Providence, R.I. On this still heavily patronized segment of the corridor route, there is tentatively planned a substantial increase in speed, frequency, and comfort, using a gas-turbine, self-propelled, lightweight train now under development by a leading carbuilder.

The only portion of the New Haven Railroad which is electrified and can be used by the new electrically powered cars to be provided under the demonstration project is the densely traveled four-track section between New York and New Haven.

The route taken by all of the railroad's suburban, and most of its long haul, trains utilizes the tracks of the New York Central to gain access to Grand Central Terminal. The Central uses an electric power system which differs from that of the New Haven's both in type of current and in method of current collection, and New Haven locomotives and multiple-unit cars serving this route must be specially and expensively equipped to adapt to both systems.

Hence operation of demonstration equipment likely will be confined to service in and out of Pennsylvania Station, New York. This could take the form of through service between New Haven, Conn., and Pennsylvania Railroad points or additional service between New Haven and New York via the Hell Gate Bridge route, with convenient transfer to Pennsylvania Railroad trains.

The demonstration project will, however, make full ticket collection and other statistical studies of the entire New York-Boston main line of the New Haven for future use. Since a portion of intercity trains on the corridor rail route run through between Boston and Washington, utilizing equipment of both railroads, it is necessary to study patronage reaction of the through corridor route as a unit, as well as the separate sections southwest and northeast of New York City.

Experimental design

Since the purpose of the demonstration project is to measure as precisely as possible public reaction to specific ingredients of service betterment, it is essential that the experiment be designed, or shaped, in advance, in detail-to produce the largest possible store of usable information, with maximum pinpointing of the individual factors of improvement.

Only a relatively short span of time has been allotted to the project; hence, a sequential type of data analysis probably will be necessary, so that data can be examined at relatively short intervals, as a basis of decision to change or continue the relative weights of individual items of service change.

An efficient sequential design will require continuous and current availability of detailed results. Careful attention must be given to the degree of sensitivity in the data collection system.

Were there unlimited time for the experiment, quite possibly the key factors of service improvement could be put into effect one at a time, and each additional factor isolated from those preceding. Obviously, since this is impossible, combinations of factors will have to be dealt with.

Fortunately, by applying and appraising varying combinations of factors on individual trains and on separate segments of the routes involved, it should be possible to obtain refinement in identification not possible with their uniform application.

Second, modern techniques of differential statistical analysis will further break down the raw data findings to isolate specific items of influence.

To insure high sensitivity, the data-collection system will be carefully planned to give an orderly, progressive sequence of events which will measure market response to a wide range of changes in speed, schedule, frequency, and convenience and fares.

The specific pattern of changes in these factors which should be made during the demonstration period will depend in large part upon information developed in the base year prior to the beginning of actual demonstration.

This information will be obtained from sample surveys of all types of passengers, both rail and nonrail, and from more intensive flow data collected from rail passengers alone. The survey data will include information on reasons for choosing present mode of travel and preference for various levels of change in speed, schedules, and other factors possible in rail travel. The flow data will show station-to-station movement of rail passengers by type of ticket, time of day, and day of week, as well as seasonally. The indicated preferences and observed flow pattern will be used to design the optimum combination of improvements possible within the capability of the demonstration. The experiment will include meaningful deviations from this optimum in such a manner as to permit measurement of the differential effects of the important factors within the time permitted.

Accurate, detailed and current passenger statistics will be collected from both the Pennsylvania Railroad and the New Haven Railroad on trips between Washington and New York and Boston. These data will be collected from July 1, 1965, to provide a normal base year for the demonstration period starting July 1, 1966, and through the demonstration period, which will probably end June 30, 1967, Each passenger moving on through trains will be counted and classified according to: (1) City of origin and destination; (1) type of ticket (e.g. coach or pullman); (3) date of trip; and (4) train number (time of day).

These data will be collected continuously by the conductors on each train and will be reported at the end of each run. The data will then be summarized according to the four classifications shown above. These classifications will permit tabulations showing the flow between each pair of the following cities: Washington, Baltimore, Wilmington, Philadelphia, Trenton, Newark, New York, Stamford, Bridgeport, New Haven, New London, Providence, and Boston.

These flows can be summarized, in total, for any period of days, or hour of the day, and by type of ticket. The detail in which the basic data are collected will also permit the development of distribution of traffic showing peakloads as well as averages and the effect of shifting schedules.

In addition to the traffic flow data, there will be a program of on-train questioning to develop patron reaction to changes in service and sample surveys to determine the effect of such changes in rail service on nonrail passengers. These sample surveys will also measure the effectiveness of advertising which will be needed to acquaint people with the demonstration program.