America's Highways 1776–1976: A History of the Federal-Aid Program/Part 2/Chapter 4

From Wikisource
Jump to navigation Jump to search
Part Two Chapter Four
Research

Research was one of the principal missions of the first national highway program in the United States and is, in fact, the oldest continuous Federal highway activity. The scope of federally conducted or assisted research encompasses all of the Federal highway programs and every facet of highway transportation—as a national resource and as a principal force in the social and economic well-being of the people and Government.

Early History

The history of highway research in the Federal Government began with the establishment of the Office of Road Inquiry (ORI) in the Department of Agriculture in 1893. Prior to that date, there had been numerous investigations and experiments, but they were primarily scattered and isolated events. With the creation of the ORI, whose primary mission was to investigate the best methods of roadmaking and to assist in disseminating this information, a formal, organized research program began. In 1894, the ORI issued 9 bulletins on such subjects as State laws, roadbuilding materials, and railroad rates for hauling those materials.

Demonstration trains, originally known as “Good Roads Trains,” traveled throughout many parts of the Nation in the 1890’s. These trains were fitted with construction and roadbuilding machinery and equipment as well as section models of macadam and other types of road construction.

States, local authorities, railroad companies and the manufacturers of earth-handling and roadbuilding machinery also cooperated in building short sections of quality roads to demonstrate good roadbuilding practices. The emphasis was on drainage, surface courses, and maintenance. Local materials and manpower were used under the general supervision of Federal engineers. Sometimes on “Good Roads Day,” as many as 500 local farmers would take a walking tour of a demonstration section being constructed. In 1910, the annual report of the Office of Public Roads reported that 55 object lesson and experimental roads had been completed during the fiscal year.[1]

Although information on specific research activities in the early 1900’s is rather fragmentary, the most substantial continuing effort was testing large numbers of samples of highway materials, including aggregate, cement, soil, asphalt, and tar. A Federal laboratory was established in 1900 for mechanical and chemical investigation of these materials. The results of the tests were analyzed to obtain a general overview of the characteristics of available materials and their suitability for roadbuilding purposes. Samples were sent in from all areas of the country and were tested free of charge until 1924, when it was announced that materials would no longer be tested for the general public. In the early years, the laboratory developed important tests to help improve bituminous construction.

In addition, statistical and economic evaluations were made in 1910 of the effect of road improvements on communities. Some work was also undertaken on coatings and coverings for iron and steel corrosion. In 1911 there were some object lesson projects on culverts, and a bulletin was published on highway bridges.

From the beginning, the Office of Public Roads included a group of men to develop new techniques through experiment and research. Historically, this activity has involved a strong element of “let’s try it and find out” as illustrated in this early bituminous macadam pavement experiment. Today’s broad program of research and development dealing with problems of safety, traffic management, environmental protection, materials, structures, and maintenance uses modern, scientific techniques in searching for solutions.

Rather than attempting to outline here a large number of technical advances, several broad historical milestones can be summarized. (1) The research mission of the Federal agency was clearly established, and, later, Federal funds were specifically earmarked for research. (2) A major research journal, Public Roads, was founded. (3) The cooperative and joint-participation concept with the States and industry was established. (4) The initial concentration on material and surfaces was rapidly expanded into areas such as structures, economic consequences, and the impact of traffic and trucks.

Federal Aid for Research

Federal Administrative Resources

The first sustained fiscal support for highway research was authorized by section 21 of the Federal Highway Act of 1921. The foundation for the Federal-aid State highway planning and research program was laid with the enactment of the Hayden-Cartwright Act of 1934.

Section 11 of the Hayden-Cartwright Act specified “With the approval of the Secretary of Agriculture, not to exceed 1½ per centum of the amount apportioned for any year to any State . . . may be used for surveys, plans, and engineering investigations. . . .” Under this authorization, some States, with the delegated approval of the Bureau of Public Boads, initiated research activities, mostly in the area of physical materials of the highway.

The Federal-Aid Highway Act of 1944 contained for the first time the term “research” in addition to planning. Thus, the States at their option and with the approval of Public Roads could use a portion of the 1½ percent planning funds for research. Funds not used for planning or research could revert to the construction program. With the enactment of the Federal-Aid Highway Act of 1962, the 1½ percent funds were restricted as of fiscal year 1964 to research and planning purposes only. If they were not used during their availability period, these funds would lapse.

To further encourage the States to increase their research and planning efforts, the 1962 Highway Act authorized, beginning with fiscal year 1964, the use of an additional one-half of 1 percent of sums apportioned for each fiscal year for planning and research, but this was optional. The initiative to conduct such efforts rests with the State highway departments.

Prevost Hubbard

Prevost Hubbard
Prevost Hubbard

There were many professional men who made major contributions to the development of America’s roads. One such man was a scientist named Prevost Hubbard. Between 1905 and 1919 he developed for the Office of Public Roads methods for testing and specifying bituminous pavement materials which eventually became standards of the highway construction industry and the foundation of modern materials technology.

Hubbard, born in 1881, was educated at George Washington University in Washington, D.C. He joined OPR in 1905 as the agency’s first laboratory chemist to help sort out and develop standards for the wide variety of petroleum based products then being experimented with as surfacing agents for highways.

He became active in the American Society for Testing Materials and spearheaded, with Logan Waller Page, Committee H (now Committee D-4 on Road and Paving Materials). He served as secretary of the Committee from 1909 to 1946.

In 1919 Hubbard joined the Asphalt Institute, first known as the Asphalt Association, to direct a research and development program sponsored by the petroleum industry. The Institute’s laboratory had its beginning in the basement of Hubbard’s home in White Plains, New York. One of the first accomplishments under his leadership was the development of Medium Curing Cutback Asphalt, a liquid form of asphalt, later used extensively in the construction of low-cost, low-volume, all-weather pavements.

During the early 1920’s an intensive effort was made to reduce the number of specifications in effect in the United States. The asphalt industry, under the direction of Hubbard, in cooperation with the States and Bureau of Public Roads, was successful in reducing the grades of asphalt cement in use from 88 to 9 and asphalt joint sealers from 14 to 4. Similarly, Hubbard’s cooperative efforts during the 1930’s led to the successful reduction of the number of grades of liquid asphalts used by 33 States from 125 to 18.

The latter study included an extensive testing program of most liquid asphalts produced in the United States. J. T. Pauls of the Bureau of Public Roads commented on the study in a progress report in 1932:

Representatives of the Bureau of Public Roads and Mr. Hubbard of the Asphalt [Institute] stressed the fact that they believed in the adequacy of the proposed cooperative tests for the testing of liquid asphaltic products and felt that many tests now being used by the States were unnecessary and unsatisfactory.

Hubbard went on to develop methods for testing the design of paving mixtures. With the assistance of F. C. Field of the Asphalt Institute, Hubbard invented the Hubbard-Field Stability Test, which became the first widely accepted test to measure the strength of asphalt paving mixtures.

He died in 1971 at the age of 90. A year later Committee D-4 on Road and Paving Materials established the “Prevost Hubbard Award” in recognition of his outstanding service to the Committee and to the field of bituminous road and paving materials.

This is a 1915 traction dynamometer mounted on a wagon, used to measure the force required to pull the vehicle over the road against the friction of the wheels in contact with the road.

The Federal-Aid Highway Amendments Act of 1963 expanded the law to include development under the research and planning section. The Act specified that the 1½ percent funds would be available, among other purposes, “. . . for research and development, necessary in connection with the planning, design, construction, and maintenance of highways and highway systems. . . .” The intent of Congress was that development would be an integral part of the overall research and development program and that this provision would stimulate the States to play a more active role in the development phase.[2]

Public Roads Magazine

The Federal Aid Road Act of 1916 authorized a 5-year road program which had barely begun when the United States entered the European war. Looking forward to the resumption of the road program after the war, Director Logan W. Page of the Office of Public Roads and Rural Engineering (OPRRE) foresaw the need for a journal devoted to the publication “. . . of the results of researches, experiments and studies of those connected with this Office, and of highway officials of the various States . . .”; and also “. . . for the dissemination of such information as the officials of the various States may desire to spread for the benefit of their contemporaries.”[3]

The first issue of the new publication, named Public Roads, appeared May 1918. It provided the State highway officials with a welcome forum for the discussion of current problems. The first issue brought the industry up-to-date by summarizing motor vehicle licensing laws and fees for registration and operators’ licenses. This wartime issue also urged highway builders to conserve scarce fuel by proper attention to the firing of boilers and the careful use of steam in road machines and in quarrying. An entire issue (June 1918) was devoted to the catastrophic road breakups caused by heavy trucking during the 1918 spring thaw. The May 1919 issue dealt with the social and economic benefits of using convict labor on the public roads. When the Government distributed the huge surpluses of military equipment to the States, Public Roads ran articles on how to take care of the equipment and convert it to civilian highway use.

Public Roads published the resolutions adopted by the American Association of State Highway Officials (AASHO) at its annual meetings of December 1918, 1919, and 1920, and also the papers read at those conventions. In effect, it was the official journal of AASHO until that organization launched its own publication, American Highways, in 1922.

Within a year of its first issue, Public Roads was an important voice of the young highway industry, with a long waiting list of would-be subscribers. In fiscal year 1920, the authorized monthly circulation was raised to 4,500 copies, but hundreds of requests for the magazine had to be refused. Budgetary cuts reduced the circulation to 4,000 copies per month for fiscal year 1921, and, without explanation, publication was suspended altogether after the December 1921 issue. The suspension drew an immediate protest from the American Road Builders’ Association, AASHO, and other organizations interested in roads and also “. . . many expressions of regret not only from its engineer subscribers, but also from the nontechnical administrative heads of county highway activities to whom it had been helpful. Not the least gratifying of such expressions were those which came entirely without solicitation from the editors of other technical engineering journals.”[4]

Public Roads resumed publication in March 1924, with the return of better times. However, the magazine was no longer a forum for the administrative and technical problems of the States, this function having been assumed by American Highways after Public Roads ceased publication. Instead, the new Public Roads was exclusively a house research journal, and all of its contributors were engineers, scientists, and economists of the Bureau of Public Roads. As the Bureau’s research activities expanded, Public Roads published papers dealing with every aspect of highway research—finance and taxation, the economics of transport systems, the properties of soil and road materials, the management of construction operations by contractors, the characteristics of highway traffic, the strength of road slabs, and many others.

Public Roads was the original publisher of many landmark papers in highway research. Most notable of these was “Highway Capacity: Practical Applications of Research” by O. K. Normann and W. P. Walker (Public Roads, October, December 1949). Another paper, “Interrelationship of Load, Road and Subgrade” by C. Hogentogler and C. Terzaghi (Public Roads, May 1929) laid the foundations of subgrade soil classification and marked a turning point in studies of subgrade soils.

The highway researchers of the twenties, thirties, and forties were breaking new ground. Often progress in a particular field of research depended on the invention of new instruments to measure what had never been measured before. A continuous stream of such instruments issued from the Bureau of Public Roads’ instrument laboratory for nearly 40 years—the Goldbeck Pressure Cell for measuring pressures under pavements; the electric-eye and road-tube traffic counters; the Benkelman Beam for measuring minute deflections in pavements under load; and many others. Information about most of these devices first reached the scientific world through the pages of Public Roads.

Publication has continued without interruption from March 1924 down to the present, although the frequency of issues has varied widely. Through the years, Public Roads again expanded to include articles on highway research and development from sources outside of the Bureau of Public Roads. Throughout its long history, Public Roads has maintained a high standard of scientific accuracy and literary clarity and, taken as a whole, is a remarkable chronology of the development of highway engineering and economics in the motor age.

Federal–State–Industry Cooperation

The historic association between Federal highway researchers and their counterparts in the States is considered unique in Federal Government programs, past and present. Over the years, all of the States have conducted research in cooperation with Public Roads and contributed in countless ways to the program described here. One other important accomplishment in this longstanding Federal-State partnership has been the establishment of strong, viable research staffs, facilities, and programs in the States.

Another successful research relationship was established between Federal and State highway agencies and interested outside groups, including other Government agencies, the academic community and associations, national organizations, professional societies, and industry groups. These various bodies have aided in identifying needs and in planning research to fill those needs. They have assisted in the conduct of studies with time, manpower, funds, and consultation and have taken an integral part in the development and implementation of results.

One organization which has played a major role in coordinating modern highway research and disseminating the results is the Transportation Research Board (TRB), or, as it was known for half a century, the Highway Research Board (HRB). It was organized in 1920 as an agency of the National Research Council in the National Academy of Sciences. Its purpose was to “ ‘assist in outlining a comprehensive national program of highway research and coordinating activities thereunder; organize committees for specific problems; deal with ways and means; and act in a general advisory capacity.’ ”[5] In recognition of the increasing emphasis on the “systems” or balanced approach to transportation problems, the TRB in recent years has modified its scope to include the development of other modes of transportation as they interact with highways. A more detailed discussion of the TRB and its relationship to the Federal Highway Administration research efforts can be found in Chapter 1 of Part II.

The Maryland State Roads Commission testing laboratory in 1929.

In the last two decades, considerable cooperative research has also been undertaken on an international scale. A notable example is the continued involvement of Public Roads in a number of research committees of the Organization for Economic Cooperation and Development (OECD). Their efforts have been particularly directed toward research for safer highway design and improved traffic operations. Since 1964 Public Roads has also worked with the International Road Federation in the collection and dissemination of information through a worldwide annual inventory of research and development activities.

Construction Materials and Structures

In the early period of highway construction in this country, an understanding of the physical behavior of the major materials was developed primarily by trial and error, which led to the development of criteria and tests based, to a great extent, on empirical relationships. Much of the early research work consisted of testing large numbers of samples of highway materials to determine the essential characteristics of available materials and their suitability for roadbuilding purposes. Developments with different materials varied somewhat in time, but in general during the 1920’s and early 1930’s, the technology with respect to soils, asphalts and other bituminous materials, cement and concrete, and pavements, bridges, and other structures was developed.

Taking soil samples for field subgrade soil studies.

Soils

In the very early stages of the roadbuilding industry, little of a scientific nature was known about soils as road materials other than that clay soils were sticky and sadly lacking in vehicle support when wet; and that very sandy soils, while having fair performance when wet, were highly unsatisfactory when dry. Thus, the first effective modification of soils for unpaved roads consisted of adding sand to clayed soils and clay to excessively sandy soils in a more or less “cut-and-try” system. However, the variable results obtained by casual blending soon made it apparent that more than eyeball engineering was needed in dealing with soils and developing their potential as road material.

Most of the experiments up until 1920 were concerned with dust prevention and road surface preservation. Then, in 1920, the Bureau of Public Roads began investigations “to obtain accurate scientific information regarding the characteristics of soils which affect their bearing value.”[6] At about the same time, several State highway departments established soil-testing units and each attacked the problem in its own way. However, through exchange of information at conferences and through American Society for Test- ing and Materials (ASTM), AASHO, and HRB, a soil classification system and standard soils tests were developed.

Soil Constants

One significant milestone was the development, during the latter part of the 1920’s, of tests to measure certain key characteristics of the various soil types and their blends. These were the so-called soil constants which included the liquid limit, plastic limit, and plasticity index, the latter being calculated from the other two.[7] These constants, which identify the moisture retention and flow characteristics of a particular soil, have become indispensable criteria for evaluating and predicting the performance of a given soil as a pavement foundation and for use as a control tool to facilitate the blending of inherently unsatisfactory soils to produce acceptable foundations.

Research in the late 1920’s also disclosed that other soil properties, in addition to plasticity, affect the performance of soils in road foundations. It became apparent that the wide range of soil types should be classified on the basis of measurable characteristics, such as fineness and chemical composition, to provide highway engineers a working soil language. Based on data obtained by testing thousands of soil samples and observation of behavior of soils under field conditions, the Public Roads’ soil classification system was developed and first published in 1929. The best soil for use in road foundations and earth structures was classified A-1. The series continued through seven more main groups, the poorest being A-8. This system, with appropriate revisions, later became the ASTM and AASHO standards.

This nuclear moisture-density gage was developed in the late 1950’s to instantly test subgrade compaction without the need to take samples and complete a laboratory analysis.


Continuing soils research included studies of resistance to frost heaving in cold climates (frost susceptibility) and consolidation or settlement characteristics of soils. It was determined that frost heave was greatest in silty soils and that the best remedy was to remove and replace the silt with coarser grained materials, such as gravel, to a depth at which frost action was not detrimental. Under the guidance of Dr. Charles Terzaghi of the Massachusetts Institute of Technology, serving as consultant to Public Roads, apparatus for measuring the consolidation characteristics of soils was also developed in the late 1920’s, and the test data were used to determine the rate and amount of settlement of highly compressible soils.

Occasional highway surface failures where the quality of the foundation soil was considered to be good indicated insufficient compaction during construction. To provide adequate compaction control, a test method for determining the moisture-density relations of soils for earth dams was adapted to highway earthwork in the mid-1930’s. Thereafter, compaction specifications for highway embankments, subgrades and soil-aggregate base courses required that the material have a proper amount of moisture when compacted so that a target density could be reached.

This is a circa 1935 electrical resistivity test along the George Washington Memorial Parkway in Virginia. An electric current is passed through steel rods to measure ground resistance. This geophysical method is used to locate subsurface rock formations.

Soil Stabilization

Significant progress was made in soil stabilization during the 1930’s. Research showed that soils for use in pavement subgrades could be improved by the addition of portland cement, lime or bitumen, and that calcium and sodium chlorides were effective dust palliatives. During World War II, the Soils Laboratory collaborated with the Engineer Board, Fort Belvoir, Virginia, in the evaluation of chemicals for soil stabilization. Beginning in 1954, and continuing until about 1970, the Bureau in cooperation with the chemical industry evaluated about 50 chemicals for soil stabilization. A few were found to be marginally useful, but none were economically competitive with Portland cement or lime for subgrades.

Remote Sensing

Throughout modern highway history, there has been a continuing search for a rapid, effective way to explore conditions on or beneath the ground without actual excavation or ground surveys. Beginning about 1932, Public Roads promoted the development and use of remote sensing methods for obtaining information on soil and ground conditions in place. The “cumulative-curve” method of presenting electrical resistivity data was developed and adapted to determine depth to bedrock, delineate sand-gravel deposits and thickness of portland cement concrete pavement. Also, aerial photographic interpretation principles were adapted to engineering-soil and materials mapping. Cooperative work between BPR field offices and other agencies (U.S. Geological Survey, National Aeronautics and Space Administration (NASA), the University of Michigan and Purdue University) in 1960-1967 demonstrated that color aerial photography was a good way to evaluate soil and terrain. Widespread use of these methods has followed. The development and evaluation of ground and aerial remote sensing methods for obtaining information on terrain and environmental features continued into the 1970’s.

Bituminous Materials

Over the years, Public Roads has played a predominant role in the development of asphalt technology and test methods for bituminous materials. In 1903 it cooperated with the American Society for Testing Materials in organizing a committee on road and paving materials to develop essential standard test methods and material specifications.

The BPR tried various tests on bituminous materials to determine their physical and chemical properties and their suitability for use in road construction. Conclusions on which properties were significant in relation to performance and on the best methods for testing materials were disseminated in a series of Department of Agriculture bulletins, beginning in 1911, which included the methods in use by the Office of Public Roads at that time, and provided a progression of guidelines for procedures in the use of bituminous materials. Many of the early ASTM and AASHO standards for material and test method specifications, which are used in highway construction, were derived from this series. The latest edition, now called Standard Specifications and Methods of Tests for Highway Materials, was published by AASHO in 1974.

Beginning in the 1920’s and continuing through the mid-1960’s, most of the research effort in the bituminous area was concerned with studies of properties of asphalts and tars produced in the United States, and the relation of such properties to the performance of pavements. Often such studies resulted in the adoption of new specification requirements.

One of the most significant developments was the simplification of specifications for liquid asphalts and asphalt cements. A survey of penetration grade asphalts in 1923 showed that 88 different specifications for these materials were being used in the United States. A joint conference of representatives of Public Roads and users and producers of asphalt recommended that only nine grades were sufficient to provide the necessary materials. Those nine grades were adopted by national groups and most of the States. Subsequently, the number of grades has been reduced to five.

In 1930 a similar survey showed that there were 119 different tests, including all variations, being used in specifications for acceptable grades of liquid asphaltic materials in the United States.[8] After a cooperative study of the materials being produced and a series of conferences throughout the United States sponsored by Public Roads, the Asphalt Institute, and the petroleum industry, 13 grades were adopted by most States and national groups. In most cases, the reduced number of grades for both penetration grade and liquid asphalts were adequate to fill the needs for pavement construction and resulted in considerable economic benefit to the user and producer of the materials.

Development of the Thin-Film Oven Test

The control of the hardening of asphalt in hot-plant mixing and in mixtures in service has been of major concern to the engineer and chemist since asphalt was first used. In 1940 the Public Roads Administration developed a laboratory test that would predict the amount of hardening of asphalt in hot-plant mix construction. This became known as the “Thin-Film Oven Test” and has been adopted as a standard specification test by ASTM, AASHO, and essentially all State highway departments. This test was also adopted by several foreign countries to measure and control asphalt hardening.

Development of the Immersion-Compression Test

The effect of moisture on asphalt paving mixtures, macadams and surface treatments has been one of the primary causes of pavement distress or early failure. Tests were available to measure the effect of moisture on different aggregates coated with asphalt, but a test was needed to measure the effect of water on complete compacted mixtures representative of those used in construction. To meet this need, the immersion-compression test was developed, based on a laboratory study made by the Bureau in 1942. The test produced valuable information on the susceptibility of asphalt mixtures to loss in strength by water action.[9] The immersion-compression test has been adopted as a standard by ASTM, AASHO, and many State highway departments and has been a valuable tool for investigating the cause of distressed pavements.

Bituminous mixtures being prepared in the Public Roads Administration research lab for immersion-compression tests to ascertain durability of materials for highway construction in 1948.

Cement and Concrete Materials

The name “portland cement” comes from an “artificial” cement patented by Joseph Aspdin in England in 1824. Aspdin kept his process a secret, although many others tried to duplicate this product. In 1845, I. C. Johnson developed a product based on high temperature calcination that was essentially “portland cement” as we know it today. Portland cement was first manufactured in the United States in 1871.[10]

Concrete technology was in its early development at the beginning of the 20th century. Today’s refined knowledge of concrete design and utilization is the product of a number of organizations, such as Lewis Institute, American Concrete Institute, American Society for Testing and Materials, Portland Cement Association, State highway departments, and the Bureau of Public Roads. In particular, BPE played an important role in developing knowledge of the relationship of concrete’s various components to concrete strengths and other characteristics and in developing better tests and specifications for the product. In addition, BPR performed studies which established the quantitative relation between some aggregates and the alkali content of cements and determined the value of fly ash in concrete to control the damaging alkali aggregate reactions as well as serve as a replacement for part of the cement. Other research assessed the sulfate resistance of various concretes and the relation of such resistance to cement composition. Air entraining agents to prevent freeze-thaw damage were also evaluated.

As the science of manufacturing portland cement developed and more knowledge of the effect of special compositions was gained, specialty cements such as “high-early-strength portland cement” or “sulfate resistant” cements began to appear on the market. This led to the adoption by ASTM and AASHO around 1940 of five types of standard cements.

Research engineer scans slice of concrete with microscope to determine size and distribution of air voids, important in the durability of concrete, especially under the action of freezing temperatures.

The development of standard cement tests was an important part of Public Roads’ research in the late 1920’s and early 1930’s. Difficulties arose when discrepancies were found in the results of tests run by different testing laboratories around the country whose function was to verify that highway construction materials met their required specifications and also to test samples of completed highways. This problem led to the establishment, in 1929, of the Cement Reference Laboratory (CRL) at the National Bureau of Standards. It was an ASTM sponsored activity actively supported by Public Roads from its inception. The CRL’s duty was to determine whether the laboratories’ equipment was in order and to observe whether the proper techniques were being used in making the tests.[N 1]

Aggregates for Construction

The simplification, standardization, and uniform application of aggregate gradations has been a major effort since 1936. This work led to the early adoption of a simplified practice of specifying sizes of coarse aggregate for concrete. In 1962 Public Roads encouraged the States and other agencies to adopt simplified procedures for aggregates used in bituminous construction. The recommendations included the development and adoption of standard aggregate sizes, standard sieves, and a standard method of reporting gradations. Public Roads also developed a new gradation chart that showed desired combinations of aggregate sizes in graphical form. Since fewer blends were required, all customers could be supplied from only a few stockpiles and producer costs are minimized. This chart is now in wide use for evaluating aggregates in bituminous paving mixtures.

In 1928 studies were made of concrete containing rounded gravel coarse aggregate versus the angular material produced by crushing ledge stone or large-size waterworn gravel. This research determined that crushed stone is generally preferred for paving work because of the superior strength of the resulting concrete.

Concrete slabs are subjected to daily applications of deicing agents, such as calcium chloride, to determine the durability of the concrete or the effectiveness of various protective treatments.

Research on concrete freezing and thawing was also being conducted at this time. Public Roads research led to the development of new concrete specifications requiring denser and more durable types of coarse aggregate. Other research in this area included the development of air-entrainment techniques and the establishment of a list of acceptable air-entrainment admixtures. Air entrainment is a process that puts air bubbles in the cement to relieve the stresses caused by freezing water. Later, Public Roads assisted in the development of ASTM and AASHO specifications and guidelines for air-entraining agents.


  1. * In 1960 the CRL was expanded into the Cement and Concrete Reference Laboratory (CCRL).

Coating Materials

Beginning in the 1920’s, Public Roads became concerned with the need for improved field performance of such common highway coating materials as paints and galvanizing for metals for guardrails, more durable materials for culverts, and more lasting and visible paint for lane markings and directional highway signs. A special chemical unit was established to conduct research in this area.

A significant result of this effort was the development in the late 1940’s of a modified anticorrosion paint primer for the bridge steel being shipped to the Philippines for rebuilding its war-damaged highway network. The standard red lead-linseed oil primer then in use was found to be badly scratched and significantly stripped from the steel after its arrival in the Philippines. To compound the problem, the exposed steel also corroded rapidly when stored under the humid Philippine conditions. Studies showed that a pigment combination of red lead and iron oxide, as well as a combined base of linseed oil and alkyd resin, would provide adequate corrosion protection and would be tough enough so that shipment damage and deterioration in storage was minimal.[11] The specification developed has since become an AASHO standard used by a number of State highway departments.

Another development in coating materials was an abrasion-resistant paint system for highway structures. Exceptionally high wind velocities combined with wind-borne ice and soil particles in the Alaskan Copper River Delta produced almost immediate and extensive abrasive damage to the standard bridge paint system in use. Laboratory and field research by Public Roads and the Alaska Department of Highways were completed in 1965 and demonstrated that rubber-based coating systems offered superior abrasive resistance to these destructive forces.[12]

Because traffic paints do not last long in heavy traffic, there has been a continuing interest in developing better and more economical lane marking material. This has led to three distinctly new and rather revolutionary types of lane markers. In general, the new materials have been developed by industry with evaluations performed by State highway departments and technical assistance and funding help from Public Roads.

The new materials are:

  • Prefabricated plastic striping which is supplied either as short rectangular segments or in rolls like paper towels and applied to the road surface with permanent cement.
  • Thermoplastic striping applied hot by either extrusion or spraying and capable of self-adhesion to the road surface. The stripe has good visibility and excellent life characteristics.
  • Plastic dots reflectorized to have high visibility in rainy weather and protected by steel or other housings or base plates, the unit being firmly cemented to the pavement surface. These devices are being designed to have a very low profile, and the cement now available holds them so firmly to the pavement surface that they are not readily torn loose by street sweepers or even by snowplows.

Each of these lane-marking developments is somewhat more expensive in the initial installation than conventional painted striping, but the initial extra cost is often offset several times over by the greatly extended life, which has proved to be as much as ten times that of painted stripes.

New Analytical Techniques

During the late 1940’s and early 1950’s Public Roads pioneered in the application of sophisticated instruments to help solve materials problems. Infrared spectroscopy was developed to identify the nature and understand the behavior of chemical admixtures for concrete, mineral composition of aggregates, and other highway materials. An infrared spectrophotometer was acquired in the late 1950’s and was successfully applied to analyzing, classifying, and regulating the use of certain admixtures for concrete, such as water reducers, hardening retarders, and air-entraining chemicals. These materials were of such complex chemical nature that they had defied earlier less sophisticated methods.

Pavement Research

In 1908, Director Page reported on 13 experimental road sections on main roads connecting New York and Boston. These experiments resulted in the successful upgrading and restoration of deteriorated travel surfaces with various water-gas and coal tar products used primarily as dust palliatives. Additional reports by Page during the period 1909 through 1918 covered numerous experiments involving dust prevention and road preservation. These early treatments did much to make automobile travel faster and more pleasant and were relatively inexpensive for a single application. However, since they required frequent renewal, the total cost was appreciable, and they finally had to be abandoned.

Infrared spectrophotometry analysis can identify substances present in such materials as paints, concrete admixtures coatings, rubber, etc., by determining the frequency at which radiation is absorbed by the substance.

In the early 1930’s, a gradual transition took place from dust control treatments to surface treatments, using tars and asphalts and better mineral cover. Also, during this period and continuing into the 1950’s, many miles of penetration and waterbround macadam pavements were built. These roads served growing highway needs very successfully, and considerable mileage has survived to the present time. Concurrent with the development of more substantial surface treatments, primarily in the east and the south, many central and western States experimented with and developed effective procedures for the type of pavement known as bituminous road mix.

In this period, Public Roads research was very active in coordinating field studies of roads in service with laboratory research on materials. In the 1930’s, for example, studies in Colorado and Wyoming demonstrated the need for sealing or “surface treating” road-mixed surfacings. As a result, bituminous seal coating became a fairly regular procedure. Another example was a study of bituminous concrete roads in Ohio which demonstrated the general serviceability of this paving.

Concrete Pavement Research

Many articles were published documenting the Bureau of Public Roads’ research on portland cement concrete pavements. Among the earliest was a study in 1919 on the behavior of concrete slabs. The measured curling and warping movements of the slab were evaluated, in addition to the effect of wheel loading, and a thickness design formula based on the corner break was developed. This was followed in 1923 by Dr. H. M. Westergaard’s theoretical analysis of slabs on an elastic medium. There followed a period of detailed studies by Bureau of Public Roads’ researchers lasting until 1936 to verify and expand his results into a comprehensive design procedure, published as The Structural Design for Concrete Pavements, which has remained the basic guide for a decade or more.

Experimental roads continued to be built to test new concepts in concrete pavement. In 1921, the Bureau of Public Roads built the first experimental continuously reinforced concrete pavement near Washington, D.C., where many of the basic principles related to that type of pavement were developed.[13] In 1937 the Bureau, in cooperation with the Indiana State Highway Department at Stilesville, began an extended field test of continuously reinforced concrete pavement, which now comprises a significant portion of the Interstate System mileage. The BPR also worked with industry in 1957 to produce the first U.S. prestressed concrete highway pavement, and, subsequently, built a 3,200-foot experimental section of pavement at Dulles International Airport in 1971.[14] This concept too is gaining widespread acceptance.

The Bureau of Public Roads roughometer measures the roughness of the pavement in units of vertical spring motion per unit of distance the wheel travels.

Research was also performed on reinforcing materials for concrete pavement. Bond studies in 1926 or steel reinforcing bars of various shapes and types of deformation included pullout tests to evaluate the contribution of the different types of bar deformation to the load-carrying capability of concrete slabs. The superior performance of several specific types of bars was established. In addition, results in 1958 of an extensive laboratory study of the performance of load transfer dowels in concrete pavement joints had an immediate effect in pavement design. Recommended dowel bar lengths and diameters for effective load transfer were widely adopted and are still in use today.

The Bureau of Public Roads was an early leader in efforts to provide a means of obtaining quantitative measures of pavement roughness. The result was a compact single-wheel trailer that measured and recorded the roughness of the wheel path in units of vertical spring motion per unit of distance traveled. With this device, commonly known as the BPR roughometer, it became possible to easily rate the roughness of both new and old pavements. The term “present serviceability index” as a function of roughness was used to express the results in numerical terms. This index was used to specify the degree of roughness for design standards and in rating construction.

Among other items of special equipment designed by Public Roads engineers was a device to measure the vertical displacement of pavement, both elastic and permanent, caused by static and moving wheel loads of various intensities and at different distances from the point of load application. With these displacement data, it was possible to make design analyses for better subgrades and pavements. This ingenious device, named the Benkelman Beam for its inventor, received nationwide attention and was duplicated by many State highway departments.

In addition to tests on roads in their normal condition, studies were made in the 1920’s to determine the impact forces of wheel loads generated by an “artificial” bump. One study was on a ramp 30 inches long with a drop-off varying from 1½ to 3 inches. These tests demonstrated the superiority of pneumatic tires versus solid rubber tires. Partly as a result, solid rubber tires were rapidly phased out in the late 1920’s.

Service Lives

Highways do wear out. Because they wear out, highway programs must include continuous resurfacing and reconstruction operations to maintain highways in a usable and safe condition. Accordingly, a knowledge of service lives of highway pavements is essential.

Taking deflection measurements with the Benkelman Beam to determine pavement displacement when a load is applied.

Testing impact of vehicle wheels on pavement.

This machine was used in 1922 to test the impact of solid rubber tires on test pavement sections. By raising the wheel and suddenly dropping it at impact forces comparable to truck traffic on the roads, comparative pavement design strengths could be determined.

Research in the field began in 1935 when road life studies were incorporated as part of the statewide planning surveys. At one time or another nearly all States cooperated in this research by providing the basic data which were analyzed by Public Roads staff. Although the purpose of the surveys was to determine average service life of road surfaces, the data also provided the means for obtaining construction costs, salvage values of retired roadway elements, and service lives of structures, gradings, and rights-of-way.

Based on an analysis of the records of surfaces and roadway elements previously constructed and depreciated, it is now possible to estimate the amount and cost of replacement of these elements. For example, the results of a study completed in 1971 indicated that of the total miles of roads remaining in service in 1968, 60 percent will be retired in 10 years and 87 percent in 20 years.[15] Such information is essential to determine construction and reconstruction programs and corresponding revenue needs of a future period.

Bridge Research

Early highway bridge research was conducted in response to the need to shape the AASHO standard specifications for highway bridges, which were gradually developed between the formation of the AASHO Bridge Committee in 1921 and the first printing of the standards in 1931. Subjects of early bridge research studies included the expanded use of welding and high-strength bolting for connections in steel structures, the widespread acceptance of continuous composite bridge design and new techniques of construction, such as orthotropic deck design and cable-stayed girder bridges. Other important developments were specifications covering the fatigue life of steel and the widespread adoption of new high-strength structural steels for bridges.

A unique aspect of early structural research activity resulted from the catastrophic failure of the Tacoma Narrows Suspension Bridge in the State of Washington on November 7, 1940, due to flutter induced by high winds. Under the auspices of the Advisory Board on the Investigation of Suspension Bridges, formed in September 1942, a wind tunnel large enough to accommodate a scale model of the entire Narrows Bridge was designed and built at the University of Washington. This tunnel was used to make exhaustive studies of the causes and possible remedies for such a failure. After extensive redesign utilizing the findings of that study, the Narrows Bridge was rebuilt and has since served well without excessive vibration or other evidence of distress. One feature of the redesign was a slotted or grid type of deck which largely relieved the vertical component of wind streams impinging on the bridge.

The extensive investigations which followed the collapse of the Narrows Bridge led to the construction in 1950 at the Fairbank Highway Research Station of the George S. Vincent Memorial Wind Tunnel, where model studies on the effects of winds on highway structures were conducted. Models of many of the major suspension bridges throughout the United States and abroad have been investigated for aerodynamic stability in this wind tunnel.

One of the subjects of those studies was the Golden Gate Bridge. Excessive wind-induced vertical oscillations at a maximum amplitude of 13 feet had occurred while the Narrows Bridge investigation was still underway, causing great official and public concern. Consequently, section-model studies of the bridge were performed in the Vincent wind tunnel. Some structural stiffening of the suspended portions of the bridge, as well as the introduction of slots in the deck to reduce the vertical component of the wind force, proved effective, as in the case of the redesigned Narrows Bridge, in remedying the excessive oscillation.

A great deal of research was also initiated on all of the common bridge construction steels in order to learn more about their sensitivity to brittle fracture. Once this research began to produce results, tentative toughness specifications for bridge steels, using the Charpy V-Notch impact specimen as a control, called for tougher steel with increased steel strength and thickness. These requirements were adopted by the AASHTO Bridge and Materials Committees for inclusion in the 1974 interim specifications.

Aerodynamic studies of suspension bridges are conducted in the George S. Vincent wind tunnel. A precise model of a portion of the bridge deck is mounted on springs that match the stiffness of the actual cable system with instrumentation to measure the oscillations produced by the wind stream.

One of the circular test tracks built in Arlington, Va., for measuring the impact forces of various wheel loads on pavements.

New Structural Concepts

In the early 1950’s, the use of prestressed concrete structural members in this country was hastened by the many research studies dedicated to the solution of problems arising in connection with the adoption of this new structural concept. As a result, the use of prestressed concrete bridges advanced rapidly in the United States, achieving predominance as a construction material in many sections of the country. Similarly, an intensive program of structural research led to the widespread acceptance of the use of high-strength bolting of steel structural connections, an improvement over the previous exclusive dependence on riveting.

A profilometer is used to measure wear and displacement of pavement caused by traffic on the Arlington, Va., test track.

Public Roads studied a number of full-scale bridge loadings in cooperation with the various State highway departments beginning with the Yadkin River Bridge Test in 1928. And in 1950, Public Roads instrumentation and assistance were made available to any State highway department requesting assistance. As a result of these studies and the road tests, which included bridges, a significant contribution was made over the years to the improved fatigue design of steel structures. Out of these studies came a better-understanding of such theoretical concepts as the effects of structural fatigue on load distribution between members and the effect of incorporating new materials, such as high-strength reinforcing steel or lightweight concrete, on the performance of concrete highway bridges.

With the beginning of the Interstate System came the need for fitting bridges to much more restrictive and sophisticated geometries to provide for the needs of high-speed traffic in congested areas while maintaining structural esthetics and economy. Straight span members were no longer always adequate for the needs, and the understandings of the structural design of curved members was limited. Research undertaken, both in the form of multi-State pooled-fund laboratory and analytical studies and the extensive use of cooperative Federal-State field studies, helped fill this urgent and critical need for improved curved girder design technology. As a result of this research, bridges today are designed and built with curved alinement and warped decks in a manner unheard of two decades ago.

Road Tests

Following World War I, a prime objective of the State and Federal highway program was to reconstruct the surfaces and pavements of main highways which had received widespread damage from the greatly increased traffic of heavy trucks with solid rubber-tired wheels. The light surfaces, designed for horsedrawn or other light vehicles, failed under the impact of the heavier vehicles. Road test data were needed to determine the supporting capability of the various subgrade soils, the stresses induced in rigid and flexible pavements by the impact of motor vehicles, the effects of expansion and contraction of road surfaces caused by variations in temperature, the wear of traffic upon pavement surfaces, the distribution of loads upon bridges, and many other factors. To solve some of these problems, the Bureau of Public Roads and the States initiated a series of road tests.

The Bates test road and test vehicles.

The Arlington Road Test. The first road test was initiated in 1918 at BPR’s Experimental Farm in Arlington, Virginia, to measure qualitatively the impact forces of various wheel loads. The tests, conducted with Army trucks equipped with solid rubber tires, were on concrete, brick and bituminous slabs on circular tracks. The early tests indicated the major effect of wheel impact forces and led to further studies with more refined measuring instruments and the inclusion of pneumatic tires.

The Bates Road Test. For the Bates Road Test, the Illinois Division of Highways constructed 68 test sections, including six types of pavement on 2 miles of road and, with the assistance of BPR engineers, conducted tests on them from 1920 to 1923. Trucks were operated with controlled wheel loads that were progressively increased from 2,500 to 13,000 pounds. The test data gave the pavement type and thickness required for a specified loading and showed the need for control of wheel loads. A direct result of this study was the use of a thickened edge concrete pavement.

The Pittsburg Road Test. Between 1921 and 1922, another road test was conducted in Pittsburg, California, to determine the efficiency of both reinforced and plain concrete pavements of variable thickness and designs on certain types of subgrade soil. The test was conducted by the Columbia Steel Company on an elliptical track 560 feet in length containing 13 concrete pavement test sections. Much new information was learned, particularly concerning the effectiveness of longitudinal joints in preventing longitudinal cracking.

The combined findings from these three contemporary road test studies led to major advances in pavement design practices. The relations were determined between concentrated wheel loads and the thickness of several common types of pavement, which were directly usable on new highway plans. From the range of wheel loads studied, highway officials were able for the first time to reach agreement on the use of a 9,000-pound maximum value as the economic standard for highway pavement designs. More directly evident to the public was the abrupt phasing out of solid-rubber tires on trucks because of their high impact destruction on all types of pavements. Within a few years after 1926 the use of pneumatic tires became universal. Also, these studies made evident the need for amplified research on soil support values to attain better pavements.

The Hybla Valley Road Test. From 1944 to 1954 a series of studies on flexible pavements was conducted on the Hybla Valley test track in Alexandria, Virginia, by BPR in cooperation with the Asphalt Institute and HRB. A 2,000-foot oval track for full size vehicles was built with selected soil foundation and paved with asphaltic concrete. Using newly developed measuring equipment, data were obtained on the vertical displacements of the pavements from both static and moving wheel loads of varying intensities.

A significant finding from these tests was that there is considerable elastic movement within a nonrigid pavement structure under load. A great deal was also learned about instrumentation and field measurement procedures which greatly aided later test road studies.

Road Test One—Md. In 1949 the Interregional Council on Highway Transportation originated a test road study on a 1.1-mile section of highway south of La Plata, Maryland. The study and report was a cooperative effort of 11 eastern States, the District of Columbia, Public Roads, truck manufacturers, the petroleum industry, the Department of Defense and HRB. The objective was to obtain data, for use both in vehicle weight regulations and in pavement design, on the relative effect of four different axle loads on the existing concrete pavement (four lanes), then in excellent condition.

Test trucks with 8,000- and 22,400-pound single axle loads were operated on adjacent lanes and trucks with 32,000- and 44,800-pound tandem axle loads operated on the other pair of lanes. Nearly 240,000 single axle truck passes were made and about half as many of the tandem axle.

The data on the progressive pavement deterioration, evidenced by cracking, internal structure failure, and joint depression, showed that damage was proportionate to axle loads. Significant data also were reported concerning soil support features and stresses at various points in the pavement slab.

WASHO Road Test. In 1951 the Western Association of State Highway Officials (WASHO) set up a test road to obtain data for use in establishing load limits and in designing flexible pavements. Cooperating in this test were 11 western States, truck and truck trailer manufacturers and three petroleum companies. The Highway Research Board supervised the construction and testing to determine the effects of four-axle loads on selected designs of flexible pavement. The Bureau of Public Roads assisted with funds, personnel, instrumentation, photography, supplies and equipment. This test was run from 1952 to 1954 at a site near Malad, Idaho.

Two four-lane test loops, each with 1,900-foot tangents, were constructed with five pavement structural sections ranging in depth from 6 to 22 inches on each tangent. On one loop the pavements were subjected to 18,000- and 22,000-pound single axle loads and on the other 32,000- and 40,000-pound tandem axle loads. The lighter loads were operated on the inner lanes. About 240,000 truck passes were recorded on each of the loops.

The test data yielded significant relations of stresses within and failures of the several flexible pavement design sections as the truck loadings progressed. Some of the general findings were:

  • The 4-inch asphaltic concrete top course was markedly superior to the 2-inch.
  • The heavier axle loads resulted in considerably more distress.
  • A paved shoulder contributed to the pavement support.
  • Greatest distress occurred in the spring and least in the fall.
  • Distress from the tandem axle was equivalent to that of a single axle of about two-thirds of the tandem weight.

The AASHO Road Test. After the WASHO Road Test, the American Association of State Highway Officials (AASHO) in agreement with the Bureau of Public Roads decided to conduct a comprehensive national test to obtain data on all significant variables. In 1955, AASHO sponsored and requested the Highway Research Board to direct this test. The purpose was to study the performance of pavement and bridge structures of known characteristics under moving loads of known magnitude and frequency. Portland cement concrete and asphaltic pavements, as well as certain types of bridges, were included in the specially constructed test facility.

This $27 million project, located near Ottawa, Illinois, was financed by the highway departments of the 48 States, Hawaii, the District of Columbia, and Puerto Rico. Financial support was also provided by the Bureau of Public Roads, the Automobile Manufacturers Association, the petroleum industry, and the American Institute of Steel Construction. The Department of Defense furnished the heavy vehicles and drivers for their test run operations. In addition, there were many services and contributions by the automotive, petroleum, tire, cement, and steel industries.

Traffic on loops during the AASHO road test.

The test road consisted of about 7 miles of two-lane pavements, half of concrete and half bituminous. The roadways included 16 short-span bridges. There were 836 separate test sections located in six loops which varied in length from 2,000 to 6,600 feet. The test sections had selected factorial combinations of surface, base, and subbase thicknesses. Axle loads differed on each loop and varied from 2,000-pound single to 48,000-pound tandem; truck trailer units were included. Full-scale tests started in November 1958 and terminated in November 1960. A total of 1,114,000 axle loads had been applied to the pavements and bridges surviving at the last date.

The project produced a reservoir of facts long needed for the development of a more refined and scientific design of pavements and short bridges.[N 1]

The test data well established the desired relations of pavement structural designs (component thicknesses) and loadings (magnitude and frequency of axle loads). These findings were developed in the form of equations and graphs that showed the effects of particular variables on pavement performance. These soon were incorporated by AASHO and others into pavement design guides and manuals. The concept and use of a “serviceability index” to define pavement performance was a major item. And the data on equivalencies of single- and tandem-axle loads for the same pavement performance was a high-use product. Also important were: a method to predict pavement performance from measurements of deflections and strains; and data on the reduction of surface skid resistance.

The test bridge findings largely verified design predictions of deflection and strain. The dynamic studies resulted in the formulation of new theoretical concepts for analytical evaluation of stresses and deflections from moving vehicles.


  1. The extensive technical data and findings are recorded in seven Highway Research Board Special Reports Series 61, and Special Report 73.

Construction and Maintenance

Production Costs

The construction of highways in the early stages of our country’s development involved mostly hand labor and draft horse or mule power. Formal organization and production efficiency were items unheard of, and the pace of construction depended, for the most part, on the forcefulness and ingenuity of the foreman. Later, as mechanized methods started to supplement the man and mule power method, the need for systematic organization increased. Early Public Roads activities in this area concentrated mostly on object lessons gained in test road construction.

In the early 1920’s, the Production Cost Study Program was initiated by the Bureau to assist the highway industry in analyzing its various operations regarding time utilization and operational efficiency of equipment used in construction. This program, led by T. Warren Allen, served two purposes: (1) To assist the contractor by direct time-motion studies on their operations, expressed in terms of production rates and costs, pointing out observed inefficiencies and measures as to how he might improve on his operations; and (2) to train junior engineers in Public Roads by direct exposure to the practicalities of construction equipment operations and the overall process of building a highway. This program continued until 1936.

There were 16 short-span bridges on the AASHO road test to determine deflection and strain on bridges from heavy loads.

Because of personnel shortage during the war emergency, the program was dormant for 10 years and was reinstituted in 1946. By then the highway construction and maintenance industry, with its far flung operations reaching into every corner of the country, was using thousands of major equipment units and millions of dollars worth of material each year. Economical and efficient use of this equipment, manpower and materials was of concern to every segment of the industry in order to stretch the highway dollar to meet the need for more and better roads and yet assure the competitive contractor a reasonable profit.

During the next 20 years, production time and motion studies were made on over 400 projects. Over 40 information reports were issued prior to 1960 by Public Roads and the Highway Research Board on performance, time utilization, and costs pertaining to equipment employed on highway construction and maintenance work. In addition to routine production studies on maintenance and construction operations, the program was directed also into other areas to gather information on a particular facet of construction. In 1952, a year-long comprehensive study was made with the objective of developing comparative data regarding net cost to the public of construction work performed by contract and by State forces in North Carolina.

These cost studies were later expanded to obtain data on economics of detours versus carrying traffic through construction and on economics of specifying a single cold feed bin versus multiple bins for hot-mix production. Other cost studies were made on various types of bridge construction and grading and paving work to develop comparative man- and equipment-hour unit cost data. Teams also studied such areas as: (1) Determining relationship between mixing time and productivity for dual drum pavers and for central mix plants, (2) mechanics of aggregate drying in hot-mix plants, (3) optimum rolling pattern on hot-mix construction, (4) benefits of blending material prior to loading the concrete mixer, (5) evaluating a newly developed nuclear gage for determining density of bituminous pavements, (6) evaluating time lapse movies technique for studying construction, and (7) demonstrating the need for surge bins. Also, a series of short-term and long-term pilot maintenance equipment and operations studies were conducted to pinpoint areas needing better organization and training.

Michigan was the site for a winter driving test to determine safe speeds and handling of heavy vehicles on ice and snow, including negotiating turns and curves.

This program was again terminated in 1966. The Federal Highway Administration resumed this program in 1971 on a more limited scale. The benefits that accrued from the early pre-World War II and the later post-World War II programs were widely reflected in many positive improvements in productivity. It can be said that they were a major factor, directly and indirectly, in the constant evolution leading to automation of construction, control, and operation.

Maintenance Management

Even though research on highway materials and construction processes had an influence on highway maintenance and, in part, directly applied in maintenance operations, the separate study of management of maintenance operations differs substantially from the other areas of highway research. These investigations involved a considerable number of States, and their results have affected all States.

Highway maintenance has experienced many changes through the years, most of which were based on intuition and practical considerations rather than factual knowledge and scientific management principles. There were some limited-scope highway maintenance management studies through the 1940’s, but these did not constitute a significant serious research program.

A 1950 joint study with the Connecticut State Highway Department developed facts concerning the performance of labor and equipment and appraised management problems. Subsequently, about 20 other small studies were conducted by the States during the 1950’s. The results were not singularly significant, but indicated a common need for better management. In 1959, a major study by the State of Iowa, in cooperation with Public Roads, developed a comprehensive system for producing facts and analyzing management aspects. This report received considerable publicity and the study’s system was adopted in various ways by many highway maintenance organizations.

During the 1960’s, emerging conditions that required a refined maintenance management role included the new Interstate System, the need for higher levels of maintenance, a rapid change in technology, and labor and budget problems. As a result, maintenance management became much more prominent in both research and practice.

Research to improve maintenance management was greatly expanded to meet these challenges. Between 1960 and 1970, about half of the State highway organizations conducted maintenance management studies. There was considerable cross-fertilization between these studies, and they covered a very wide range of elements, from equipment development and roadside practice to statewide planning and economic investigations. In these studies, a number of major management consultants were involved as well as State universities. The results led to major improvements in maintenance operations throughout the United States, and more refined studies are in progress.

Traffic and Safety

Capacity and Design

In response to Chief MacDonald’s concern, research was directed in the 1920’s into three major areas: (1) Road construction and maintenance, (2) the economics of road operation, and (3) the economic value of highways to a community. Although elements of highway capacity and design were mentioned, it was W. K. Hatt, Director of the Highway Research Committee of the National Research Council, who in 1921 raised the following questions: “What is the capacity of a road of a given width for any particular type of vehicle as expressed in vehicles per hour, ton-miles per year, etc.? What is the appropriate unit for expressing traffic for various purposes?”[16]

Research on Highway Geometries

In March 1925, Public Roads first reported a number of significant facts concerning the lateral distribution of traffic. These factors included the effect of road width, curves, shoulder conditions, grades, surface crown, and other physical features. Observers recorded the positions of tires in relation to lines painted across the pavement at 1-foot intervals. The results indicated that 18 feet was the minimum pavement width which would permit passenger cars and trucks to pass in safety and with a reasonable amount of clearance. Speeds were not considered in this study.

Four years later, in 1929, when equipment for measuring speeds became more reliable, it was found that the safe passing of “rapidly” moving automobiles and trucks required a surface width of at least 20 feet for two-lane rural highways.

Since 1929, researchers have progressively provided the basic information required to set design standards and operational controls for main roads and freeways. In December 1944, one of the recommendations of a BPR study of the effect of roadway width on traffic operations was that lane widths should be 12 feet for safety and comfort of traffic operations. The AASHO soon adopted this width as a standard for primary highways in the United States. Within a short time this standard was also adopted in many foreign countries. Vehicle widths remain about the same today, and this lane width remains the basic standard.

Horizontal and Vertical Curves

An early design policy statement that is still current says that “In the design of highway curves it is necessary to establish the proper relation between design speed and curvature and also their joint relations with superelevation. While these relations stem from laws of mechanics, the actual values for use in design depend upon practical limits and factors determined more or less empirically over the range of variables involved.”[17] A series of studies have been made to relate the speed of operation on highways to the horizontal and vertical alinement. One of the earliest studies on the effect of speeds on geometric design was conducted in 1929. Even though the legal speed limits in the 1920’s were 35 to 45 miles per hour, A. Bruce, in his article “The Effect of Increased Speed of Vehicles on the Design of Highways,” concluded that there was need for easier curves, greater superelevation of curves, more extensive vertical curves, and greater sight distance on both horizontal and vertical curves.

In 1953 a comprehensive study of driver behavior on vertical and horizontal curves was completed by Public Roads in cooperation with the New York Department of Public Works that showed that the percentage of drivers exceeding a safe speed for the curve radius increased greatly on the sharper curves. The combination of curves over 5 degrees (radius less than 1,200 feet) and grades in excess of 5 percent also were found to result in an especially high accident rate on conventional rural highways.

A comprehensive study of “Driver Performance on Horizontal Curves” was conducted during 1951–1954 in cooperation with five States at 35 horizontal curves. The separate studies were on two-lane, two-directional roads. Among the more important conclusions were the following:

  • Drivers did not change their speeds after entering a horizontal curve.
  • The existing superelevation had no effect on speeds.
  • Operating speeds and the radius of curve are linearly related.
  • The curve radius had a greater effect on speeds than sight distance.[18]

In 1969 data in NCHRP[N 1] Report 68 on the application of vehicle operating characteristics to geometric design substantially verified these results.


  1. National Cooperative Highway Research Program.

Highway Capacity

Highway capacity has been the subject of careful and painstaking study for nearly six decades. A rational and practical method for the determination of highway capacity was essential for the sound economic and functional design of new highways and for the many existing roads and streets which must continue in use for extended periods of time. Basically, highway capacity concerns the effectiveness of various highways to serve traffic and involves the many elements of highway design, speeds, vehicle and driver performance, and traffic control. In recent years, due to the development of refined traffic study methods, instrumentation and equipment, substantial amounts of reliable field data have been used to develop new insights into problems of and solutions to traffic operations.

In 1934 the BPR set up and took a dominant part in a coordinated series of studies to obtain the basic data on highway capacity. This widespread effort included studies on: methods of counting traffic;

Olav Koch Normann

Olav Koch Normann
Olav Koch Normann

In a career foreshortened by his untimely death at age 57, Olav Koch Normann’s contributions to safe and efficient highway travel could be matched by few others in a full lifetime of effort. Endowed with a rare combination of brilliance of mind, skill of hand, and enormous physical strength and endurance, O. K. Normann set a pace that few could follow. He was a leader in many fields but became best known as the ‘father’ of the world’s knowledge in the area of highway capacity.

Normann was born in Kansas City, Missouri, on October 1, 1906, and earned his Bachelor’s degree in Civil Engineering at the University of Minnesota in 1928. Immediately entering the Bureau of Public Roads as a junior engineer in its training programme, Normann began the career that kept him in the Bureau until his death.

His inquisitive mind led him early into the field of research, and in 1935 he moved into the field of geometric design and highway capacity.

The approach he followed would probably now be called operations research or system analysis in today’s more sophisticated language. Then it was recognized simply as the only logical way to approach the problem. It accepted the movement of traffic as a dynamic system involving the vehicle, the driver, and the road. Organizing studies in each of these segments, Normann himself synthesized the separate results into principles derived from actual experiment on the road and in the traffic stream to reflect the countless possible combinations of the individual and collective actions of drivers and differing vehicle performance and roadway design elements.

This approach might not have been so successful had it not been backed by Normann’s own broad talents. As an analyst, he had an uncanny ability to detect trends and relations within masses of data that customary statistical methods did not seem to reveal. While taking full advantage of modern data processing and analysis techniques, he did not forsake simple graphical and other methods that on more than one occasion proved out after the more complex methods failed.

It was this depth of investigation that showed clearly, as research progressed, what specific facts were lacking or where data in different form would have helped. Here another of Normann’s talents, mechanical skill, came to the fore, for seldom, it seems, was there available the apparatus or equipment needed to measure particular facets of driver behaviour or vehicle performance. Not only could Normann specify what was needed — he could design and construct it, sometimes with the design only in his mind, not on paper.

The impact of Normann’s early research and writing led to his selection as Chairman of the Highway Capacity Committee of the Highway Research Board when it was first organized in 1944. His selection as chairman was unique, for at that time, as a general policy of the Bureau of Public Roads, its employees could serve as members or as secretaries of Board committees, but never as chairmen. Normann’s accomplishments were so predominant in the field, however, that there could be no sensible alternative to disregarding the policy in his case.

Years of work by the committee members, aided by heavy contributions of data obtained by the Bureau of Public Roads and many state highway departments and city traffic engineering organizations, led to the publication of the Highway Capacity Manual in 1950. Much of the willing support of this committee can be attributed not only to the desire of various agencies to cooperate in the Board’s activities, but also to the personal regard individuals in these organizations felt for Normann, the man, not just the chairman. Published with some early reservations as to the breadth of interest it would attract, more than 26,000 English language copies have been sold, and the manual has been translated into nine other languages.

Disbanded upon publication of the manual, the committee was subsequently reorganized by the Highway Research Board in 1953, again with O. K. Nbrmann as its chairman. Again, there could have been no other sensible choice. The committee was charged with extending and updating the original manual, and at an appropriate time, presenting a revised version for publication. The new manual was nearing completion when death took the chairman. It will be his monument.

Although perhaps most widely known for his work in highway capacity, Normann’s research led him into all facets of geometric design of highways, and many of today’s standards of curvature, sight distance, gradients, lane widths, and intersection design are traceable directly to his efforts. Again, it was his understanding of the highway as but one element of a system that enabled him to design his experiments so as to produce results of immediate applicability. Important as he regarded the adequate reporting of research results, he felt their greatest value lay in prompt application.

As his technical field broadened, he was called upon, as is often the case, to shoulder increasing administrative duties. He advanced steadily through positions of heavier responsibility, until at the time of his death, he was serving as Deputy Director for Research of the Office of Research and Development in the Bureau of Public Roads. Despite the burden of administrative duties, his interest in technical matters in no way slackened and he seemed to regard the administrative functions as merely added, not different, responsibilities. In effect, he did two jobs, each in his characteristically thorough and exacting way—something he could do because of his energy and physical endurance, but which evoked mixed feelings of admiration and despair among those whom he hopefully expected to keep pace.

He was always a part of his community and accepted civic responsibilities with the same energy and enthusiasm that marked his professional life. He was active in and served as president of his Civic Association, and as chairman of Better Government Committee of his county’s federation of civic associations.

In all aspects of life, not merely in his professional area, Normann was a confirmed skeptic. In his view, there must be some better way to do it, whatever it might be, than the way it was traditionally being done, Little with which he was connected failed to show some evidence of innovation before its completion—a healthy attribute in research, but not always so desirable in other fields, such as, for example, home building. Many are the anecdotes which will live long in retelling among his vast circle of friends that stemmed from Normann’s innovations, and from his amazing resourcefulness in turning apparent setbacks and failures into advances and success.

Through his highway work he became widely known around the world. Many engineers from other countries, in travelling through the United States, benefited by technical discussion and enjoyed his warm hospitality. Along with his activity in the United States he had the unique opportunity to direct the changeover from left-hand to right-hand driving in Panama during World War II. He conducted a training course in highway capacity in Madrid in 1961 and was one of a small group of United States highway engineers who were given an extensive tour through Russia in that same year.

In his professional field, he was active in the Institute of Traffic Engineers, and served as a member of its Board of Direction from 1959 to 1961, as well as President of the Washington Section, ITE, in 1956 to 1957.

Among the many testimonials and honours he received were the Department of Commerce Silver Medal for Meritorious Service in 1960, and the Theodore M. Matson Memorial Award in recognition of the advancement of the science of traffic engineering in 1957. A unique feature of the latter award is that he was the first recipient.

Normann was deeply appreciative of these awards, and sincerely felt undeserving. But a little-known gesture only a few months before his death meant more to him than any other. As chairman of the Highway Capacity Committee as it was reconstituted in 1953, his role had not been easy. The committee included in its membership men of brilliance, but whose brilliance was in some cases matched by their strong wills. As the work on the new manual was nearing completion, some issues became increasingly difficult to resolve, some philosophies were at wide variance with others, and schisms within the committee threatened. Eventually agreement was reached, but few knew of Normann’s concern over what he construed as disharmony for which he felt somehow responsible, and perhaps even a lack of confidence of the committee in its chairman. His doubts and fears were relieved, however, during the 1964 meeting of the Highway Research Board when his committee members, in a carefully staged luncheon, presented him with a very fine watch bearing the engraving ‘to Mr. Capacity’. Such tributes are not uncommon upon completion of assignments or retirement from office, but how often does a committee make such a heartfelt expression of appreciation to a chairman who is continuing in office?

No one is indispensable, but O. K. Normann came close to being irreplaceable.

Reprinted with permission from Australian Road Research, Vol. 2, No. 1, September 1964.

lateral placement of vehicles; distances for passing; speeds of vehicles; spacing of vehicles in the traffic stream; and hill-climbing abilities of larger vehicles. Improved measuring instruments and methods for analyzing the large volume of data were also developed.

The Bureau immediately applied these early results in design and construction programs. One direct application was the analysis of data on transverse positions of vehicles on bridges to assist in the establishment of standards for bridge widths. The basic relations found between vehicle speeds and spacings with varying volumes of commercial traffic and under various alinement conditions have been used and have aided in estimating the likelihood of congestion in tunnels and on bridges.

These early studies in principles of highway capacity were developed and later expanded by O. K. Normann, the “Mr. Capacity” of highway research. He established that the practical or working capacity of a highway is a relative value, being the number of vehicles that a highway can carry without restricting the speed or movement of vehicles to an extent that drivers find intolerable, He also determined that the minimum spacing the average driver allows between his vehicle and a vehicle ahead varies for different highway conditions, as well as the different speeds, and that the theoretical or possible capacities vary accordingly.

A major milestone was the determination of possible capacities, which are still used by design, traffic, and operating engineers today. On analyses of the data assembled by 1941, Normann concluded that “The possible capacities are about 2,000 vehicles per hour for both lanes of a 2-lane highway, 4,000 vehicles per hour for two lanes of a 4-lane highway, and up to 3,600 vehicles per hour for the best 3-lane highways.”[19] The first edition of the Highway Capacity Manual, published in 1950, and the latest edition, published in 1965, provide the identical capacity values for two- and four-lane highways. The value for three-lane, two-way highways has been increased only slightly to 4,000 vehicles per hour in the latest edition.

Both manuals have been widely used, the first having been translated into nine languages. These successively refined evaluations proved to be major tools for engineers developing the details of current highway programs. With them, designers were able to determine the number of traffic lanes and other geometric features that should be provided to accommodate the predicted volumes and types of vehicles in operation at practical speeds on a given highway.

Continuing research in the 1940’s regarding characteristics of individual drivers showed that “persons traveling long distances drive faster and generally have newer cars than local travelers; young persons drive somewhat faster than older persons; men drive somewhat faster than women; and the newer vehicles are driven faster than older vehicles.”[20] Two important conclusions were developed from this series of studies: (1) Highways built to accommodate a high percentage of drivers traveling on long trips should, therefore, be designed for higher speeds than highways on which trips are predominantly short, and (2) there is no justification for a design of high-ways to accommodate speeds in excess of 70 miles per hour under any condition. For about two decades 70 miles per hour has been the advocated upper design speed for the Interstate System and other main highways.

Truck Performance

Data obtained and procedures developed in the analysis of passing practices in the late 1930’s have resulted in guidelines for proper design of two-lane, two-way highways, including the principles of stopping and passing sight distances. Also, the more fundamental data then developed from the hill- climbing or gradeability studies are still in use today in solving problems of highway design and traffic regulation. A basic conclusion of the gradeability study was that for motor trucks even to approach reasonable speeds on grades, grades must be reduced to 3 percent or less. Where such grades are not practical, these data led to the concept of an additional uphill or climbing lane on highways carrying substantial numbers of heavy trucks.

The desirability of design with grades as flat as practical was recognized in the 1954 AASHO design policy guides. The 1956 Interstate standards named maximum design grades of 3, 4, and 5 percent (except in rugged terrain) for design speeds of 70, 60, and 50 m.p.h., respectively.

In 1948, a major study was conducted on the fuel consumption of trucks in relation to their weight and power. Cooperating in the study were vehicle manufacturing and operating groups, the Department of the Army, the Pennsylvania Department of Highways, the Pennsylvania Turnpike Commission, and the Bureau of Public Roads. Important findings developed then, and still in use today, for commercial vehicles were: (1) On any highway section, gasoline consumption and travel time vary in a definite manner with the rate of rise and fall on the highway; (2) gasoline consumption was definitely related to the gross weight and the travel time to the weight-power ratio of the vehicle; and (3) the results were applicable to paved highways and gasoline-powered vehicles in any part of the country.[21]

This automatic traffic recorder was installed on US 240 in Maryland in 1938. It used photoelectric cells to project two beams across the road, and when a vehicle broke the two beams simultaneously, a cumulative count was recorded on a tape.

Stopping Distance

Braking performance of motor vehicles is another important area which the Public Roads researchers have been studying for over three decades. The braking ability of all types of motor vehicles is an essential factor in the safe and efficient movement of traffic on our highways. The sight distance needed for stopping is one of the important geometric design elements.

In 1941, the Public Roads Administration started a broad program of brake performance studies of motor vehicles in everyday traffic. The Advisory Committee on Motor Vehicle Brake Research was formed to direct and guide the program. This was the beginning of a series of studies conducted at 7- to 10-year intervals, the latest of which was completed in 1974. The objectives of these studies were to: (1) Establish a better understanding of the features of vehicle braking performance; (2) determine what levels of brake performance can reasonably be met by vehicles in service; (3) determine the essential elements of reasonable brake regulations; and (4) determine practical means of enforcing brake regulations.

Military Highway Transport

Most highway research was initiated to provide basic information for the development of street and highway design and traffic operations for civilian traffic. Much of this information has been applicable also to military traffic movements, but military convoys required special study. In the spring of 1940, a study to determine the effect of certain highway conditions on the operation of military convoys was conducted by BPR at the request of the Department of the Army.

Ruote 82 in Connecticut winds its way down hill, but note, that there are two lanes for up-hill traffic, allowing slower traffic such as trucks, to use the outer lane and faster traffic to pass on the inner lane.

This study dealt with the effect of alinement, grade, and pavement type and condition on convoy operations. It involved an analysis of the individual speeds of all convoy vehicles and the longitudinal spacings between them. It was found that grades were the only highway feature that materially affected the speed of the military convoys. Highway grades of no more than 6 percent, and preferably 5 percent, would allow practically all military vehicles to maintain speeds of 25 miles per hour as long as the road surfaces were dry and solid. These data help produce better controls for military convoy operations.

Traffic Control Devices

Units along a highway that regulate or advise the vehicle operators are broadly known as traffic control devices. Presently they consist of traffic signals, signs, pavement markings and other markers or delineations along the roadway. The earliest forms were milestones and notched trees. With the advent of the automobile, signing and markings became prevalent in the early 1900’s on principal highways in and around the larger cities. Most of these early devices were developed by innovative people based primarily on their imagination and ingenuity.

The U.S. literature until the mid 1920’s is practically void of records of research on devices used to control traffic. Instead, a unit was conceived, installed, observed and conclusions drawn for future guidance. Naming, numbering, and marking of roads began with motorist clubs, chambers of commerce, women’s clubs and, to some extent, the States. Being individually initiated, the overall result was a colorful chaos. To bring order out of this chaos and thus help to speed the traveler upon his way, the American Association of State Highway Officials agreed in 1925 that the main roads of the country should be marked with standardized information and direction signs. A 1927 manual, developed by a joint board of Federal and State highway officials, published sign details and the numbering system of U.S. highways. This incorporated the numbering system adopted in 1925 for the Federal-aid highway system, which is still in effect.

In the late 1920’s, a uniform route numbering system for U.S. highways was adopted, and these routes were identified with a distinctive shield design.


The various control devices now used for traffic management were mostly developed through private industrial research. However, as early as 1934 the Bureau of Public Roads had begun a series of intersection studies to determine the delay caused by different control methods. A report in that year showed that for an intersection having a total volume of 2,000 vehicles per hour, operation without any mechanical control incurred the least delay to traffic. “Of all the control methods [studied], officer control permitted the fastest movement of traffic, closely followed by the shortest fixed-time control, and traffic-actuated control.”[22] This study not only set the pattern for much future research in this area, but also established methods and procedures for the conduct of studies of the effect of traffic control devices on the safety and efficiency of traffic operations.

The first national rural manual on traffic control signs was issued by AASHO in 1927, and the Manual on Street Traffic Signs, Signals and Markings (urban) was issued in 1930 by the National Conference on Street and Highway Safety. These first efforts at national standards gave details on sign shape, color, legend and mountings. While colors were prescribed, data were lacking as to their suitability. To fill this need, the Bureau of Public Roads undertook a study, completed in September 1933, on the visibility and legibility of several alternative color combinations of signs visible by day and by night, with and without reflector buttons. This study showed that the black on yellow combination was more effective than either black on white or white on black. It also determined for the first time the effectiveness of reflecting buttons in various sizes and spacings.[23]

The first combined rural-urban Manual on Uniform Traffic Control Devices for Streets and Highways was issued in 1935. With each new edition of the Manual, revisions were made based on research conducted by Public Roads, State highway departments, universities, cities, industries, and others. For example, in the 1955 Manual the stop sign was changed from black on yellow to white on red, a standard in use today. This came about as a result of research tests which demonstrated that motorists observe the red sign more effectively than the yellow.[24]

The Interstate manual adopted by AASHO and approved by the Bureau of Public Roads in 1958 for signing and pavement marking of the Interstate System was the result of recommendations based on their joint engineering and psychological research. The use of lowercase lettering became widespread after research indicated that these letters are more easily read by the average driver. There was diversity of opinion on the color and reflectorization of directional signs. A research project was undertaken by Public Roads in 1957, which gave substantial support for the final selection of green as the approved color for directional sign background on the Interstate System.[25]

The general advantages of pavement markings were recognized after their first use in 1911, and they were included in the early manuals to some extent. However, it was not until 1947 that the effectiveness of centerlines, particularly on two-lane, two-way roads, was confirmed through research. The result of studies at 12 locations in seven States showed that centerline markings provided a general improvement in the transverse positioning of vehicles, that vehicles were driven closer to their proper position on the roadway, and that they encroached on the left lane much less frequently where there was a centerline.[26]

Edge lines were added as a standard device in the current Manual on Uniform Traffic Control Devices for Streets and Highways (1971) and are in universal use. Their effectiveness was determined through a comprehensive study of driver behavior as related to pavement edge markings, conducted by BPR in 1957 in Louisiana and the western States. This study showed that edge lines were very effective aids in confining traffic in the normal travel lane and in reducing partial shoulder use.

Freeway Control Systems

By the late 1950’s a substantial number of freeways had been developed in most of our larger cities. As a result of increasing traffic on them, peak-hour congestion became a major problem. Research studies showed congestion during the morning and evening peak travel periods could be greatly reduced by the use of traffic surveillance and addition of further controls. The application of the latest developments in electronic and other technological equipment served to increase operational efficiency and safety.

The Bureau of Public Roads, in cooperation with a number of State highway departments, spearheaded the research and development in this area and has promoted the practical results of its studies. The first project was initiated in 1960 on John C. Lodge Freeway in Detroit. The second one was initiated in 1961 on the Eisenhower Expressway in Chicago and the third in 1964 on the Gulf Freeway in Houston, Texas. After 1965, a large number of States started installing freeway surveillance and control systems, the most extensive of which was the 42-mile loop pilot project in Los Angeles.

Demonstration moving merge control systems have been used since about 1969 on freeway entrance ramps to help ramp drivers enter a freeway in situations where it would be difficult to determine adequate gaps in the freeway traffic. This is an artist’s concept of the moving greenband system, just one type under consideration.

In the late 1960’s, Public Roads continued to expand and refine the technology of freeway operations by undertaking a major travel corridor control investigation in Dallas. This research was a major effort to explore the total corridor concept, including the freeway, service road, arterial street, and surrounding system. Public transportation, bus operations, roadside communications and park-and-ride were modern-day additions to the effort. Public Roads sponsored many other freeway operations studies, among them the improved control of diamond interchanges.

The concepts of integrated surveillance and control are equally important to urban nonfreeway systems. Traffic signals, intersections, and individual aspects have been mentioned. In the late 1960’s, Public Roads stimulated and assisted in “systems” approaches to control traffic for the whole of a certain urban area. One significant output has been the computerized concept of arterial traffic control. Another major project was an urban traffic control research laboratory operating on certain streets in Washington, D.C., which was still active in 1975. This study has had many influences in the broad area of “software” for urban systems, ranging from preprogramed to instantaneous timing. Another major component was an integrated bus priority system. As meaningful results from these studies have become available, nearly all of the States have applied the new technology to other locations.

Lighting

During the late 1930’s when driver behavior, highway capacity, and traffic control were under intensive study, considerable research was devoted to the effect of highway lighting on traffic operations and safety. For several years prior to this time, illuminating engineers had studied means of reducing nighttime accidents, which were proportionally higher then those in the daytime. There were only a few hundred miles of rural highways that were lighted, many of which were temporary installations to demonstrate and experiment with the safety aspects of lighting.

In 1939, the Public Roads Administration and the Ohio Department of Highways concentrated their research efforts on the effect of lighting on the lateral position, passing practices, headways, and speeds of vehicles on rural highways. The results showed that the behavior of drivers operating at night without overhead light differed measurably from that in daytime, but that under artificial light they conformed very nearly to their behavior in the daytime. This indicated the advantage of highway lighting for certain speed-volume-geometric highway conditions.[27]

By the late 1950’s, an increasing number of miles of freeways were being put into operation. Because very little was known about freeway lighting with respect to driver behavior and safety, the Bureau and the Connecticut State Highway Department undertook in 1959 a comprehensive field study on the effect of illumination and delineation. Nine different conditions of illumination and delineation were studied. No significant difference with respect to vehicle speeds, lateral position, and clearances between vehicles under the nine study conditions were noted. In general, it appeared that some benefit resulted from full-level illumination in the deceleration area and that even greater benefit occurred when illumination was combined with roadside delineation. Since then HRB has reported a number of similar studies which generally corroborate the original findings. As a result of these research conclusions, recent lighting installations have been largely located at interchanges and other points of conflict or decision.

The photolog system is a sequential set of photographs, usually taken at one-hundreth mile increments and recorded on a continuous film strip, of the highway and its immediate environment. Each photograph normally provides the viewer with the date it was made, the route, milepoint, and direction of travel, but it can include other data as well. Photologging was developed more or less on the “let’s try it and find out” method by research and operations engineers. Today it is an operational tool in some 40 States. The varied uses to which photologs may be applied include evaluating the adequacy of traffic control devices, providing information for project design, identifying and evaluating high accident locations, acquiring planning inventory data and supplying data for research studies.

Highway Safety

The basic principle of highway safety, as summarized by Thomas H. MacDonald in 1949 in Public Roads magazine, is that maximum safety is provided by designing, building and operating the highway and vehicle to fit the driver’s known capabilities and limitations. This concept recognizes the tremendous variation in age, ability, experience, skill and physical and mental condition of tens of millions of drivers. It prescribes that the highway and the vehicle should accommodate the maximum amount of this variation and assist the maximum number of drivers in their task. It has been the foundation for nearly all useful and productive research in highway safety.

Since the 1930’s, many studies have provided the bases for safer design criteria for highways and vehicles. The highway safety research has included on-highway experiments and observational studies of passing practices, lane position, braking and acceleration capability, grade-climbing ability, and other aspects of driver behavior and vehicle performance.

Control of Access

Research studies over the past several decades have shown that the most effective way to facilitate the driver’s task is to provide him with a highway having full control of access. Such a highway, best exemplified by the Interstate Highway System, prohibits access from abutting property, thus eliminating many of the roadside conflicts which confront the driver. Crossroads are grade separated, eliminating angle collisions occurring at intersections. Wide medians eliminate head-on collisions, and carefully planned interchanges with long speed change lanes minimize rear end and turning collisions. Research undertaken in the early 1950’s demonstrated that full control of access reduces accidents, injuries and fatalities by 50 percent or more.[28]

Control of access also has safety benefits on two-lane highways. A study on a two-lane rural highway with average daily traffic of about 8,000 vehicles showed that, if the number of at-grade intersections and roadside business driveways per mile were increased a hundredfold, the expected accident rate would increase approximately 14 times.[29]

Roadside Safety

One of the major accident types is the single vehicle run-off-the-road accident. Often the vehicle strikes an object such as a tree, rock, sign, or guardrail, resulting in injury or death to the vehicle’s occupants. Although studies designed to relieve this problem began in the 1920’s, the most significant advances have occurred in the recent past.

In December 1966 a short-range Public Roads research program was established for quick-payoff concepts and devices to substantially reduce the severity of single vehicle collisions with fixed roadside structural obstacles. The primary objective of this program was the development of devices that would protect vehicle occupants against severe injury during impact with rigid gore structures, bridge piers, sign bridge supports, and open areas between twin bridges. The studies emphasized the use of existing technology in such fields as impact absorption barriers, new and improved types of redirectional barriers, vehicle entrapment devices, cable supported structures, and improved computer simulation techniques. The impact acceptance criterion was tentatively set at a maximum deceleration limit corresponding to a 10-foot stopping distance from 60 miles per hour for passenger vehicles, a survivable situation for shoulder and lap-belted vehicle occupants.

Several types of impact attenuators, or crash cushions, were successfully developed and tested. These include the steel drum barrier (an assemblage of empty 55-gallon drums), a water cushion (an array of waterfilled vinyl cells), and clusters of sand-filled plastic drums. More than 3,000 such devices have been installed on the highways to date, with evidence that they are extremely effective in reducing severity of single vehicle collisions.

The overall concept and application of breakaway sign supports were also developed primarily during the past 10 years. Research efforts began with the inception of multi-State, pooled-fund studies of the breakaway sign and luminaire support concept. The research included full-scale field experiments and crash testing of vehicles into prototype breakaway sign supports. In addition, much work was also developed or verified through the use of computer simulation and theoretical studies. Private industry also joined in this effort and produced a number of breakaway concepts.

Motorists on I-84 in Connecticut are protected from a possible fatal injury at this median strip bridge pier by a guardrail, the approach end of which is buried in the ground as the result of research into the causes and severity of accidents.

Reducing Skid Accidents

Skid accidents, which result from slippery pavements, especially when wet, are of such severity that they have received considerable attention in research and safety programs for more than four decades. Major skid factors include: water-on-pavement conditions, tires and their treads, pavement surface characteristics, aggregates and mixes used, pavement age and surface wear, speeds and vehicle operations, and highway geometries. There have been many contributing research studies on individual factors and their various combinations, but the complex interactions involved have inhibited easy and low-cost solutions.

In the late 1920’s Iowa State College began measuring road skid resistance with a towed trailer. Extension of these studies were reported in 1934; this proved to be a milestone report on the theory of skidding and means for measuring. Within the next few years many States, universities, industries and other groups were engaged in related research. Both field measurements and laboratory studies on aggregates, mixes, their polishing features, and means for measurement were under study. The Hybla Valley and other test roads included these features. In the 1950’s hydroplaning was identified as one of the skid conditions of much concern on airports as well as roads. Also the first steps toward grooving of concrete pavements were being made to reduce slipperiness.

Skid prevention attained nationwide attention and emphasis at an important international conference in Charlottesville, Virginia, in 1958. This conference dealt with many new research and development studies to attain better skid resistance and served to redirect the research that has continued since then. Because of the magnitude and severity of the problems in recent years, FHWA has instituted a large scale research project under its federally coordinated program. It deals with friction requirements, pavement materials and surfaces, accident and cost effectiveness studies, and a national program to standardize the skid measurement process.

Early results from the research have provided: (1) Three regional test centers and the calibration of dozens of State skid trailers; (2) a standard two-wheel test trailer and testing procedure; (3) guidelines to design appropriate variable message signs under adverse weather conditions; and (4) improved laboratory methods to select skid-resistant materials.

The Accident Prone Driver

Accident research studies have often been deficient because of limited data and the difficulties in obtaining the data. Despite these limitations, key accident studies have provided much useful information that has dispelled certain myths and provided support for research and practical day-to-day programs.

One such myth prevalent during the 1940’s and 1950’s that has been exploded is that of the so-called “accident prone driver.” A study in 1962-63, showed that “. . . in any one year 0.5 percent of the drivers have two or more reported accidents and that they account for 13.9 percent of all accidents.”[30] The accident prone concept suggested that these drivers be removed from the road, thereby reducing accidents by 13.9 percent. Such a conclusion, however, was erroneous. Nearly all of those who were “accident repeaters” in any one year became so by chance alone; the research revealed that 87 percent of the so-called accident repeaters would not have even one reportable accident the second year.

This trailer measures pavement skid resistance. A prescribed film of water is spread on the pavement junk prior to brake application. Instruments on the trailer measure wheel motion and traction during the entire braking cycle.

Speed and Accidents

For many years, highway professionals have known that highway safety could best be achieved by minimizing traffic turbulence and speed variance. Cooperative research by the Bureau of Public Roads and several State highway departments in the late 1950’s and 1960’s verified these views. Both two- and four- lane main rural highways without control of access, as well as urban and rural freeways, exhibited the lowest accident involvement rate for vehicles operating near the average speed for all traffic. For vehicles traveling at both lower and higher speeds, the involvement rate increased sharply. For example, at speeds 15 miles per hour below the average speed on Interstate highways, the accident involvement rate was five times as great as for vehicles operating at the average speed. When an accident occurred, the severity increased with speed, particularly at travel speeds above 60 miles per hour. The chance of being killed in an accident was found to be about four times as great at travel speeds of 73 miles per hour or higher than at 60 miles per hour.[31] These findings have been particularly useful in stimulating official and public awareness of the effects of speed.

Driver and Vehicle Characteristics

In addition to speed, studies showed that other characteristics for groups of drivers and vehicles were directly related to accidents on two- and four-lane main rural highways without control of access. Included were age, sex, and residence of driver; type, age, and horsepower of vehicle; and the seat location of individuals within passenger cars.

It was found that local drivers tend to have higher accident involvement rates than out-of-county drivers, particularly at night, and that drivers of passenger cars with low horsepower have higher involvement rates than drivers of cars with higher horsepower, regardless of several other variables studied, including travel speed.[32]

Highway Safety Study

In the late 1950’s in response to a congressional requirement, a comprehensive study of highway safety was undertaken by the Bureau of Public Roads. The report included a discussion of the traffic accident problem and its setting, review of the highway transportation system, evaluation of current highway safety activities, and a brief description of an adequate highway safety program. Results of this study were used to expand the Federal role in highway safety, including creation of a special National Highway Safety Bureau in the Federal Highway Administration, and eventually creation of the National Highway Traffic Safety Administration. Also, a nationwide computerized system was developed for identifying drivers with suspended or revoked licenses attempting to obtain drivers’ licenses in other States.

Other Safety Research

Other important studies have delineated the cost of traffic accidents in considerable detail, permitting development of useful benefit-cost analyses of various safety problems. Studies related to sizes and weights of trucks in connection with the speed study mentioned earlier have been useful in justifying a combined reduction in speed to 55 miles per hour to save fuel and lives coupled with a 10 percent increase in allowable axle and gross weight limits on the Interstate System to further save fuel and reduce freight haulage costs.

Since 1970 research studies have provided methods for designing improved guardrails, median barriers and bridge rails, criteria for safer roadsides, improved highway lighting, improved traffic signal configurations, warrants for enhancing safety at rail-highway grade crossings, use of earth heat to melt snow, crack detectors for bridges, and driver aids in fog. Also, there has been increased attention to pedestrian and bicycle safety, including the development of bicycle safe grate inlets for city streets, wide dissemination of a bikeway state-of-the-art and evaluation of several pedestrian countermeasures.

The in-motion vehicle weighing and measuring system obtains a description of the traffic traversing the area — number of vehicles, speed and classification (car or truck). Then the weight and dimension subsystems add a complete description of the truck portion of the traffic, including height, width, length, axle spacing and weights, vehicle type, loading, etc. Installations such as this are being used in Texas and as demonstrations in several other States.

Human Factors Engineering

During World War II, behavioral scientists pooled their talents to help military systems engineers develop vehicle operating systems to be used by military trainees with little or no engineering or scientific training. Fitting machines and environments to man’s capabilities and limitations resulted in the emergence of new disciplines known as engineering psychology or human engineering. The relationship between highway design and operational techniques became a major concern of the BPR’s research program.

In the 1950’s a few engineering psychologists began to translate the findings of classical experimental psychology to tasks that confronted motorists operating their automobiles between towns and cities. They also attempted to determine the basic skills required of drivers. The early models of driver behavior are now recognized to be overly simplistic, but they alerted traffic engineers and other road authorities to the fact that most people using highways differ from highway designers and traffic engineers in their perception of the driving situation.

The 1950’s were characterized by three major approaches for incorporating driver characteristics into highway designs and traffic operations techniques. The first was an attempt to scale the difficulty of specific types of driving situations by measuring driver stress. The second was performance oriented research characterized by measurement of the driving patterns of large numbers of drivers on a variety of road environments and was an extension of studies undertaken by the Bureau of Public Roads in the 1930’s and 1940’s. The third was epidemiological research, aimed at establishing patterns of responses of drivers which would permit officials to improve their selection techniques or develop remedial training programs. Incorporation of driving behavior into highway design and operational techniques was a major concern of the BPR’s research program.

Behavior research in the 1960’s began to mature. A programmatic approach to determining driver capabilities and limitations was begun in the late 1950’s when a team of behavior scientists began to conduct sophisticated analyses of the perceptual environment faced by motorists traversing high-speed roadways on potential collision courses with other vehicles and obstacles. Fundamental work on road tracking, speed sensing, car following, overtaking and passing, object avoidance, sign reading, and path finding were performed jointly in BPR, several universities and by private contractors. Most notable among the collective agencies was the portion of a program of automated or semiautomatic vehicular control conducted by Ohio State University’s Industrial Engineering Department.

In the mid-1960’s, fundamental work aimed at delineating driving skills abated in favor of development of driving aids systems to assist motorists. Several such systems were identified to overcome high accident situations and basic limitations of motorists; developmental research study of them continues. Some of the traffic control and information systems were:

  • A passing aid system to assist drivers in overtaking and passing on two- and three-lane highways.
  • A merge control system to assist drivers in making complex decisions at freeway entrance ramps.
  • An experimental route guidance system for assisting motorists in finding their way to unfamiliar destinations.

The knowledge developed through these programs and efforts of behavioral scientists have been inte-grated into modern highway design and traffic engineering.

Environmental Factors

Hydraulics and Hydrology

Hydraulics and hydrology, simply stated, are sciences developed to control the deleterious effects of water. There is adequate evidence that ancient cultures constructed amazingly elaborate conduits, aqueducts and other drainage structures to control the flow of water. In the early development of rail transportation, bridges and culverts were certainly an integral part of the design. With the advent of the automobile and accelerated highway construction programs, drainage and water problems became more critical. Bridge damage, washouts, and flooded roadways became intolerable, and paved surfaces required good drainage to reduce hazardous conditions for the movement of traffic and to prevent pavement failures. Progress in the control of water flow and drainage was slow, and in the early days, designs were based on judgment without uniform policy guidelines or well-developed engineering technology. It was common to size highway drainage structures by using a formula developed by Professor Talbot of the University of Illinois for the design of railroad structures being built in the western States in the 1890’s. That formula was the first attempt to provide a rational approach to drainage design; it was crude but an improvement over previous methods. However, it is interesting to note that in 1961, despite the fact that by then more sophisticated techniques had been developed, a survey by the American Society of Civil Engineers concerning drainage practice showed 12 States still using Talbot’s formula.

In the early 1920’s, David Yarnell pioneered in experimental hydraulic research with his study on the flow of water through culverts based on extensive full-scale tests conducted at the University of Iowa. Unfortunately the report covered only culverts flowing full. Most culverts do not flow full, and the factors affecting them are significantly different, Yarnell went on with experiments on flow resistance caused by pile trestles and flowover embankments. His work on rainfall intensity-frequency relations was published by the U.S. Department of Agriculture in the mid-1930’s and for many years was widely used across the country in determining the rainfall rate to apply in the use of the rational method of estimating runoff.

Highway engineers had also been concerned about erosion of highway ditches and slopes, but it took the demonstration projects of the Soil Conservation Service in the late 1930’s to show how flattening of slopes and rapid establishment of sod could control erosion. About the same time the Soil Conservation Service also began setting up experimental watershed stations in several locations to study rainfall-runoff relationships as affected by land use practices.

During World War II, the Public Roads Administration became responsible for constructing satellite airfields and found that data on rainfall and runoff were inadequate. Using rainfall simulators provided and operated by the Soil Conservation Service, a series of experiments on overland flow was conducted on paved and grass surfaces. The data analyzed and reported by Public Roads in 1944 enabled engineers to estimate for the first time flows over overland routes. This paper has become a standard reference on the subject.

After World War II greater interest developed in the application of hydraulic engineering principles to highway design. One of the first applications of the overland flow equations was in the design of a storm drain system in Chicago. From this, Tholin and Kiefer developed the “Chicago Hydrograph Method” which involved routing runoff through the drainage system.

The Public Roads Administration developed a highway engineering drainage manual. The manual provided step-by-step instructions, together with numerous nomographs and charts based on research data, to facilitate the solution of hydraulic equations for design of culverts, open channels, and storm drains. The manual was widely used in draft form by the States in their highway programs and on most of the toll roads because it provided uniform and improved drainage design techniques. It was never formally published and has since been superseded by the series of Hydraulic Engineering Circulars published by BPR beginning in 1960. The original manual and the Circulars have been used as text for training courses and design practices throughout the United States and abroad.

Following a series of disastrous floods, the Iowa State Highway Commission began a research project in 1948 at the University of Iowa to investigate scour around bridge piers. Scour is the severe erosion of firm supporting soils around bridge piers and is the principal factor in the failure of bridges. The results of that investigation enabled bridge designers to estimate the probable depth of scour during floods and provide adequate supports.

A formal hydraulics research program was instituted by Public Roads in 1949 in recognition of the need for systematic study of hydraulics and hydrology as an integral part of highway engineering. The agency conducted in-house research principally in the field of hydrology (analyses of runoff data); supervised contract research at a number of university and Government laboratories; and monitored research undertaken in several States.

In the 1950’s major research was continued on bridge scour at the University of Iowa; culvert design was studied at the National Bureau of Standards; backwater caused by bridges was studied at Colorado State University; head, or energy, losses in storm drain junctions were studied at the University of Missouri; resistance losses in concrete pipe were studied at the University of Minnesota; and urban stormwater runoff was studied at the Johns Hopkins University. All of the research was undertaken as a result of expressed needs evolving from field experience. All reports provided design data and methods readily used by highway designers.

Another field of research begun in the 1950’s in a few States was the Cooperative Highway Program of the U.S. Geological Survey for measurement and analysis of runoff from small watersheds, preferably for at least 10 years in each State. The objective was to estimate the magnitude and frequency of peak flows from watersheds generally under 25 square miles in area. The program has grown to include at present about 30 States with an annual total expenditure of about $1 million.

To protect this cut slope from erosion during highway construction, the first section of the cut already has new grass growing. The second section has been seeded and mulched, while earthwork is still in progress on the third section.

In succeeding years hydraulic research for the Federal Highway Administration has been conducted on hydraulic roughness of corrugated metal pipes by the U.S. Waterways Experiment Station, on use of riprap[N 1] to minimize scour at culvert outlets, on unsteady flow in a pipe at Colorado State University, and on design of riprap lining for open channels at the University of Minnesota, not to mention numerous smaller studies undertaken elsewhere, usually in cooperation with a State highway department.
  1. Riprap is a layer, facing or protective mound of stones, concrete, or other material, randomly placed to prevent erosion, scour or sloughing of a structure or embankment.

Roadside Development

One of the earliest publications to identify the elements of and the need for improved road environments was by Louis C. Haupt in 1891.[33] His theme was that good roadside development is a move toward better roads. Little further attention was given to this subject until the early 1930’s when AASHO and HEB organized their first roadside committees; at that time only 10 States were represented. Even then, where roadside improvement was performed, it was done years after the highway was constructed.

Finally, however, the research and development work of the committees, with support from the Bureau of Public Roads, received progressively greater attention, and for some 20 years adequate roadside treatment has been a recognized part of design and construction. Roadside development research has covered such areas as erosion, esthetics, rest areas, resource conservation, planting and vegetation management.

The large number of studies and reports in this area of research was oriented mostly to local problems and conditions. A 1972 study showed that roadside development efforts still seem to be very much a local matter. Efforts to broaden this activity on a national scale have thus far met with little success.

Vegetation Management

A major part of roadside development is vegetation management. For 50 years, most of the work was directed toward the prevention of erosion. While erosion control is still a primary goal, for more than a decade the Federal Highway Administration and most States have recognized other values in vegetation. Research was directed to the selection of proper vegetation to encourage the establishment of certain species of wildlife. In addition, the esthetic value of vegetation is recognized for its potential in maintaining balance between the highway and the natural environment.

During the 1920’s and 1930’s, maintenance of roadside vegetation was usually done by hand. Although research had greatly improved mowing equipment and techniques by 1960, breakthroughs in the use of herbicides greatly altered roadside maintenance techniques. As this work progressed, however, concern over the long-term effect of herbicides created new interest in development of dwarf ground covers.


Today, with rising maintenance costs and public concern for our environment, the need is recognized for a highly refined roadside development program. Research and implementation efforts by FHWA, State agencies and others are focused on reduced mowing, refinement of herbicide technology, and the development and selection of vegetation that will reduce maintenance costs and enhance the overall environment.

Recent Environmental Research

In the past 10 years, two areas of interest have been receiving considerable attention: (1) Firm criteria for esthetic enhancement of the view from the highway are evolving from key studies and are having a major impact on new highway design and upgrading of existing roads, and (2) the need to protect the wildlife environment has opened a broad vista of studies addressing the compatibility of small and large game, birds and fish with the highway environment.

Work on erosion control, roadside vegetation, management transportation economics, and hydrology were the forerunners of today’s environmental research. This was highlighted by the establishment of a formal research program in environmental design and control. In response to the National Environmental Policy Act of 1969, the research is directed to understanding the interaction between highways and the environment and developing technology to protect the environment.

Since 1970 this research has dealt with a wide range of environmental elements including air, noise and water quality; social and economic effects; roadside rest areas; sewage treatment; vegetation management; esthetics; water runoff; de-icing chemicals; spills of hazardous materials; erosion control; and wildlife.

In summary, roadside development has progressed during the past 50 years from a somewhat haphazard approach to a highly sophisticated and comprehensive science.

Nationally Coordinated Programs

In the early 1960’s several significant events led to dramatic changes in the concept and activity of highway research. Enlarged research and development legislative authority was enacted in 1962 and made effective beginning with fiscal year 1964. There were growing problems for highway transportation along with population growth, urban concentration, and changing national priorities and goals.

One major legislative change was the requirement that 1½ percent highway planning and research (HPR) funds must be used for planning and research purposes. No longer was there an option of using this money for construction. This requirement ensured the strength and vitality of federally aided planning and research programs in the States. Another event was the organizational separation of planning and research in the Bureau of Public Roads late in 1961. Research and development finally became a formal separate program in the Bureau.

At the same time, there was high interest and strong urging from scientists and technologists in the Department of Commerce and elsewhere to push the Bureau of Public Roads beyond its historic materials and physical research and undertake so-called “soft” research in operations, human factors, systems analysis, safety, society and the environment. Research programing and management could no longer be a loose collection of a number of separate, isolated, totally unrelated studies. The problems were too complex and the resources too small. Comprehensive, integrated and balanced research was the only answer. Starting in 1963, a multidisciplined task force reviewed a wide range of problems and opportunities for solution in the human, physical, environmental and public policy fields. This culminated in the creation of “A National Program of Research and Development for Highway Transportation,” which officially began in 1965.

If the National Program could be summed up in three words, they would be coordination, concentration and flexibility. The Federal Highway Administrator called it a program to coordinate and concentrate our efforts on the most urgent problems with flexibility, responsiveness and recognition of local and regional problems.[34] The program consisted of detailed work plans and problem statements for each; it served as a stimulus and guide for R&D efforts to develop and apply solutions rapidly.

The National Program was directed to three issues of highest concern—highway safety, urban transportation, and reduction in the costs of construction and maintenance. These issues were important because of the considerable human losses and $10 billion annual costs of accidents; the increasing problems of congestion, pollution and adverse socioeconomic impacts; and the critical need to optimize technological processes and cost.

In 1970 when the National Program was 5 years old, FHWA initiated a new, nationally designed and coordinated program for highway research and development called the Federally Coordinated Program of Research and Development in Highway Transportation (FCP). The FCP was specially designed to meet the needs of the customer–highway program managers and operating personnel in State and local agencies as well as the general public. It was also designed to be particularly responsive to urgent new problems in the environment, energy and resource conservation, utilization of waste, the integration of multitransportation goals, and increasing the efficiency of the present system.

Research Into Practice

For a number of years, there had been a strong movement throughout the highway community to expand and strengthen the practical utilization of research. The problems of highway transportation were becoming too large, and it seemed that too little was being accomplished in putting the information gained from research to work.

Both AASHO and FHWA acted in the late 1960’s to formalize implementation. It was suggested that each State name a high-level implementation coordinator and each FHWA region organize an implementation committee. Both of these suggestions were carried out, but a national program still did not develop. It seems that plans, procedures and resources were lacking. There still was no formal national focus. Skilled promotion was needed to gain managerial support, to overcome inertia and reluctance to change, and to secure the essential ingredients of public acceptance.

In response to this need, FHWA in 1970 created a separate Office of Development. Thus a formal, substantial program was begun and within a couple of years, the resources, plans, procedures and programs were prepared. Development/implementation has become a full and substantial partner in the business of “research to practice.”

REFERENCES

  1. Bureau of Public Roads Annual Report, 1910, p. 769.
  2. Federal-Aid Highway Act Amendments of 1963, S.R. 552, 88th Cong., 1st Sess., p. 6.
  3. L. Page, Salutatory, Public Roads, Vol. 1, No. 1, May 1918, p. 3.
  4. Bureau of Public Roads Annual Report, 1922, p. 36.
  5. Ideas and Actions: A History of the Highway Research Board, 1920–1970 (Highway Research Board, Washington, D.C., 1970) p. 29.
  6. Subgrade Investigations Begun by Bureau of Public Roads, Public Roads, Vol. 2, No. 24, Apr. 1920, p. 29.
  7. A. Wintermeyer, Adaptation of Atterberg Plasticity Tests for Subgrade Soils, Public Roads, Vol. 7, No. 5, Jul. 1926, pp. 119–122.
  8. E. Kelly & P. Hubbard, Rationalisation and Simplification of Test Requirements for Liquid Asphaltic Materials, Public Roads, Vol. 13, No. 6, Aug. 1932, p. 89.
  9. J. Pauls & H. Rex, A Test for Determining the Effect of Water on Bituminous Mixtures, Public Roads, Vol. 24, No. 5, Jul.–Aug.–Sept. 1945, pp. 115–129.
  10. R. Bogue, Chemistry of Portland Cement (Reinhold Publishing Corp., New York, 1947) pp. 10–14.
  11. W. Halstead, The Behavior of Red Lead-Iron Oxide Primers When Exposed Directly to Weathering, Public Roads, Vol. 29, No. 9, Aug. 1957, p. 213.
  12. B. Chaiken, Abrasion Resistance of Bridge Paints for Use in Alaska—Field and Laboratory Tests Evaluated, Public Roads, Vol. 33, No. 10, Oct. 1965, p. 201.
  13. J. Pauls, Reinforcing and the Subgrade as Factors in the Design of Concrete Pavements, Public Roads, Vol. 5, No. 8, Oct. 1924, pp. 1–9.
  14. Prestressed Concrete Pavement Demonstration at Dulles International Airport, Public Roads, Vol. 37, No. 1, Jun. 1972, pp. 16–21.
  15. I. Corvi & J. Houghton, Service Lives of Highway Pavements—A Reappraisal, Public Roads, Vol. 36, No. 9, Aug. 1971, p. 192.
  16. W. Hatt, The Field of Highway Research, Public Roads, Vol. 4, No. 5, Sept. 1921, p. 17.
  17. Policy On Geometric Design of Rural Highways (American Association of State Highway Officials, Washington, D.C., 1965) p. 152.
  18. A. Taragin, Driver Performance on Horizontal Curves, Public Roads, Vol. 28, No. 2, Jun. 1954, pp. 27, 28.
  19. O. K. Normann, Highway Capacity, Proceedings, 21st Annual Meeting, Vol. 21 (Highway Research Board, Washington, D.C., 1941) p. 379.
  20. Public Roads Administration, Highway Practice In the United States of America (Federal Works Agency, Washington, D.C., 1949) p. 65.
  21. C. Saal, Time and Gasoline Consumption In Motor Truck Operation, Research Report 9-A (Highway Research Board, Washington, D.C., 1950) p. 16.
  22. E. Holmes, The Effect of Control Methods on Traffic Flow, Public Roads, Vol. 14, No. 12, Feb. 1934, p. 240.
  23. F. Mills, The Comparative Visibility of Standard Luminous and Nonluminous Highway Signs, Public Roads, Vol. 14 No. 7, Sept. 1933, p. 111.
  24. G. Sessions, Traffic Devices: Historical Aspects Thereof (Institute of Traffic Engineers, Washington DC 1971) p. 121.
  25. Id., p. 123.
  26. A. Taragin, The Effect of Driver Behavior of Center Lines on Two-Lane Roads, Proceedings, 27th Annual Meeting, Vol. 27 (Highway Research Board, Washington, D.C 1947) p. 273.
  27. W. Walker, Effects of Highway Lighting on Driver Behavior, Public Roads, Vol. 21, No. 10, Dec. 1940, p. 187.
  28. C. Prisk, How Access Control Affects Accident Experience, Public Roads, Vol. 29, No. 11, Dec. 1957, p. 266.
  29. J. Cirillo, R. Beatty, S. Dietz, S. Kaufman & J. Yates, Interstate System Accident Research Study I (Federal Highway Administration, Washington, D.C, 1970) p. I-1.
  30. D. Solomon, Highway Safety Myths, North Carolina Symposium On Highway Safety, Vol. 2 (the University of North Carolina Highway Safety Research Center, Chapel Hill, 1972) p. 41.
  31. D. Solomon, Accidents On Main Rural Highways (Federal Highway Administration, Washington, D.C, Reprinted 1974) p. 12.
  32. The Federal Role in Highway Safety, H. Doc. 93, 86th Cong., 1st Sess., pp. 71–83.
  33. Roadside Development—Evaluation of Research, NCHRP Report No. 137 (Highway Research Board, Washington, D.C, 1972) p. 45.
  34. R. Whitton, Preface to A National Program of Research and Development For Highway Transportation (Bureau of Public Roads, Washington, D.C, 1965) p. 11.