• Author / Uploaded
• askarah

• Categories
• Toma de decisiones por consenso
• Superficie de la carretera
• Hormigón
• La carretera
• Evaluación

Citation preview

1-32

Systems for Design Of Highway Pavements Final Report

Prepared For National Cooperative Highway Research Program Transportation Research Board National Research Council

Prepared By Michael I. Darter Harold Von Quintus Emmanual B. Owusu-Antwi Jane Jiang

Champaign, Illinois May 1997

Acknowledgments This work was sponsored by the American Association of State Highway and Transportation Officials, in cooperation with the Federal Highway Administration, and was conducted under the National Cooperative Highway Research Program administered by the Transportation Research Board of the National Research Council.

Disclaimer The opinions and conclusions expressed or implied in the report are those of the research agency. They are not necessarily those of the Transportation Research Board, the National Research Council, Federal Highway Administration, American Association of State Highway and Transportation Officials, or of the individual States participating in the National Cooperative Highway Research Program.

Report Preparation NCHRP Project 1-32 was conducted under the direction of Michael Darter, Harold Von Quintus, and Emmanuel Owusu-Antwi. Other team members include Yan (Jane) Jiang, Brian Killingsworth, Jerry Daleiden, Wayne Seiler, and Kenneth McGhee. ERES Consultants, Inc. served as the prime and Brent Rauhut Engineering, Inc. served as a subcontractor on this project.

i

Table of Contents Abstract ...................................................................................................................................... iv Summary ......................................................................................................................................v Chapter 1 Introduction and Research Approach .......................................................................1 Problem Statement ...................................................................................................1 Research Objective ..................................................................................................2 Research Approach ..................................................................................................5 Chapter 2 Findings.........................................................................................................................5 Success of Pavement Design Catalogs/Standards In Europe...................................5 Pavement Design Methods Versus Presentation Formats .....................................11 Catalog of Current State Pavement Design Features.............................................12 Feasibility study.....................................................................................................13 The Consensus Meeting.........................................................................................19 Catalog of Recommended Pavement Design Features ..........................................21 Prototype Knowledge-Based Expert System (KBES) ...........................................25 Chapter 3 Interpretation, Appraisal, Applications...................................................................27 Catalog of Current State Pavement Design Features.............................................27 Catalog of Recommended Pavement Design Features ..........................................27 Supplemental Prototype KBES—Designer ...........................................................43 Chapter 4 Conclusions and Suggested Research ......................................................................53 Conclusions............................................................................................................53 Suggested research.................................................................................................53 References.....................................................................................................................................58 Appendix A:

Summary of European Design Catalogs

Appendix B:

Minutes of NCHRP Project 1-32 Consensus Meeting

ii

Table of Contents (continued) List of Figures Figure Page 1 Percent respondents indicating the prospective users of a catalog .................................15 2 Percent respondents checking the categories of the ways of using a catalog .............................................................................................................................15 3 Percent respondents indicating the potential advantages................................................16 4 Percent respondents selecting the disadvantages categories ..........................................17 5 Example structural design details for flexible pavement................................................34 6 Example structural design details for rigid pavement ....................................................41

List of Tables Table 1 2 3 4

Page Calculation of seasonally adjusted effective resilient modulus for subgrade.................31 Site condition design cells and alternatives for flexible pavement catalog ....................33 Determination of seasonally adjusted effective subgrade k-value..................................38 Site condition cells and design alternatives for the rigid pavement ...............................40

iii

Abstract

Three practical and useful products were developed for the state highway agencies in this study. The Catalog of Current State Pavement Design Features provides a comprehensive summary of the design practices for flexible and rigid pavements for each state highway agency in the U.S. The Catalog Of Recommended Pavement Design Features includes (1) recommended (good practice) design features for highway engineers, administrators, and others in a format that is easy to use and understand, (2) a model catalog presentation format for potential usage by agencies, and (3) recommended consensus on many design features for pavement design for varying site conditions. Many European countries have developed and successfully used design catalogs for years. The prototype Knowledge-Based Expert System (KBES) supplements the catalog by (1) providing interactive guidance to the designer in obtaining design inputs for the catalog site conditions and (2) a rapid catalog database search and presentation medium that represents the paper catalog and that quickly and efficiently identifies feasible design alternatives for a given set of site and design conditions.

Keywords: flexible pavement, rigid pavement, pavement design, subdrainage, design catalog.

iv

Summary

The pavement design dilemma (as stated by a former state pavement design engineer): There is a large amount of knowledge about pavement design available that is not being used. Much of this knowledge resides with experienced engineers, in FHWA and State highway agency manuals, pavement performance databases, and in research publications. A large percentage of experienced engineers and contractors have retired leaving pavement design and construction to far less experienced engineers. New engineers entering the highway field often do not have the barest knowledge about pavements. Technical support from industry has been decreased. Existing design manuals and procedures address thickness design, but do not directly consider many important details that affect performance and future rehabilitation needs. Many of these “details” are specified in agency “standards” which are seldom improved or updated consistently. Is there a way to bring more of the available knowledge directly into the pavement design process, especially for relatively inexperienced engineers? This study has produced three practical and useful products for the state highway agencies to address the above pavement design dilemma. These include a Catalog of Current State Pavement Design Features, a Catalog Of Recommended Pavement Design Features, and a prototype Knowledge-Based Expert System (KBES). The development of the design feature recommendations through a unique consensus building process and the potential uses of each of these products by state highway agencies are summarized. Many European countries use pavement design catalogs. The catalogs, or pavement standards as they are sometimes called, were initially developed in the 1960s and 1970s. They v

started out as very simple documents of only a few pages showing diagrams of pavement sections as a function of traffic and often subgrade support. These simple catalogs or standards have since evolved into very comprehensive documents. The presentation format of the European catalogs is such that they are relatively easy to use by practicing engineers. There is no doubt that pavement design catalogs have become highly popular and successful in many European countries. In addition to the national highway system, they are actually being developed for a variety of other roadways such as cities, ports, and toll roads.

Pavement Design Methods Versus Presentation Formats. There is a difference between a pavement “thickness design method” and a pavement design “presentation format" by which a design method is made available for use by practicing engineers. They are completely different things, and this is crucial to understanding the potential value of a pavement design catalog which is a “presentation format”, not a design method. A catalog presentation format can be used to include a combination of engineering experience, road test results, and mechanistic based thickness design procedures.

Catalog of Current State Pavement Design Features. A Catalog of Current State Pavement Design Features was prepared that provides a highly informative and practical guide on the details of the design practice for flexible and rigid pavements in the U.S. The catalog first and most importantly provides information in the form of a large Synoptic Tables Of State Pavement Design Features. This section provides for the first time for both flexible and rigid pavements a comprehensive summary of nearly all state’s design practices.

vi

Second, the catalog provides, for each state highway agency (SHA), a State Factorial Design Matrix of Pavement Design Features that gives design layer thicknesses for a wide range of traffic, subgrade, and climatic site conditions for major types of main highway pavement constructed by the SHA. Third, the catalog provides a Catalog of Key Design Features by Climatic Region and Pavement Type.

Feasibility study. A detailed evaluation was conducted into the feasibility of developing a comprehensive catalog of recommended design features for both flexible and rigid pavements and a supplemental prototype knowledge-based expert system (KBES). It was concluded that it is feasible to properly develop a such a catalog and a supplemental KBES having specific objectives and scope. This conclusion was based upon the positive ratings and general comments of respondents (State highway agencies, FHWA, industry), many years of successful European experience, and the recent development of catalog-like procedures by several states (Washington, New York, Missouri).

Consensus meeting. Input from all sectors of the highway industry was required to develop the catalog. Thus, a resource group of experienced pavement engineers from all sectors of the industry was assembled. The consensus group discussed and reviewed the draft Catalog of Recommended Pavement Design Features. The meeting produced a consensus on many design features and performance criteria. The most valuable aspect of the meeting was the many interesting and beneficial improvements proposed by the highly experienced and diverse

vii

resource group. This approach was so successful that it should be considered for use on other studies where expert human experience is important. Catalog Of Recommended Pavement Design Features. This document presents a catalog of “good practice” recommendations for design features of highway pavements for highway engineers, administrators, and others in an easy to use format. Guidelines are provided for three main site conditions: traffic loadings, subgrade support, and climate. Based on these inputs, design feature recommendations are provided in “design cells” including the pavement cross section, structural design, materials, and other features required to meet minimum performance requirements. This pavement design catalog is a relatively simple but effective mode of presentation of an underlying pavement design methodology that includes both empirical and mechanistic components. The catalog includes: •

Recommended (good practice) design features for highway engineers, administrators, and others in a format that is easy to use and understand.

•

A model catalog presentation format for potential usage by agencies.

•

Recommended consensus on many design features for pavement design for varying site conditions. The catalog provides recommendations on design features for highways ranging from heavily

trafficked Interstate and primary highways to secondary highways. The specific pavement types included in the catalog are as follows:

Flexible Pavements •

Asphalt concrete pavement with a crushed aggregate base. viii

•

Asphalt concrete pavement with an asphalt treated base.

•

Asphalt concrete pavement with a cement treated base.

•

Asphalt concrete full-depth pavement.

Rigid Pavements •

Jointed plain concrete pavements (JPCP).

•

Jointed reinforced concrete pavements (JRCP).

•

Continuously reinforced concrete pavements (CRCP).

The catalog, as it stands, is not intended for direct use in pavement design by an agency. Design feature recommendations are provided in the form of acceptable ranges within each cell of site conditions. The catalog will, however, provide recommendations that are adequate to identify design features for flexible and rigid pavements that will help guide highway authorities in selecting suitable and reliable designs. In general, the catalog provides recommended design features that meet specific minimum performance requirements for a given set of site conditions. The catalog recommendations are based on many sources, however, the most significant source is the recommendations achieved by consensus of a resource group of pavement design experts from Federal, state, industry, consulting, and academia. Contributions were also made by the NCHRP based on reviews of the documents. In addition, use was made of current SHA design practices, FHWA design manuals, the 1993 AASHTO Guide for Design of Pavement Structures, and mechanistic-empirical performance models that were used to limit the occurrence

ix

of key distress types for flexible and rigid pavements and adjusted as needed to limit key distress types within specified performance criteria.

Prototype Knowledge-Based Expert System (KBES). A prototype microcomputer-oriented, KBES for selecting pavement design features was developed under this study. The prototype KBES is a Windows 3.11 program with standard user friendly Windows graphic user interface. The prototype KBES includes three main parts: •

Input assistant to provide interactive guidance to the designer in obtaining design inputs for the catalog site conditions and other inputs;

•

Database searching and presentation to access a project database which represents the paper catalog, to quickly and efficiently identify feasible design alternatives for a given set of site and design conditions;

•

Evaluation assistant to provide interactive guidance to the designer in evaluating the advantages and disadvantages of the various design alternatives and explanations about various design features.

x

Chapter 1 Introduction and Research Approach

Problem Statement A rational pavement design must consider the effects of roadbed soil, climate, traffic loading, construction materials, and other design details and features on pavement performance and lifecycle costs. The objective of the design process is to identify pavement structures that will provide acceptable performance and economy over the intended design life. For a given combination of roadbed soil conditions, climate, and traffic loading, therefore, the goal of a pavement engineer during design is to determine the pavement structure and the related design features that will meet the requirements at a particular location. Given the similarities in environment, traffic, and roadbed soil within and among States, one would expect similarities in the practices for design and construction in the nation. However, this is not the case. The practices for design and construction vary widely. A catalog for pavement design that identifies recommended design structures and features for flexible and rigid pavements would help guide highway authorities in selecting suitable and reliable designs. Nearly all European highway agencies have developed and adopted such design catalogs. Supplementing such a catalog with a microcomputer-oriented, knowledge-based expert system would further enhance the catalog's use and facilitate its updating.

1

Research Objectives The objectives of this research are (a) to evaluate the feasibility of developing a comprehensive catalog of recommended design features for both flexible and rigid pavements and a corresponding prototype expert system, and (b) if feasible, to develop such a practical catalog and prototype expert system for selecting the recommended design features.

Research Approach The research approach used to accomplish the objectives included the following major items: (1) A detailed review of existing design catalogs from European countries was conducted and interviews were held with several experts involved with the development of those catalogs. In addition, three states (Washington, New York, and Missouri) have recently developed interesting catalog like documents which were also reviewed. (2) Three key site conditions were identified for the catalog factorial matrix: traffic loads, subgrade support, and climate. These site conditions define the factorial matrix which serves as the main structure of the catalog. Each “cell” in the factorial matrix represents a specific combination of traffic loading, climate, and subgrade support. (3) Key design features were identified for flexible and rigid pavements that should be included in each of the factorial “cells” of the catalog. Design features include subgrade treatments, subdrainage features, shoulder design, layer thickness, material requirements, joint design and other features. Thus, a large majority of the design features that are specified for design and construction are included in one document and are easily obtained. (4) The current pavement design procedures, design inputs, and design features used by each State highway agency (SHA) for flexible and rigid pavements were obtained, summarized, 2

and checked twice by each state. The many summary tables of design features represent the first ever compilation of such information for both flexible and rigid pavements. (5) Based upon this information, a Catalog of Current State Pavement Design Features was developed (15). This catalog consists of three main parts: (a)

State synoptic tables of many pavement design features (by State).

(b) Individual State factorial design matrix of pavement design features including design layer thicknesses for a wide range of traffic, subgrade, and climatic site conditions for major types of main highway pavement constructed by the SHA. (c )

Catalog of key design features averaged over climatic region and pavement type.

(6) Feedback on the usefulness and advantages/disadvantages of a catalog was obtained directly from the various segments of the U.S. highway industry through questionnaires and interviews with experienced engineers. (7) The objectives, scope, and some critical developmental issues of the catalog were then further defined, along with key advantages and disadvantages. All of this information was then analyzed to obtain a comprehensive feasibility evaluation of producing a catalog of recommended design features. The feasibility of developing a catalog of recommended design features and a supplemental computer-oriented, Knowledge-Based Expert System (KBES) were determined. Both were judged to be feasible and after approval by NCHRP, both were then developed. (8) A draft catalog of recommended design features for flexible and rigid pavements was first prepared based upon a review of European catalogs, existing design procedures, mechanisticempirical design checks, and experience of the research staff.

3

(9) A consensus meeting was held where many experts were brought together to deliberate and revise the draft recommendations. A consensus was reached on almost all of the issues and the draft catalog was extensively revised. The catalog was then further reviewed by the NCHRP and many further revisions were made. Thus, the catalog represents the composite knowledge of many practitioners as well as traditional sources of existing design procedures and mechanistic-empirical design checks (16). (10) A prototype knowledge-based expert system (KBES) was developed to supplement the paper catalog. The prototype KBES represents the paper catalog in a electronic format and also adds further capabilities. When fully developed, the prototype KBES will supplement the catalog in the following ways: •

Provide interactive guidance to the designer in obtaining design inputs for the catalog site conditions and other inputs. This part is well developed and includes all the information that is included in the printed catalog.

•

Provide access to a relational database which represents the printed catalog, to quickly and efficiently identify feasible design alternatives for a given set of site and design conditions. This goal was also achieved. The prototype KBES can search for the viable design alternatives very efficiently.

•

Provide interactive guidance to the designer in evaluating the performance of the various design alternatives. The prototype KBES provides mock-ups to illustrate this important concept.

•

Provide explanations about various design features. An on-line technical help file is under development to achieve this goal.

4

A plan for extending the prototype KBES into a fully operational KBES was developed. The prototype KBES must be developed into a fully operational KBES if its true value is to be realized.

5

Chapter 2

Findings

Success of Pavement Design Catalogs/Standards in European Countries Nearly all European countries, including Germany, France, Austria, Spain, Belgium, Italy, and Switzerland, use pavement design catalogs. To learn from their experience, the catalogs from these countries were thoroughly reviewed, and first-hand information was sought from representatives of the highway agencies. A summary of most of these catalogs is included in Appendix A. The pavement design catalogs, or pavement standards as they are called in some countries, were initially developed in the 1960s and 1970s. They started out as very simple documents of only a few pages showing diagrams of pavement sections as a function of traffic and often subgrade support. These simple catalogs or standards have since evolved into very comprehensive documents. (1-13) These European catalogs and standards have evolved into what would be considered in the U.S. as portions of geotechnical manuals, pavement design manuals, geometric design manuals, a variety of standard drawings and cross sections, drainage manuals, materials manuals, and some aspects of construction specifications all placed into a single document. However, the presentation format of the European catalogs is different There is no doubt that pavement design catalogs or standards have become highly popular and successful in many European countries since the 1970s. They are actually being expanded into a variety of other highway systems in some countries such as cities, ports, and toll roads. Some comments from various European countries about design catalogs are provided. 6

Spain. Professor Carlos Kraemer of the Universidad Polite'cnica de Madrid was involved in the first catalog developed in Spain in 1975. He noted the following advantages of Spain's pavement catalog (personal communication): •

Establishes uniform design criteria based on standard specified materials of adequate quality.

•

Improves the average performance of pavements by eliminating through experience underand over-designed solutions.

•

Provides nonspecialized engineers with a practical tool which reduces the risk of misunderstandings and errors in theoretical calculations, or even when following a design guide like AASHTO. The designer can focus on the choice of the most appropriate solution in each specific case on the basis of availability of materials and of cost.

•

Reduces the number of alternative pavement structures, which is beneficial for contractors who soon become familiar with them.

•

Simplifies the supervision of design, construction quality control, assessment of performance, and, in short, maintenance management (through a reduction in alternative pavement structures).

•

Allows the progressive optimization of the design process through performance observations and analysis.

•

Professor Kraemer also listed some disadvantages (personal communication):

•

There is a risk to introduce routine in the design process. Pavement design is sometimes considered to be a solved problem; hence, fewer engineers are motivated in becoming specialists in this area.

•

New materials may be more difficult to be introduced, even with a standard structural design open to initiatives (for instance, high-modulus asphalt mixes instead of conventional ones). 7

•

It may be less adaptable to regional or local conditions, like extreme climates, local marginal materials, particular local aggregates (for instance, some volcanic materials).

•

A certain lack of flexibility to adapt to budget limitations.

Professor Kraemer had this to say about achieving a consensus: "Before final approval, the catalog was submitted for comments to a committee where the asphalt and cement industries were represented, as well as the union of the most important contractors of the country . . . It was not difficult to arrive at a consensus. A certain balance in construction cost has been achieved between pavement structures designed for a given traffic and subgrade category. The different solutions are reasonably competitive." Some entities in Spain (regional administrations, particularly those with low traffic volume roads and ports) have developed their own catalogs, better adapted to their special experiences, materials, and needs. This demonstrates the popularity of the catalog concept in Spain. "The catalog is fully accepted in Spain by government officials, asphalt and cement industries, consulting engineers, and contractors. After 19 years, its practical use can be taken as proved. The same approach has been extended to other areas such as urban roads, port pavements, and even mine haulage roads. A revision should be considered about every 10 years. If after a shorter period, evidence shows a pavement structure as under-designed or a new solution appears as useful, an addendum should be issued. Every new edition should particularly take into account new materials (with standard specification drawn up), the evolution of freight transport and truck wheel loads and the general economic situation." (C. Kraemer, personal communication)

8

Germany. Professor Gunther Leykauf of the Technische University at Munich provides the following comments on the German experience: "The first design catalog in Germany was published in 1966 and consisted of only one sheet. Since that time the thickness standardization has been revised several times, and the thickness of the catalog increased, too. A new version has been installed this year to revise our last RSto 86/89; main goal is to introduce structures for widening, renewal, and reconstruction of pavements. "(1) Initially, engineering judgement was the technical basis for the catalog. However, with each revision the results of theoretical calculations, laboratory and field tests, as well as practical experiences were introduced. Failure criteria were fatigue cracking for asphalt pavements (however, not absolute values were considered but a comparison between different structures) and bending stresses for concrete pavements. Rutting scarcely can be influenced by thickness, but by the asphalt mixture (crushed aggregates) and a good compound at the layer interfaces. "(2) A minimum bearing capacity is required on top of the subgrade (foundation level). Further advantages are a limitation of possible technical solutions; hence, an improved communication between client and contractor, rationalization effects, and better quality due to the standardization. "(3) There is a good cooperation between government engineers, university researchers, and industries. "(4) Compromises have been necessary for the final thickness catalog (to achieve a consensus). 9

"(5) Complaints arise with a certain regularity every 5 to 10 years. Just now there is a discussion [as to] which kind of wearing course for asphalt pavements must be considered in a cost-comparison with a concrete structure. However, the problems are settled by the price or by decisions of government agencies to demand either an asphalt or a concrete pavement (for very heavily trafficked highways). "(6) The RSto 86/89 is fully accepted in practice. There exists the need to introduce structures for the renewal of pavements and for widening of pavements." (G. Leykauf, personal communication). Dr. Andreas Zachlehner of Germany stated the following: "The main concept of the German pavement catalog is to provide ready-made structural pavement alternatives for a given amount of traffic and, in a minor degree, climatic conditions. Because of this, the traffic catalog is of great importance, especially for local administrations without a large amount of technical staff or experience. "One of the main characteristics of the design procedure is that a fixed value of bearing capacity is required for the subgrade independent of local conditions. By this means, the subgrade conditions are eliminated as design parameters. This may not always lead to the most economical solution. "Another characteristic is that the different pavement structures considered for given traffic and climatic conditions are supposed to be technically equivalent. The difficulty consists precisely in establishing the technical equivalence between those different pavement structures . . . as the not very well-known, long-term performance may be decisive for this comparison. The once declared technical equivalence of two different pavement structures theoretically allows the tender to submit proposals including any of them." (A. Zachlehner, personal communication) 10

Dr. Peter Canisius of the Federal Highway Administration in Germany stated the following: "Prior to the development of standards or catalogs, lots of people were doing designs, some good and some poor designs. People said 'Let's find the best design and use that.' Now you can select alternative designs for given conditions. In Germany, they use the standards (catalog) for most projects, and thus different designers would get the same designs. If the contractor follows the standards they will get good performance. Germany designs against frost heave by providing granular blanket of sufficient thickness. Advantages of catalog: not enough experienced pavement design engineers in Germany to design roads. The catalog reduces errors in pavement design from inexperienced engineers. The subgrade is not a factor in the catalog because every subgrade must meet this minimum standard bearing capacity. Procedures are included about what to do with poor soils. Contractors try to optimize standards, if [problems develop with a] pavement, they have to repair because of their warranty for construction. Does a catalog affect innovation? Germany is not against innovations. Industry builds roads to standards; however, this does not hamper innovations." (P. Canisius, personal communication)

France. Mr. J. P. Christory of the Laboratoire Regional de L'Ouest Parisien in France has provided information concerning his active participation in developing pavement design catalogs for various cities and regions of France. Pavement design catalogs have been very successful in France, as evidenced by the fact that the first catalog was developed by the government in 1977 for freeways and has since been updated and expanded greatly. In addition, other catalogs for toll roads and urban areas (such as Paris) are now available and have had strong support and success. The French see the catalog format as a highly effective format to communicate 11

pavement design information to all persons interested in pavements. Mr. Christory is involved in the active development of these catalogs for several entities in France. Thus, in summary, pavement design catalogs/standards have been very successful in European countries, and are in use in everyday design.

12

Pavement Design Methods Versus Pavement Design Presentation Formats This section distinguishes between “pavement thickness design methods or procedures” and "pavement design presentation formats" by which a design method is made available for use by practicing engineers. They are completely different things, and this is crucial to understanding the potential value of a “pavement design catalog” which is a “presentation format”, not a design method. Several different “thickness design methods” exist, such as those developed by American Association of State Highway and Transportation Officials (AASHTO), Asphalt Institute (AI), Portland Cement Association (PCA), Illinois DOT Mechanistic Design, Washington DOT, and the Texas DOT FPS and RPS. Each thickness design method is based on some underlying theoretical and/or empirical concepts. Along with the thickness design methods for each agency there are corresponding materials specifications, mix designs, subgrade treatments, and standard design drawings for cross-sections, joints, reinforcement, and subdrainage. In addition to these “thickness design methods,” and not to be confused with them, are different "design presentation formats" by which a design method is made available for use by a designer. For example, the AASHTO thickness design procedure has been utilized by designers in several different formats: nomograph, spreadsheet, and various computer programs. A catalog for second overlays of composite pavements was recently developed for the Illinois DOT based on the AASHTO procedure. The New York DOT used a modified AASHTO thickness design procedure plus other models to develop a tabular or catalog format for presentation. The following underlying thickness design methods have utilized various presentation formats for design engineers:

13

•

Completely empirical and experience-based methods have utilized tables, graphs, and catalogs.

•

Closed form design equation (empirical or semi-empirical) methods have used hand calculator, nomograph, tables, graphs, spreadsheets, micro-computer software, and catalog.

•

Complex design algorithms (usually mechanistically based) methods have used computer software, but have also been simplified into tables, graphs, and catalogs.

Therefore, the design presentation format is not necessarily related to the sophistication of the thickness design method. Also, in the subsequent feasibility evaluation, a pavement design catalog utilizes both some underlying thickness design theory and a specific type of design presentation format (i.e., the catalog format). The underlying design theory can be anything from engineering experience, to empirical equations from road tests, to sophisticated mechanistic algorithms, to any combination of these. Some European catalogs that were originally developed from experience have since been checked and revised using mechanistic theory but retain their catalog presentation format. In addition to layer thicknesses, the catalog approach provides an easy-to-use format to present recommendations for other design features such as shoulders, subdrainage, materials criteria, joint design, subgrade treatments, etc. all in a single document.

Catalog of Current State Pavement Design Features A Catalog of Current State Pavement Design Features (15) was prepared that provides a highly informative and practical guide on the details of the design practice for flexible and rigid pavements in the U.S. 14

The catalog first and most importantly provides information in the form of a large Synoptic Tables Of State Pavement Design Features. This section provides for the first time for both flexible and rigid pavements a comprehensive summary of nearly all state’s design practices. Second, the catalog provides, for each state highway agency (SHA), a State Factorial Design Matrix of Pavement Design Features that gives design layer thicknesses for a wide range of traffic, subgrade, and climatic site conditions for major types of main highway pavement constructed by the SHA. This information is provided for both flexible and rigid pavements, as constructed in the individual states. Third, the catalog provides a Catalog of Key Design Features by Climatic Region and Pavement Type. These results are presented graphically in the form of a catalog of designs for flexible and rigid pavements in four climatic regions. Fourth, a summary of key design features for flexible and for rigid pavements are given. This information will be of interest to highway administrators, pavement designers, contractors, industry, and others involved in various aspects of pavement design and construction.

Feasibility study A detailed evaluation was conducted into the feasibility of developing a comprehensive catalog of recommended design features for both flexible and rigid pavements and a supplemental prototype knowledge-based expert system (KBES). The feasibility study included a review of design catalogs and procedures from European countries and U.S. States, identification of key site conditions and design features, and finally feedback on the usefulness and advantages/ disadvantages of a catalog obtained directly from the various segments of the U.S. highway industry. 15

(1) This was accomplished through the development of a "strawman" catalog (and knowledgebased expert system) that illustrates the main catalog format and design features to be included. This "strawman" was based upon the catalogs from European countries, recent catalog-like designs by State highway agencies (SHAs), and the Catalog Of Current State Pavement Design Features and Practices developed under this study. (2) The "strawman" catalog and accompanying questionnaire were sent to 27 experienced engineers representing the State highway agencies, Federal Highway Administration (FHWA), and the asphalt and concrete paving industries. Nineteen (19) responses were received and analyzed. Results of the surveys are as follows: (A) Who might use the catalog? Figure 1 shows the percentage of respondents indicating who might use the catalog: local agencies (79%), consultants (79%), central offices (72%), districts (72%), experienced engineers (68%), new engineers (68%), contractors for information purposes (58%), and industry (58%). These results show that the respondents in general believe that there is a lot of interest in the catalog. (B) How might the catalog be used? Figure 2 shows the percentage of respondents indicating how the catalog might be used: to review and check designs (95%), to obtain information on recommended design features for comparative purposes (84%), to "customize" or adapt the catalog to specific agency conditions and use for routine design (43% of SHAs responded affirmative), to train personnel in pavement design (42%), and to update agency design procedures (37%). These results show that catalog could have several uses, if developed properly.

16

(C) What advantages will the catalog provide? Figure 3 shows the percentage of respondents indicating what advantages the catalog would provide: simplicity of use (89%), improvement of the efficiency of the pavement design process (68%), improvement of communication with administrators about pavement design (63%), easy implementation of a new (or modified) design procedure (58%), provision of an easy to use source of key design features for a given project (58%), and enhanced communication in pavement design and construction (42%). These advantages agree well with those mentioned by European experts. Central Offices 72 % Experienced pavement design engineers 68%

Local Agencies 79%

Industry 58%

Consultants 79%

CATALOG SURVEY "Who might use a catalog?"

Contractors 58%

Districts 63%

New inexperienced engineers 68%

Figure 1. Percent respondents indicating the prospective users of a catalog.

17

Make comparisons 84 % Review and check design 95%

CATALOG SURVEY "How might a catalog be used?"

Update agency design 37%

Training 42%

Customize - Adapt to agency 43% (SHA respondents only)

Figure 2. Percent respondents checking the categories of the ways of using a catalog.

18

Simplicity of use 89%

Improvement of communication with administrators 63%

Easy-to-use source of key design features for a project 58%

Improvement of the efficiency 68 % Reduction in error 53%

CATALOG SURVEY "What Advantages?"

Easy implementation 58%

Enhance communication 42%

Figure 3. Percent respondents indicating the potential advantages.

(D) What disadvantages will the catalog provide? Figure 4 shows the percentage of respondents indicating what disadvantages the catalog would provide: provides new engineers with a false sense of security (83%), applicability of the design catalog to some local site conditions (71%), gives engineers and management too simplified a view of pavement design (67%), recommended design features may not fit local project conditions (67%), and incompatibility with current design methodology (44%). The respondents clearly indicate that there are some significant disadvantages with the catalog also. Some of these can be dealt with in development. (E) What aspects of the design catalog would be the most valuable? The percentage of

19

respondents indicated the following aspects would be most valuable: Layer material properties

89 percent

Layer thickness

74 percent

Subgrade treatments

68 percent

Subdrainage design features

63 percent

Shoulder design features

42 percent

Joint design features

37 percent

Reinforcement design

26 percent

One respondent indicated that they were all equally important to a pavement design.

New engineers false sense of secuity 83 % Applicability to some local conditions 67% Too simplified a view of design 67%

CATALOG SURVEY "What Disadvantages?"

Recommended design features may not fit local conditions 71%

Incompatibility with current design methodology 44%

Figure 4. Percent respondents selecting the disadvantages categories.

20

(F) Do you feel that a microcomputer-based knowledge-based expert system (KBES) would enhance the value of the printed catalog? The percentage of respondents indicated the following: Yes, most definitely Yes, somewhat No, not likely

66 percent 28 percent 6 percent

There appears to be strong support for the KBES. (3)

Interviews with experienced pavement researchers, designers, administrators, and contractors about the feasibility of pavement design catalogs and about a computerbased, knowledge-based expert system (KBES) were also conducted. Several European experts who have actually been involved in the development of design catalogs for their counties were also interviewed. The NCHRP 1-32 panel also provided valuable information.

(4)

The objectives, scope, and some critical developmental issues of the catalog were then further defined, along with key advantages and disadvantages. All of this information was then analyzed to obtain a comprehensive feasibility evaluation of producing a catalog of recommended design features.

Based on all of this input, it was concluded that it is feasible to properly develop a Catalog of Recommended Pavement Design Features and a supplemental KBES having specific objectives and scope. The catalog and the KBES had overall good support from a cross-section of the

21

Federal, State, and industry highway engineers and contractors. However, several disadvantages were identified that must be eliminated or minimized during development. It is also concluded that a properly developed catalog and KBES has a good chance of being used for a variety of purposes by a variety of agencies and individuals. These conclusions are based upon the positive ratings and general comments of respondents (State highway agencies, FHWA, industry), many years of successful European experience, and the recent development of catalog-like procedures by several States (Washington, New York, Missouri)(references).

The Consensus Meeting Input from all sectors of the highway industry was required to develop the catalog. Thus, a large resource group of experienced pavement engineers from all sectors of the industry was assembled. The resource group meeting was held in Chicago at the O Hare Hyatt Regency Hotel from January 22-26, 1996. Sixteen of the eighteen members of the expert resource group attended along with the research team members and Professor Lorenzo Domenichini from Italy who gave an overview of the development and use of catalogs in Europe and answered many related questions. Members of the consensus group, Professor Domenichini s presentation, and the minutes from the meeting are given in Appendix B. The consensus group spent the week discussing and reviewing the draft Catalog of Recommended Pavement Design Features that had been previously provided to them. The meeting produced two major results: many recommendations on design features and a consensus on nearly all of the final recommendations.

22

The meeting was very lively, especially during the first two days. Obtaining a consensus (meaning specifically that no one in the group opposed a given recommendation, although not everyone was completely satisfied either) was very difficult, and often required considerable efforts to find a recommendation that could be supported by all. The following definition of “consensus” was used as provided by management consultant Peter Scholtes who states that consensus represents decisions that best reflect the thinking of all persons involved.

"Consensus is . . . finding a proposal acceptable enough that all members can support it; no member opposes it." "Consensus is not . . . a unanimous vote—a consensus may not represent everyone's first priorities. . . a majority vote—in a majority vote, only the majority gets something they are happy with; people in the minority may get something they don't want at all, which is not what consensus is all about. . . Everyone totally satisfied." "Consensus requires . . . time, active participation of all group members . . . skills in communication: listening, conflict resolution, discussion facilitation, creative thinking and open-mindedness."(14) The entire group balloted the general design recommendations related to all pavement types (i.e., approach used for structural design, design checks, design reliability, initial serviceability, final serviceability). The group was then divided into two groups, one for asphalt pavements and one for concrete pavements for specific design features. These groups then balloted on design recommendations specific to that pavement type. Each day, a summary of the previous days results was provided to the entire group assembled together. Ballots were conducted on over 40 recommended design features. The research team is very pleased that a consensus was eventually achieved on practically every design feature. The most valuable aspect of the meeting was the many interesting and 23

beneficial improvements proposed by the highly experienced and diverse resource group. The research team was frankly surprised at the ability to eventually reach consensus on so many recommendations.

24

The consensus building process consisted of the following steps: (1) A description of the recommendation by a member of the 1-32 research team. (2) Discussion by the resource group. (3) Secrete balloting (each group member filled out a ballot that contained a rating scale that ranged from “strongly disagree” to “strongly agree” and the reason if the person disagreed or strongly disagreed). (4) A display of a frequency distribution of ratings on a computer screen for all to see. (5) The announcement that “a consensus has been reached” if no one rated either “strongly disagree” or “disagree”, OR the announcement that “a consensus has NOT been reached.” (5) If a consensus was not reached, the reasons why were read from the ballots and further discussion was held until a the recommendation was modified or clarifying notes attached that made it possible for all members to support the recommendation. Another ballot was then taken to verify that indeed a “consensus” had been reached.

This approach was so successful that it should be considered for use on other research projects where expert human experience is important. The results from this meeting were used to develop the catalog as described in the next section.

Catalog of Recommended Pavement Design Features (16) This document presents a catalog of “good practice” recommendations for design features of highway pavements for highway engineers, administrators, and others in an easy to use format. Guidelines are provided for three main site conditions: traffic loadings, subgrade support, and

25

climate. Based on these inputs, design feature recommendations are provided in “design cells” including the pavement cross section, structural design, materials, and other features required to meet minimum performance requirements. This pavement design catalog is a relatively simple but effective mode of presentation of an underlying pavement design methodology that includes both empirical and mechanistic components. The entire document is included in Reference 16. There are several potential uses for this catalog, as described in Chapter 3.

Description of Catalog. The contents of this Catalog of Recommended Pavement Design Features for highway pavements include the following: •

Recommended (good practice) design features for highway engineers, administrators, and others in a format that is easy to use and understand.

•

A model catalog presentation format for potential usage by agencies.

•

Recommended consensus on many design features for pavement design for varying site conditions.

The printed design catalog is organized into the following parts: Part 1

An introduction to the catalog.

Part 2

How to use the catalog and design criteria.

Part 3

Project site condition inputs (climate, traffic, and subgrade).

Part 4

Guidelines on recommended design features for alternative pavement structures that will meet the minimum performance requirements of the site condition cell.

Part 5

Special subsurface conditions.

References 26

Glossary

Definition of terms used in catalog.

Appendices

Detailed information on inputs, design examples, design check models, and notes on design feature recommendations.

Scope of Catalog. The catalog provides recommendations on design features for highways ranging from heavily trafficked Interstate and primary highways to secondary highways. The specific pavement types included in the catalog are as follows:

Flexible Pavements •

Asphalt concrete pavement with a crushed aggregate base.

•

Asphalt concrete pavement with an asphalt treated base.

•

Asphalt concrete pavement with a cement treated base.

•

Asphalt concrete full-depth pavement.

Rigid Pavements •

Jointed plain concrete pavements (JPCP).

•

Jointed reinforced concrete pavements (JRCP).

•

Continuously reinforced concrete pavements (CRCP).

The catalog, as it stands, is not intended for direct use in pavement design. Design feature recommendations are provided in the form of acceptable ranges within each cell of site conditions and, thus, are not suitable for use in design. The catalog will, however, provide recommendations that are adequate to identify design features for flexible and rigid pavements

27

that will help guide highway authorities in selecting suitable and reliable designs, and to check designs. The catalog is applicable to project site conditions and construction practices encountered in the United States with guidelines provided for the appropriate adjustments for special subsurface conditions. In general, the catalog provides recommended design features that meet specific minimum performance requirements for a given set of site conditions. A critical issue that is addressed for each type of pavement and cell site condition is material requirements. The structural and durability requirements for the materials vary between some site condition cells due to traffic and climatic conditions.

Basis for the Catalog. The catalog recommendations are based on many sources, however, the most significant source is the recommendations achieved by consensus of a large resource group of pavement design experts from Federal, state, industry, consulting, and academia. The resource group met for an entire week and debated and revised many proposed recommendations until a consensus was reached (Appendix B). Contributions were also made by the NCHRP based on reviews of the documents. In addition, use was made of current SHA design practices (15), FHWA design manuals, the 1993 AASHTO Guide for Design of Pavement Structures, and mechanistic-empirical performance models that were used to limit the occurrence of key distress types for flexible and rigid pavements and adjusted as needed to limit key distress types within specified performance criteria.

28

Structural Sections in the Catalog. This catalog provides structural sections that are expected to carry a specified amount of mixed traffic that has been projected to occur over a given design period. Differing amounts of maintenance and rehabilitation may be required to reach the end of the design period, and of course life-cycle costs may vary between structural sections.

29

Prototype Knowledge-Based Expert System (KBES) A prototype Knowledge-Based Expert System (KBES) was developed to supplement the Catalog of Recommended Pavement Design Features. This chapter and the User’s Guide (16) document the development and the prototype KBES, called Designer. The software development will be described first, followed by an overview of the prototype KBES, and a discussion on the use of the KBES is given in Chapter 3.

Software Development. A prototype microcomputer-oriented, KBES for selecting pavement design features was developed under this study. The inference engine of the KBES uses the expert system shell, CLIPS 6.04. CLIPS is an acronym for C Language Integrated Production System. It was developed by the Software Technology Branch of the National Aeronautics and Space Administrations (NASA). CLIPS is disseminated under the sponsorship of NASA by the Computer Software Management and Information Center (COSMIC) in the interest of information exchange. There is no royalty required to use and further develop the KBES using CLIPS. CLIPS 6.04 is written in C-language for interactive execution on IBM PC compatible computers running MS-DOS v5.0 or higher and MS Windows v3.x. Features of CLIPS include a conventional rule-based expert system, procedural programming ability, and CLIPS object-oriented programming capability. All these features are utilized in developing the prototype KBES. The rule-based expert system part is used frequently in the KBES whenever there is any reasoning or pattern matching involved. The procedural and object-oriented programming languages make the pattern matching of the pavement conditions to the catalog factorial cells very easy and efficient.

30

The prototype KBES is a Windows 3.11 program with standard user friendly Windows graphic user interface. Microsoft Windows programming language, Visual Basic (VB) 4.0, was used to develop a friendly user interface. The core part of the prototype KBES is programmed using CLIPS. It can be executed on any IBM PC compatible 486 computer with at least 4Mb RAM and Windows 3.11 or higher. The prototype KBES program architecture is a standard modular design and is very flexible and easy to modify. The link between the VB interface and CLIPS programs is dynamic in the sense that the inferences and screen framework are completely separated. All the logical inference, derivations, and technical contents are provided in CLIPS code. CLIPS also provides all the necessary outputs and screen display instructions to VB. Therefore, the prototype KBES is easy to modify and enhance in the future.

Overview Of The Prototype KBES. The prototype KBES developed in this project includes three main parts: •

Input assistant to provide interactive guidance to the designer in obtaining design inputs for the catalog site conditions and other inputs;

•

Database searching and presentation to access a project database which represents the paper catalog, to quickly and efficiently identify feasible design alternatives for a given set of site and design conditions;

•

Evaluate assistant to provide interactive guidance to the designer in evaluating the advantages and disadvantages of the various design alternatives and explanations about various design features.

31

Appendix H of Reference 16 describes in detail each of the above modules.

CHAPTER 3 INTERPRETATION, APPRAISAL, APPLICATIONS

As discussed in Chapter 2, several important products have been developed under this study. The main products of this project include the following: Catalog of Current State Pavement Design Features, Catalog of Recommended Pavement Design Features, and the supplemental prototype KBES. The usefulness and the potential applications to State highway agencies, as well as the advantages and the disadvantages or limitations of these products are presented in this chapter.

Catalog of Current State Pavement Design Features A wealth of information about pavement design features has been collected from each State in the United States and has been compiled into the Catalog of Current State Pavement Design Features (15). This State catalog provides much useful information to pavement engineers and administrators. The State catalog may be used in the following ways: •

Provides a highly informative and practical guide to the details of design practice being used for flexible and rigid pavements throughout the U.S.

•

A reference of pavement design features that may be useful in developing designs;

•

Can be used for comparison studies of the pavement designs between agencies;

•

Can be used as a training tool to learn more about the current pavement design practices used in the U.S. 32

•

The summary and the statistical analysis part of the key pavement design features provides a quick reference of the current common pavement design practices for highway authorities.

Catalog of Recommended Pavement Design Features A catalog of “good practice” recommendations for highway pavement design features was developed for highway pavement engineers, administrators, researchers, and others. It is a relatively simple but effective mode of presentation of an underlying pavement design methodology that includes both empirical and mechanistic components.

The catalog recommendations are based on many sources, however, the most significant source is the recommendations achieved by consensus of a large resource group of pavement design experts from Federal, state, industry, consulting, and academia. The resource group met for an entire week and debated and revised many proposed recommendations until a consensus was reached. Contributions were also made by the NCHRP based on reviews of the documents. In addition, use was made of current SHA design practices, FHWA design manuals, the 1993 AASHTO Guide for Design of Pavement Structures, and mechanistic-empirical performance models that were used to limit the occurrence of key distress types for flexible and rigid pavements and adjusted as needed to limit key distress types within specified performance criteria. The catalog developed under this study is not intended for direct use in design. Design feature recommendations are in the form of acceptable ranges with each site condition cell, and thus not suitable for use in design. The catalog provides recommendations that identify design features for flexible and rigid pavements that can help guide highway authorities in selecting suitable and reliable designs.

33

The following are the possible applications of the general catalog: •

Obtain information on recommended design features for comparative purposes.

•

Train personnel.

•

Update aspects of agencies’ current design procedures.

•

Review or check current pavement designs.

•

Remind pavement engineers of design alternatives they might consider for a given set of conditions.

For example, the process of using this catalog to compare with agency designs can be summarized as follows: 1. Select pavement type to be compared. 2. Identify the site condition design cell for the project under consideration through estimation of traffic, subgrade, and climate inputs. 3. Obtain the recommended design features from the design cell. 4. Compare the recommended design features in the catalog with those of the agency. 5. Investigate the reasons for significant differences between the design features.

To illustrate the use of the catalog, examples of using the catalog to identify flexible pavement design features and rigid pavement design features are provided.

Example to Identify Flexible Pavement Design Features

34

Step 1 - General project description Conversion of two lane highway into four lane highway by construction of two additional lanes that will carry one directional traffic. Primary highway, two lanes in one direction. Crowned cross-section with 12 ft lanes and 6 ft shoulders on each side. Design period = 20 years Climatic zone = wet-freeze Step 2 - Determine site condition input for traffic (16) ADT (initial) = 10,000 two-directions Mean percent trucks = 10.0 Lane distribution factor (trucks) = 0.87 Directional distribution factor (trucks) = 0.5 (same number in each direction) Future growth of trucks = 5 percent per year, compounded GF = [ (1 + 0.05)20 - 1 ] / 0.05 = 33.06 Mean truck equivalency factor over 20 years (flexible ESAL/truck) = 1.1 Total ESALs = 10,000 * 0.1 * 365 * 0.87 * 0.5 * 1.1 * 33.06 = 5,774,012 outer lane, one direction. Step 3 - Determine site condition input for subgrade (16) Natural subgrade soil = silty clay Resilient modulus backcalculated form FWD deflection data from existing pavement. Mean backcalculated resilient modulus = 15,000 psi Mean adjusted resilient modulus = 15,000 * 0.35 = 5250 psi Seasonal adjustment required as shown in table 1.

35

Table 1. Calculation of seasonally adjusted effective resilient modulus for subgrade. Season

No. of

In Situ Resilient

Damage

Seasonal

Months

Modulus, psi

Ratio

Damage Ratio

Summer

3

5,000

0.40

1.20

Fall

3

6,000

0.30

0.90

Winter

3

10,000

0.06

0.18

Spring Thaw

1

2,000

2.60

2.60

Spring Total

2

3,500

0.70

1.40 6.28

12

Average Damage Ratio = 6.28 / 12 = 0.52 Effective Resilient Modulus = 4,000 psi

Step 4 - Determination of subgrade preparation/improvement needed (Part 4A and Part 5 of Reference 16) The seasonally adjusted resilient modulus falls into the “Very Soft” class. Subgrade “improvement” is strongly recommended for this class of soil to provide a good construction platform and a more uniform support condition. Since soil was found to be reactive with

36

hydrated lime, the top 12 in of the subgrade will be stabilized and compacted prior to construction of the pavement layers. Step 5 - Determine site condition input for climate Pavement is located in a wet-freeze area of the United States. The freezing index = 500 degree days below freezing, average annual precipitation = 40 in.

Step 6 - Determination of recommended alternative flexible pavement types (Part 4A of Reference 16) The pavement design matrix with recommended AC pavement design alternatives is shown in table 2. The site condition cell and the design alternatives of this example are identified as: Cell 13: 4-8 million ESALs and Subgrade resilient modulus = 4,000 psi. Alternative 1: HMAC Surface/Binder over Conventional Unbound Granular Base Alternative 2: HMAC Surface/Binder over Asphalt Treated Base Example structural design details of Cell 13 are shown in figure 5.

37

Table 2. Site condition design cells and alternatives for flexible pavement catalog (16).

*These flexible pavement types are considered “marginal” for the specific site condition cell noted.

38

Structural Layer Thickness - Asphalt Treated Base

Dense Graded Asphalt Concrete Surface Thickness, in. Asphalt Treated Base Thickness, in. Granular/Aggregate (Pit Run Gravel) Subbase Thickness, in.

5.5-6.5

Dense Graded Asphalt Concrete Surface Thickness, in.

5.5-6.5

10.0 (Plant mixed)

11.0 (Roadway Mixed)

13.0-14.0

13.0-14.0

Asphalt Treated Base Thickness, in. Granular/Aggregate (Pit Run Gravel) Subbase Thickness, in. Improved Subgrade Thickness, in.

Prepared Subgrade(See Section 5)

5.5-6.5

5.5-6.5

8.0 (Plant mixed)

9.0 (Roadway Mixed)

14.0

14.0

6.0 - 12.0

6.0 - 12.0

Note: A filter layer (or separator) is recommended between the subbase and very soft subgrades (See Section 5).

Controlled by AASHTO-PSI Criteria

Dense Graded Asphalt Concrete Surface Thickness, in. Asphalt Treated Base Thickness, in. Crushed Stone Aggregate Subbase, in. Granular/Aggregate (Pit Run Gravel) Subbase Thickness, in.

5.5-6.5 10.0 (Plant mixed) 7.0

Controlled by AASHTO-PSI Criteria

Dense Graded Asphalt Concrete Surface Thickness, in.

5.5-6.5 11.0 (Roadway Mixed)

Asphalt Treated Base Thickness, in.

7.0

5.0-6.0

Crushed Stone Aggregate Subbase

5.0-6.0

Granular/Aggregate (Pit Run Gravel) Subbase Thickness, in.

Prepared Subgrade (See Section 5)

5.5-6.5

5.5-6.5

8.0 (Plant mixed)

9.0 (Roadway Mixed)

6.0

6.0

7.0

7.0

6.0 - 12.0

6.0 - 12.0

Improved Subgrade Thickness, in.

Note: A filter layer (or separator is recommended between the subbase and very soft subgrades (See Section 5).

Controlled by AASHTO-PSI Criteria

Controlled by AASHTO-PSI Criteria

Structural Layer Thickness - Cement Treated Base

Dense Graded Asphalt Concrete Surface Thickness, in.

6.5-7.5

6.5-7.5

12.0

12.0

Dense Graded Asphalt Concrete Surface Thickness, in.

Cement Treated Base Thickness, in. Granular/Aggregate Subbase Thickness, in.

Cement Treated Base Thickness, in. 7.0-8.0 Crushed Stone

8.0-9.0 Pit Run Gravel

Improved Subgrade , in.

Prepared Subgrade(See Section 5)

Controlled by AASHTO-PSI Criteria

Controlled by AASHTO-PSI Criteria

Note: The site condition cells that are shaded represent designs that have been used, but are considered “marginal” and may not be able to sustain the expected traffic for the given subgrade conditions (performance is highly dependent on materials).

Figure 5. Example structural design details for flexible pavement. 39

6.5-8.0 13.0

6.0 - 12.0

Flexible

Traffic: Subgrade:

Cell 14

4-8 million flexible ESALs Weak (Resilient Modulus, 4.5-9.0 ksi)

General Structural Design Inputs

Initial serviceability Terminal Serviceability Overall standard deviation Reliability

4.5 2.5 0.49 95%

Elastic modulus of surface HMAC Resilient modulus of subgrade 5 ksi Drainage coefficient, m

450 ksi 1.00

Note: Subgrade is very weak; some type of improvement should be considered (See Section 5). Note: See Sections 4A.2 through 4A.5 for detailed guidelines on other asphalt concrete pavement design features.

Structural Layer Thickness - Conventional Unbound Granular Base Dense Graded Asphalt Concrete Surface Thickness, in. Crushed Stone Aggregate Base Thickness, in. Granular/Aggregate Subbase Thickness, in.

7.0-8.5

7.0-8.5

9.0

9.0

15.0-16.0 Crushed Stone

16.0-17.0 Pit Run Gravel

Prepared Subgrade (See Section 5)

Dense Graded Asphalt Concrete Surface Thickness, in. Crushed Stone Aggregate Base Thickness, in. Granular/Aggregate Subbase Thickness, in. Improved Subgrade Thickness, in.

Note: A filter layer (or separator) is recommended between the subbase and weak subgrades (See Section 5).

Controlled by Asphalt Concrete Tensile Strain Under Wheel Load

6.5-7.5

6.5-7.5

9.0

9.0

11.0-12.0 Crushed Stone

12.0-13.0 Pit Run Gravel

6.0-12.0

6.0-12.0

Controlled by Asphalt Concrete Tensile Strain Under Wheel Load

Structural Layer Thickness - Full-Depth Asphalt Concrete A full-depth asphalt concrete pavement is not recommended for this site condition cell.

Figure 5. Example structural design details for flexible pavement (continued).

40

Step 7 - Determination of material requirements (Part 4A of Reference 16) Asphalt cement binder: AC-20 viscosity grade, or 60-70 penetration grade. HMAC surface and binder, crushed stone aggregate base, asphalt treated base, granular subbase. Improved subgrade. Step 8 - Recommended design features for “Conventional Unbound Granular Base” (Part 4A of Reference 16) HMAC Surface and Binder

= 7 - 8 in

Crushed Stone Aggregate Base

= 10 in

Crushed Stone Subbase

= 13 - 14 in

Improved Subgrade

= 12 in hydrated lime stabilized

Step 9 - Recommended design features for “Asphalt Treated Base” (Part 4A of Reference 16) HMAC Surface and Binder

= 5.5 - 6.5 in

Asphalt Treated Aggregate Base

= 8 in (plant mixed)

Granular (Pit Run Gravel) Subbase

= 14 in

Improved Subgrade

= 12 in hydrated lime stabilized

Step 10 - Subdrainage feature recommendations (Table 31 of Reference 16) Table 31 of Reference 16 recommends Level 3 subdrainage (requiring a treated base) or Level 4-Full subdrainage system for Design Cell 13. One option would be the asphalt treated base section with edge drains. Another option would be to design a full subdrainage design that includes a 4 in permeable asphalt treated layer just beneath the

41

asphalt treated base course (the granular subbase is reduced to 10 in and serves as a separator for the layer).

42

Step 11 - Cross-Section Of Pavement The cross section shown in Figure 4 of Reference 16 is appropriate for this project.

Example to Identify Rigid Pavement Design Features Step 1 - General project description Reconstruction of rural freeway pavement Interstate highway, two lanes in one direction. Uniform cross-section with 14 ft widened outer slab and 10 ft outer shoulder (2 ft widened lane plus 8 ft shoulder). Design period = 20 years. Step 2 - Determine site condition input for traffic (16) ADT (current)

= 20,000 two-directions.

Percent trucks = 6.0 Lane distribution factor of trucks = 0.81 Directional distribution factor of trucks = 0.5 (same number in each direction) Future growth of trucks = 6 percent per year, compounded GF = [ (1 + 0.06)20 - 1 ] / 0.06 = 36.79 Mean truck equivalency factor over 20 years (rigid ESAL/truck) = 2.0 Total ESALs = 20,000 * 0.06 * 365 * 0.81 * 0.5 * 2 * 36.79 = 13.1 million, outer lane, one direction. Step 3 - Determine site condition input for subgrade (16) Natural subgrade soil = silty clay Elastic k-values backcalculated form FWD deflection data from existing pavement from different seasons are shown in Table 3. The effective k-value is then calculated from those values using the procedure provided in Reference 16.

43

Table 3. Determination of seasonally adjusted effective subgrade k-value. Seasons

Backcalculated

Static k-value

W18

Relative

(3 months

Dynamic k-value

(psi/in)

(millions)

Damage

each)

(psi/in)

Spring

154

77

12.75

0.0784

Summer

196

98

13.15

0.0760

Fall

222

111

13.37

0.0748

Winter

336

164

14.20

0.0704

(1/W18)

Mean damage W18 Effective k-value

0.0749 13.3 million 110 psi/in

Step 4 - Determination of subgrade preparation/improvement needed (Part 4B, Part 5 of Reference 16) The seasonally adjusted subgrade k-value falls into the

Weak-fair

class.

Subgrade “improvement” is strongly recommended for this class of soil to provide a good construction platform and a more uniform support condition. Step 5 - Determine site condition input for climate (16) Freezing index

= 500 degree days below freezing 44

Average annual precipitation = 33 in The project site is located in a wet-freeze area.

Step 6 - Determination of recommended alternative rigid pavement types (Part 4B.2 of Reference 16) The rigid pavement design matrix with recommended PCC pavement design alternatives is shown in table 4. The site condition cell and the design alternatives of this example are identified as: Cell 14: Traffic T5 (13.1 million ESALs) and Subgrade Weak-Fair (effective k-value = 110 psi/in) Base type selected:

Lean concrete base

Feasible alternatives selected:JPCP with Dowels CRCP Example structural details for Cell 14 of the rigid pavement design catalog is given in figure 6.

45

Table 4. Site condition cells and design alternatives for the rigid pavement (16).

Note: Climatic site condition is considered in both the determination of the subgrade Seasonally adjusted k-value and by specific climatic variables specified within each design cell. ** Cumulative in design lane over design period.

46

Rigid

Cell 14

Traffic:

12-18 million rigid ESALs

Subgrade:

Weak/Fair (k-value of 100-200 psi/in)

General Structural Design Inputs Initial serviceability Terminal serviceability Overall standard deviation Reliability

4.5 2.5 0.39 95%

Elastic modulus of PCC PCC mean flexural strength Load transfer coef, J-value Drainage coef, Cd

4,000,000 psi 650 psi Varies w/ edge support type 1.05

Structural Layer Thickness - Doweled JPCP/JRCP Aggregate base

Treated base

Edge support type

Type I

Type II

Type III

Type I

PCC slab thickness, in

10-11

10.5-11.5

11-12

9.5-10.5

10.11

10.5-11.5

4-6

4-6

4-6

4-6

4-6

4-6

Base thickness, in

Type II

Type III

Prepared subgrade (See Section 5)

Transverse Joint Design Maximum Joint spacing, ft. JRCP:

Joint reservoir and other joint design features

45 ft

JPCP:

For recommended transverse joint reservoir width and other transverse joint design details, see Section 4B.8, "Joint Sealant Reservoir and Joint Sealants" and Section 4B.4, "Transverse Joints for JPCP and JRCP".

Edge support Climate

Type I

Type II

Type III

WF, DF

17-19

18-20

18-20

WNF

16-17

16-18

17-19

DNF

14-16

15-16

15-17

Other Design Features Tie bar design for longitudinal joints: Subdrainage design:

No. 5 (0.625 in diameter) deformed reinforcing bars spaced at 30 in

Level 3 - Edge drains and non-erodable treated base, or Level 4 - Full subdrainage system with permeable base.

Minimum % reinforcement content for JRCP: 0.19%-0.21% Dowel bar design: 1.25 in diameter corrosion-resistant dowel bars spaced at 12 in Note See Sections 4B.2 through 4B.12 for additional detailed guidelines on all the rigid pavement design features.

Figure 6. Example structural details for rigid pavement. 47

Step 7 - Determination of material requirements (Part 4B.2, 4B.11, 4C Table 28 of Reference 16) Portland cement concrete: 650 psi mean flexural strength of third-point loading at 28 days Lean concrete base:

Class A base is recommended, e.g. lean concrete with 7-8% cement

Improved subgrade:

12 in of granular material

Step 8 - Recommended design features for

JPCP with Dowels

(Part 4B of

Reference 16) Doweled JPCP Slab

= 9.5-10.5 in

Lean Concrete Base

= 4-6 in

Improved Subgrade

= 12 in aggregate

Transverse joint spacing = 17-19 ft Dowel bar design

= 1.25 in diameter corrosion-resistant dowel bars spaced at 12 in

Tie bar design

= No. 5 (0.625 in diameter) deformed bars spaced at 30 in

Step 9 - Recommended design features for

CRCP

(Part 4B of Reference 16)

CRCP Slab

= 9.5-10.5 in

Lean Concrete Base

= 4-6 in

Improved Subgrade

= 12 in aggregate

Reinforcement content

= 0.70%

Tie bar design

= No. 5 (0.625 in diameter) deformed bars spaced at 30 in 48

Step 10 - Subdrainage feature recommendations (Part 4C of Reference 16) Table 31 in Reference 16 of Part 4C recommends either Level 3 or Level 4 drainage design for Design Cell 14. Level 3 requires a treated base such as the lean concrete proposed for this project. Step 11 - Cross Section (Part 4B.1 of Reference 16) The cross section provided in Figure 18 in Reference 16 is appropriate for this project.

SUPPLEMENTAL PROTOTYPE KBES—Designer A prototype microcomputer-oriented KBES named “Designer” for selecting the recommended pavement design features was developed in this project. The inference engine of the prototype KBES is an expert system shell, CLIPS 6.04, and the user interface of the software is programmed in Microsoft Windows 3.11 environment using programming language, Visual Basic 4.0. The goal of the prototype KBES is to supplement the Catalog of Recommended Pavement Design Features. A fully developed KBES companion to the pavement design catalog adds value to the catalog in several ways: •

Increased speed and efficiency,

•

Guidance on obtaining inputs for site conditions (traffic, subgrade, climate)

•

Guidance to the designer in searching for solutions,

•

Explaining the logic of “best practice” recommendations, 49

•

Coordinating decisions about many different design features,

•

Making the catalog more dynamic and easier to update,

•

Enhanced value as a teaching and training tool,

•

Increased consistency in considering all feasible options and screening out infeasible options, and

•

Evaluate the selected pavement design features.

Most important, the KBES enhances the effectiveness, the implementability, and the adaptability of the catalog, all important factors in the catalog’s success. The KBES will help to accomplish the catalog’s goals and will help the user make better pavement design decisions more rapidly and more efficiently than before. One of the goals of the catalog is to put pavement design expertise in the hands of less experienced engineers in a format which permits them to apply that expertise appropriately. This is an ideal application for a KBES. The computerized KBES is highly adaptable, as the knowledge base (made up of the input guidance rules, catalog relational database, and alternative evaluation rules) is developed and expanded over time. Indeed, the typical program structure of a rule-based expert system is not rigid and is well suited to later additions and modifications. If the KBES is customized to an agency, no major change will be required since the rules and database are separate. If implementability is defined as the likelihood of the tool actually being put into routine use in State highway agencies, then the implementability of the KBES is believed to be as great as, if not greater than, that of the paper catalog. 50

The supplemental KBES will be a very powerful tool when fully developed. However, what was developed in this project is only a prototype KBES as specified in the work statement. The prototype KBES includes most of the current printed catalog. All the structural designs in each factorial design cell and key design features are included in the program. The prototype KBES is a very useful tool for knowledgeable users in its current state. It can be used to obtain the recommended design features for all the site condition cells included in the printed catalog. These recommended features and structural design ranges can then be used to compare and check the pavement design the user has selected. However, the prototype KBES needs to be improved in many ways before it can be used as an operational software product. A plan for extending the prototype KBES through implementation, evaluation, and validation has also been developed. An expert system normally goes through the following development stages: Development Stage

Description

Demonstration prototype

Solves a portion of the problem undertaken, suggesting that the approach is viable and system development is achievable.

Research prototype

Displays credible performance on the core problem but may be fragile due to lack of testing and revision. Some areas of the software may be incomplete.

51

Field test prototype

Implements all the technical and software features. Displays good performance with adequate reliability and has been revised based on extensive testing in the user environment.

Operational model

Exhibits high quality, reliable, fast, and efficient performance in the user environment.

Implementation

A production model used on a routine basis.

52

Designer is currently a research prototype expert system. It is very important to note that the scope of the current pavement design catalog and KBES is not adequate for project level design or for pavement type selection. The prototype KBES is intended to provide recommendations of design features and the framework to demonstrate the concept and the possible capability of a fully developed pavement design KBES. Through full development of the software, verification and validation, and rigorous field testing and evaluation, the prototype KBES can be extended into an operational KBES. The operational KBES development can be achieved through the following tasks. Task 1

Identify Additional Features for the Operational KBES

Designer fulfills most goals of the KBES as described in the previous section. Credible performance can be expected in representing the paper catalog. A detailed site condition input assistant was also well developed. On-line help texts provide explanations, background information, and general guidelines about the selected design features. However, this is only a research prototype expert system, many improvements should be made before it becomes a full blown KBES. This section presents several enhancements that can be made to improve the prototype KBES and extend it into a fully operational expert system. Task 1a

Enhancements for input assistant The input assistant for the site

conditions and other design inputs is well developed in the prototype KBES, and provides all of the information that is in the paper catalog. However, more details can be 53

added to this area, especially for traffic determination. A separate module can be developed to guide the designer to calculate the design ESALs for example. Task 1b

Improvements for catalog search capability The paper catalog is also well

included in the program data base and on-line help. Viable pavement design alternatives are efficiently searched using CLIPS. More work can be done to further refine the presentation of the design features. Task 1c

Fully development of the evaluation assistant Only a frame work and

mock-up demonstrations were provided to assist designers evaluating different design alternatives for certain site condition cells. This is the area that great deal of work is needed. In the prototype KBES, it is demonstrated through mock-ups that performance models of the key pavement distresses can be used to predict the future performance of the pavement design alternative selected. A few synoptic tables of the States' current pavement designs and practices are also included to illustrate how States' design practices can be presented to the designers for their reference. These capabilities need to be implemented fully for all the cells, prediction models, and design alternatives. Furthermore, linking the program to the LTPP performance data base, States' PMS data base, and other key performance databases can also be implemented in the next phase of the KBES development. New research products, new expertise acquired, and state-of-the-art technologies can also be incorporated into the KBES in the future. For example, SuperPave's binder grade selection scheme can be included in a future KBES. 54

Task 1d

Identify additional software features The prototype KBES has most

Windows software features such as multiple project interface, standard Windows buttons, menus, and mouse functions, some useful icons, and complete save of the user inputs during a project duration which gives consistent results. However, some additional features should be added for the operational KBES. For example, save and open menu items, complete print preview for each project, some additional useful icons, etc. Task 2 Complete development of the KBES After the identification of the additional technical and software features being identified, the following subtasks can then be established to fulfil the development of the KBES. Task 2a

Complete development of the input assistant

Task 2b

Improve the catalog search capability

Task 2c

Fully development of the evaluation assistant

Task 2d

Implement all the software features

The research prototype KBES is fragile due to incomplete testing and revision. Before the KBES becomes a field prototype, extensive debugging, verification and validation of the program must be accomplished. This is described in the next section. Task 3 Verification and Validation According to a research report from a recent FHWA study, Verification, Validation, and Evaluation of Expert Systems—An FHWA Handbook, verification shows that the system is built right while validation shows that the right system was built. Evaluation 55

reflects the acceptance of the system by the end users and its performance in the field. The following questions should be answered during the verification of the KBES: •

Does the design recommendations reflect the requirements?

•

Does the detailed design recommendations reflect the design goals?

•

Does the code accurately reflect the detailed design?

•

Is the code correct with respect to the language syntax?

•

When the program has been verified, it is assured that there are no bugs or technical errors.

The following issues should be addressed in the validation process: •

How well do inferences made compare with historic (known) data?

•

What fraction of pertinent empirical observation can be simulated by the system?

•

What fraction of model predictions are empirically correct?

•

What fraction of the system parameters does the model attempt to mimic?

During the verification and validation process, all the questions identified in this section will be addressed. Detailed guidelines for the rigorous testing of the program will be provided to ensure each and every feature of the software is tested. This is similar to the Beta testing of the general software products. This will include verifying that all the menus, buttons, and all other items are working as expected. The program should also be robust and user friendly. 56

57

Task 4 Evaluation of the KBES through Field Tests Real world problems will be used to test the program, and at a minimum, the following evaluation issues need to be addressed: •

Is the system user friendly and do the users accept the system?

•

Does the system give “correct” results and is the logic of the system correct?

•

Does the expert system offer an improvement over the practices it is intended to supplement?

•

Is the system useful as a training tool?

•

Is the system in fact maintainable by other than the developers?

At the evaluation stage, volunteers from various States will be identified to perform the software testing. Based on the discussion in the pavement design features' consensus group meeting conducted under Task 5, there is a lot of interest in performing the testing of the software. With the detailed testing guideline, the above stated questions will be addressed in the testing.

Task 5

Feedback and Improvements

Based on the feedback from the field test and evaluation, appropriate improvements will be made to the KBES and this could be an iterative process. Therefore, if additional testing is needed, the revised software will be tested in the field again, as necessary.

58

Implementation Plan of the KBES The ultimate goal of the development of the KBES is to provide the States with a useful design feature recommendation tool. Since the prototype KBES is based on the printed catalog, it is not intended for design or for pavement type selection. The current paper catalog was developed to provide general guidelines and recommendations on the best practices. The structural design outputs are in ranges, and the recommendations for the design features are the general guidelines. The KBES should be customized to the specific agency's conditions when being implemented for design. When implementing, the agency's specific policies, design methodologies, and local site conditions need to be taken into consideration. Also, accurate thickness design and very detailed pavement design feature recommendations need to be provided for each cell combination. Several States across the United States can be selected as demo installation States for the implementation of KBES. The general KBES will be first customized to the State’s specific policies and local conditions. The customized KBES will be provided to the States' engineers, and review comments will be requested from them. Then any problems or barriers will be identified toward implementing the software in the States. Once the problems are understood, the customized KBES can be improved by addressing all the problems and concerns. Some short training courses or workshops may be necessary during the implementation of the software. The primary objective of the workshop should be to 59

show the engineers that the KBES can help them do their jobs more efficiently and with few errors. The KBES can also be used to obtain all kinds of pavement design checks, performance predictions, and train young engineers. For the distribution of the software, McTrans and PCTrans software distribution centers are a good possibility. It is also possible to provide the program through Internet forums software distribution. Another good possibility is distribution as an AASHTOWare product because the States may wish to sponsor future enhancements to the KBES.

60

Chapter 4 Conclusions and Suggested Research

Conclusions This study into European and American highway pavement design systems has resulted in three major practical products for use by the state highway agencies.

Catalog of Current State Pavement Design Features. This document is a highly informative and practical guide on the details of the design practice for flexible and rigid pavements in the U.S. This catalog first and most importantly provides information in the form of a large Synoptic Tables Of State Pavement Design Features. This section provides for the first time for both flexible and rigid pavements a comprehensive summary of nearly all state’s design practices. •

The catalog provides, for each state highway agency (SHA), a State Factorial Design Matrix of Pavement Design Features that gives design layer thicknesses for a wide range of traffic, subgrade, and climatic site conditions for major types of main highway pavement constructed by the SHA.

•

The catalog provides a Catalog of Key Design Features by Climatic Region and Pavement Type.

61

•

A summary of key design features for flexible and for rigid pavements are given. This information will be of interest to highway administrators, pavement designers, contractors, industry, and others involved in various aspects of pavement design and construction.

62

Catalog of Recommended Pavement Design Features. This document includes the following. •

Recommended (good practice) design features for highway engineers, administrators, and others in a format that is easy to use and understand.

•

A model catalog presentation format for potential usage by agencies.

•

A recommended consensus on many design features for pavement design for varying site conditions.

The catalog recommendations are based an expert resource group, NCHRP panel, current SHA design practices, FHWA design manuals, the 1993 AASHTO Guide for Design of Pavement Structures, and mechanistic-empirical performance models that were used to limit the occurrence of key distress types.

Prototype Knowledge-Based Expert System (KBES). This software includes the following: •

An input assistant to provide interactive guidance to the designer in obtaining design inputs for the catalog site conditions and other inputs.

•

A database searching and presentation to access a project database which represents the paper catalog, to quickly and efficiently identify feasible design alternatives for a given set of site and design conditions.

63

•

An evaluation assistant to provide interactive guidance to the designer in evaluating the advantages and disadvantages of the various design alternatives and explanations about various design features.

The KBES is a prototype and must be further developed to be fully useful to state highway agencies. This is discussed in Chapter 3 under the title: Supplemental Prototype KBES - Designer.

Consensus Building. The catalog development process required input from all sectors of the highway industry. Thus, a large resource group of experienced pavement engineers from all sectors of the industry was assembled. The consensus group spent the week discussing and reviewing the draft Catalog of Recommended Pavement Design Features. The meeting produced a consensus on many design features and performance criteria. The most valuable aspect of the meeting was the many interesting and beneficial improvements proposed by the highly experienced and diverse resource group (See Appendix B for minutes). This approach was so successful that it should be considered for use on other studies where expert human experience is important.

Suggested Research •

The Catalog Of Recommended Pavement Design Features cannot be used for design by states and local highway agencies until it is adapted or customized to fit 64

their own cross-section designs, thickness design procedure, materials specifications, mix designs, and other standards. This would result in a “customized design catalog” for an agency that could then be used by the agency for project-level design purposes. The Prototype Knowledge-Based Expert System (KBES) also requires substantial additional development to make it fully useful to state highway agencies. A plan for accomplishing that is given in Chapter 3. This electronic version of the catalog was highly regarded in the survey taken of experienced pavement engineers. The electronic version would also provide for relatively easier upgrades than the printed version of the catalog.

65

References

1.

German Federal Ministry of, “Guidelines for the Standardization of the Upper Structure of Traffic-Bearing Surfaces.” RSto 86/89.

2.

German Federal Ministry of, "Guidelines for the Standardization of Pavements During the Renewal of Traffic Surfaces." RSto-E.

3.

Helenven, L., J. Verstraeten, and V. Veverka, "Latest Developments in the Analytical Methods for the Design of New Pavements and Strengthening Overlays in Belgium." Proc. Sixth International Conference, Structural Design of Asphalt Pavements, Vol. I, The University of Michigan, Ann Arbor, MI, July 1317, 1987, pp. 121-125.

4.

Veverka, V., J. Verstraeten, and F. Fuchs, "The Design of Concrete Pavements and Overlays(Belgium)." Proc. 6th International Symposium on Concrete Roads, Vol. I, Madrid, Spain, October 8-10, 1990, pp. 49-57.

5.

Rocci, S. and C. Kraemer, "The New Spanish Standard on Pavement Design." Proc. 6th International Symposium on Concrete Roads, Vol. I, Madrid, Spain, October 8-10, 1990, pp. 129-137.

6.

Director General of Highways, "Standards 6.1 and 6.2 -IC for Flexible and Rigid Roadways." MOPU, Madrid, 1975 (in Spanish)

7.

Director General of Highways, "Instructions 6.1 and 6.2 -IC for Roadway Sections." MOPU, Madrid, 1975 (in Spanish), 200 pp.

8.

Spanish Institute of Cement and Its Applications, "Manual of Concrete Pavements for Low-Volume Roads." Madrid, Spain, (1988) 47 pp.

9.

State Ports Board, "Guidelines for the Design and Construction of Port Pavements." Madrid, Spain, (1994) 164 pp.

10.

Litzka, J. and G. Herbst, "A New Specification for the Structural design of Pavements in Austria." Proc., The 1986 International Conference on Bearing Capacity of Roads and Airfields, Plymouth, England, Sept. 16-18, (1986).

66

11.

Scetauroute, "Manual of Design for Superhighway Pavements." Paris, France, March 1994.

12.

LROP and LREP, "Design Strategies for Pavement Structures for Paris Roadways." May 1993.

13.

Boissoudy, A. D., M. T. Goux, and P. Genre, "New Concrete Pavement Structures in the Revised French Catalogue for New Pavements and Guidelines for Overlay Design." Proc. 6th International Symposium on Concrete Roads, Vol. I, Madrid, Spain, October 8-10, (1990) pp. 21-30.

14.

Scholtes, Peter R., The Team Handbook, Joiner Associates, Inc. 1988.

15.

Jiang, J., B. Killingsworth, M.I. Darter, H. Von Quintus, E.B. Owusu-Antwi, “Catalog of Current State Pavement Design Features,” Interim Report, NCHRP Project 1-32,, Transportation Research Board, 1996.

16.

Darter, M.I., H. Von Quintus, J. Jiang, E.B. Owusu-Antwi, B. Killingsworth, “Catalog of Recommended Pavement Design Features,” Interim Report, NCHRP Project 1-32, Transportation Research Board, 1997.

67

APPENDIX A SUMMARY OF EUROPEAN DESIGN CATALOGS

There are several countries, including Germany, France, Austria, Spain, Belgium, Japan and Switzerland, that use pavement design catalogs. To learn from their experience, the catalogs from these countries were thoroughly reviewed, and information was sought from representatives of their highway agencies. This appendix presents information obtained from the state-of-the-art review of the catalogs. It is important to note that these countries are smaller in size than the United States and generally have more uniformity in climate and, perhaps, other conditions. In addition, the information was primarily obtained from translation of agency document and some interpretation problems may exist. However, all possible attempts have been made to have representative from these countries verify the information obtained. Following are the details of the findings from the review of the European catalogs.

Germany The first German catalog was published in 1966. According to Mr. Peter P. Canisius of the Federal Transportation Ministry, the development of a pavement design catalog in Germany arose out of the desire by the government to provide guidelines for the standardization of pavement designs. The goal was to provide standard pavement structures for building and maintaining pavements for highways and other traffic bearing A-1

surfaces, that are based on technically suitable and economic methods of construction. The current German catalog for new pavement design that is fully accepted in practice is entitled Guidelines for the Standardization of the Upper Structure of Traffic-Bearing Surfaces (RSto-86/89). A companion catalog entitled Guidelines for the Standardization of Pavements During the Renewal of Traffic Surfaces (RSto-E) was being finalized in 1994. That catalog supplements RSto-86/89 and is for the renewal (rehabilitation) and reconstruction of existing pavements. The guidelines provided in RSto 86/89 relate traffic loading, the method of construction, and the properties of soil and other materials. Although the catalog was initially based on engineering judgment, the results of theoretical calculations, and laboratory and fields tests, as well as practical experience have been incorporated into it over the years. The failure criteria for the design procedure on which the catalog is based are fatigue cracking for asphalt pavements (instead of absolute values, a comparison between the fatigue for different structures was considered), and bending stresses for concrete pavements. The table of contents of the German pavement design catalog is provided in table A-1.

A-2

Table A-1. Table of contents of the German pavement design catalog. Chapter

Content

Page

1

General

11

2

Fundamentals

12

2.1

Terms

12

2.2

Evaluation and choice of method of construction

14

2.3

Construction classes and traffic loading

15

2.3.1

Carriageways

16

2.3.2

Bus-traffic-bearing surfaces

17

2.3.3

Parking areas

18

2.3.4

Traffic-bearing surfaces in subsidiary installations and subsidiary enterprises on Federal trunk roads

18

2.3.5

Abnormal stresses

19

Technical Regulations

19

Determination of the thickness of the frost-resistant upper structure

21

3.1

Frost sensitivity of the soil

22

3.2

Standard values for the thickness of the frost-resistant road structure

22

3.3

Increased or decreased thickness due to local conditions

23

Standardized upper structure for carriageways

24

4.1

Methods of construction and layer thickness

24

4.2

Construction and execution

25

4.2.1

Subsoil or subgrade

25

4.2.2

Base courses

26

4.2.3

Bituminous surfaces

27

2.4 3

4

A-3

Table A-1. Table of contents of the German pavement design catalog (continued).

Chapter

Content

Page

4.2.4

Concrete surfaces

28

4.2.5

Sett surfaces

28

4.2.6

Drainage

29

4.2.7

Special features

29

Traffic-bearing surfaces in a closed local situation

30

Standardized upper structure for other traffic-bearing surfaces

31

5.1

Carriageways at autobahn junctions and access points

31

5.2

Multipurpose lanes

31

5.3

Hard shoulders

31

5.4

Central reserve crossovers

31

5.5

Buss-traffic-bearing surfaces

31

5.5.1

Methods of construction and layer thicknesses

31

5.5.2

Construction and execution

32

5.6

Cycle tracks and footways

32

5.6.1

Methods of construction and layer thicknesses

32

5.6.2

Construction and execution

33

Parking areas

33

5.7.1

Methods of construction and layer thicknesses

33

5.7.2

Construction and execution

34

Traffic-bearing surfaces in subsidiary installations and subsidiary enterprises on Federal trunk roads

34

Methods of construction and layer thicknesses

34

4.3 5

5.7

5.8 5.8.1

A-4

Table A-1. Table of contents of the German pavement design catalog (continued).

Chapter 5.8.2

Content

Page

Construction and execution

35

5.9

Fire brigade roads

35

5.10

Rail areas in roads

35

Appendix I

Determination of the relevant traffic loading number

36

Appendix II

Illustration of construction methods for roads arranged according to construction classes

40

The following factors are used in the German catalog to classify the site conditions for the different structural categories: 1. Traffic is classified using a traffic load index (VB) with 7 levels. VB is the decisive factor for the classification of the structural categories in the German catalog and is determined by the following parameters: •

the average daily volume of vehicles of heavy traffic type;

•

the average change in this traffic during the design life

•

the number of traffic lanes in the cross-section

•

the width of a traffic lane

•

the longitudinal profile gradient.

If the VB cannot be computed for a specific pavement because of lack of data, then the road type—such as a high speed traffic road, traffic lanes in bus stations, parking areas that are used continuously—is used to assign the traffic level. A-5

2. Frost sensitivity classes are determined by soil classification as F1, F2 and F3. They are used to determine the thickness of the granular blanket (sand and gravel) for frost protection for each construction class. Germany believes strongly in preventing frost heave. 3. The following five local conditions are used to adjust the thickness of the pavement structure if necessary: •

Frost heave protection

•

Subgrade (cut/fill, embankment)

•

Location of project (e.g. northern/southern exposure)

•

Drainage condition

•

Type of marginal area (e.g. lateral strips, cycle tracks, footway)

4. Guidelines for soil consolidation and improvement with binders are contained in the German standard ZTVV-StB. The classification of the soil—suitable soil, conditionally suitable soil, or unsuitable soil and rock types—is used to determine the applicability of the standard. Rather than include subgrade strength as a design factor, a minimum bearing capacity is specified on top of the embankment and also on top of the granular blanket if provided. This provides a uniform subgrade support (guaranteed minimum support) upon which the AC or PCC pavement is constructed. Also, construction site traffic and paving equipment can be utilized without problems. The minimum foundation/embankment bearing capacity required is 45 MPa (6,000 psi), and the minimum bearing capacity on top of granular blanket layer required is A-6

110 MPa (16,000 psi). Bad areas that are located are removed and replaced with sand and gravel. Specific procedures are provided for the disposal of bad soils. The steps involved in following these guidelines include the evaluation and choice of method of construction (type of pavement), and the determination of construction class, traffic loading class, and thickness of the frost-protection upper structure. Based on this information, a pavement section can be selected from the catalogs for asphalt concrete or portland cement concrete surfaces. Examples from the German catalog for new AC pavements and PCC pavements fully bonded with the upper pavement layers, and JPCP overlays are shown in figures A-1 and A-2, respectively. The German catalog is supported by several specifications and technical guidelines, including guidelines for earth work and the improvement of the existing soil for road construction. For example, in German practice, it is believed that rutting of flexible pavement is scarcely influenced by thickness guideline, but by the quality of the asphalt mixture (crushed stone) and a good bond at the layer interfaces. Concrete slab are bonded to treated base courses, and joint are used in lean concrete and cement treated bases. In addition to these catalogs, guidelines, and technical recommendations that are issued by the Federal Government, certain industry organizations have developed publications using the information available in national catalogs and guidelines, for the transfer of technology to their constituents. An example is the Concrete Highway

A-7

Engineering and Subbases published by the Federal Association of the German Cement Industry.

Figure A-1. A portion of the German pavement design catalog for AC and PCC pavement layers. (Note thicknesses given are in cm.)

A-8

JPCP overlay on interlayer of lean concrete or AC on fractured old pavement.

JPCP overlay on geotextile on fractured old pavement

Figure A-2. A portion of the German pavement design catalog for JPCP overlays.

With the guidelines provided in these catalogs, for the same inputs, designers, contractors, and the industry that use them will obtain the same designs. This reduces errors in pavement design, especially by inexperienced engineers, and good performance is obtained if these proven guidelines are followed. According to Professor Dr.-ing G. Leykauf of the Technical University of Munich, another advantage of using the German catalog for design is that the number of possible technical solutions for any particular case is limited. This improves communication between the client and contractor. However, Professor Leykauf points out that, in Germany, there seems to be a certain regularity in complaints that arise every 5 to 10 years regarding the incorporation of new innovation into the catalog. The most recent example is a discussion concerning which

A-9

kind of wearing course (dense asphalt concrete or split-mastic asphalt or Gubasphalt) for asphalt pavement must be used in a cost comparison concrete structures. As a result, the catalog needs to be regularly reviewed to allow for new innovations to be incorporated. Also, according to Dr. -ing A. Zachlehner of Germany, the main concept of the catalog is to provide ready-made structural pavement alternatives for a given amount of traffic and, to a minor degree, climatic conditions. Because of this, the treatment of traffic loading in the catalog is of great importance, especially for local administrations without a large technical staff or experience. Since one characteristic of the design procedure is to fix the bearing capacity required for the subgrade independent of local conditions, the influence of the subgrade conditions are eliminated as design parameters. This may not always lead to the most economical solution.

Belgium Two catalogs are available for pavement design in Belgium. They are the Code of Good Practice for Cement Concrete Pavement Design and the Code of Good Practice for Bituminous Pavement Design developed by the Belgian Road Research Center. The recommendations provided in the catalogs were prepared in consultation with representatives from various professional circles: highway agencies, contractors, asphalt and concrete industries, and material producers. The major advantage of the methods for design provided in the catalog is that the designs recommended are based on Belgian

A-10

conditions (traffic, pavement materials, soils, and climate). The catalog is also based on a rational structural design method. Citing the need to avoid the development and use of complex models that hinder practical applications as a goal, the Belgian Road Research Center adopted the elastic layered theory for the development of the structural design method on which the catalogs are based. Fatigue cracking and rutting due to excessive deformation of the subgrade were used as the failure criteria for developing the design method for asphalt pavements. For concrete pavements the failure criteria used were fatigue failure due to combined traffic loading and thermal curling stresses, and vertical deformation of the subgrade. The critical state of the pavement is assessed from cracking, rutting, and longitudinal profile measurements. JPCP and CRCP are the types of the concrete pavements utilized. The following variables are required as inputs for the catalog: 1. Traffic is categorized into three levels using the cumulative number Nc of commercial vehicles which is determined from the following: •

the spectrum of traffic loads

•

the average number of axles per commercial vehicle

•

the impact coefficient of the dynamic effects on different roads

•

the maximum gradient of temperature: 70 oC/m (frequency 0.1%)

2. Subgrade is characterized by the dynamic modulus Es, which can be estimated by the soil CBR value or the soil class and drainability of the subgrade. For Nc greater than 104 (i.e. heavy trafficked highway), Es is required to be equal to or greater than 20 A-11

MPa (2,900 psi). The subgrade must be treated or improved with added material if the 20 MPa minimum bearing capacity requirement is not met. However, there are four other levels of Es for low volume roads.

The table of contents of the Belgian pavement design catalogs for asphalt and concrete roads are shown in the tables A-2 and A-3, respectively. Examples of the thickness design for concrete, flexible and semi-rigid (e.g. asphalt over lean concrete) pavements are provided in figures A-3 through A-6.

A-12

Table A-2. Table of contents for the Belgian design catalog of asphalt roads. Chapter Part One

Content Synopsis

1

1

Input parameters

2

Input parameters for conditions generally encountered in Belgium

3

Design method for conditions generally encountered in Belgium

4

Cases in which typical conditions are realistic, with respect to traffic

5

Design for freezing and thawing

6

Design examples

Part Two

3

Comments on the design method

Destructive effects of loads: equivalency laws (rules)

1.2

Taking traffic into consideration

1.3

Determination of the number of commercial vehicles traveling in the right lane during the service life of the pavement Consequences on the design of pavements of error in the estimation of the number of commercial vehicles Particular traffic conditions

2

23

33

41 43 46 46 51 52

Soil Soil modulus (Es)

57

2.1

Introduction

2.2

Geotechnical study

2.3

Modulus of the foundation soil

2.4

22

Traffic

1.1

1.5

13

35

Number of commercial vehicles to consider

1.4

4

26

Introduction 1

Page

59 59

Frost-susceptibility of foundation

A-13

62 68

A-14

Table A-2. Table of contents for the Belgian design catalog of asphalt roads (continued). Chapter 3

Content

Page

Climatic effects

3.1

Introduction

75

3.2

Seasonal variations in water content

75

3.3

Seasonal variations in temperatures

76

3.4

Methods for protection against climatic effects

78

4

Materials Choice of materials

4.1

Introduction

83

4.2

The role and functioning of the layers constituting the structure

83

4.3

The rules governing the behavior of materials

88

5

The design method Variations of the solution researched

5.1

Introduction

97

5.2

The rational method

97

5.3

The practical method

98

Economic aspects

105

References

109

6

A-15

Table A-3. Table of contents for the Belgian design catalog of concrete roads. Chapter

Content

Page

1

General

11

2

Fundamentals

12

2.1

Terms

12

2.2

Evaluation and choice of method of construction

14

2.3

Construction classes and traffic loading

15

2.3.1

Carriageways

16

2.3.2

Bus-traffic-bearing surfaces

17

2.3.3

Parking areas

18

2.3.4

Traffic-bearing surfaces in subsidiary installations and subsidiary enterprises on Federal trunk roads

18

2.3.5

Abnormal stresses

19

Technical Regulations

19

Determination of the thickness of the frost-resistant upper structure

21

3.1

Frost sensitivity of the soil

22

3.2

Standard values for the thickness of the frost-resistant road structure

22

3.3

Increased or decreased thickness due to local conditions

23

Standardized upper structure for carriageways

24

4.1

Methods of construction and layer thickness

24

4.2

Construction and execution

25

4.2.1

Subsoil or subgrade

25

4.2.2

Base courses

26

4.2.3

Bituminous surfaces

27

4.2.4

Concrete surfaces

28

4.2.5

Sett surfaces

28

2.4 3

4

A-16

Table A-3. Table of contents for the Belgian design catalog of concrete roads (continued). Chapter

Content

Page

4.2.6

Drainage

29

4.2.7

Special features

29

Traffic-bearing surfaces in a closed local situation

30

Standardized upper structure for other traffic-bearing surfaces

31

5.1

Carriageways at autobahn junctions and access points

31

5.2

Multipurpose lanes

31

5.3

Hard shoulders

31

5.4

Central reserve crossovers

31

5.5

Buss-traffic-bearing surfaces

31

5.5.1

Methods of construction and layer thicknesses

31

5.5.2

Construction and execution

32

5.6

Cycle tracks and footways

32

5.6.1

Methods of construction and layer thicknesses

32

5.6.2

Construction and execution

33

Parking areas

33

5.7.1

Methods of construction and layer thicknesses

33

5.7.2

Construction and execution

34

Traffic-bearing surfaces in subsidiary installations and subsidiary enterprises on Federal trunk roads

34

5.8.1

Methods of construction and layer thicknesses

34

5.8.2

Construction and execution

35

5.9

Fire brigade roads

35

5.10

Rail areas in roads

35

Appendix I

Determination of the relevant traffic loading number

36

Appendix II

Illustration of construction methods for roads arranged according to construction classes

40

4.3 5

5.7

5.8

A-17

Figure A-3. Examples of Belgian standard concrete pavement structures for heavily trafficked roads.

Figure A-4. Belgian standard design of semi-rigid pavement structure.

A-18

Figure A-5. Belgian design chart for the thickness of the bituminous layer and thickness of the crushed stone base in relation to traffic.

Figure A-6. Belgian design thickness of the granular subbase in relation to traffic and subgrade modulus.

A-19

The catalogs are available in a simple computer program that includes the equations of the various charts and structures (flexible, semirigid and rigid pavements) provided in the catalog. JPCP and CRCP are the types of the concrete pavements utilized. The design methods on which the catalogs are based, are periodically updated to take new conditions (mainly traffic) into account.

Spain The first Spanish catalog was issued in 1975. Using the experience acquired from the application of the 1975 catalog, and to take into account new materials and the evolution of traffic, the second and current edition of the catalog was issued in 1989. Given the time that elapsed between the two editions, there were corrections of some underdesigned structures, the incorporation of new possible solutions, improvement of standard specifications for existing materials, and the inclusion of standards for new paving materials like porous asphalt, lean concrete and RCC in the current catalog. The current catalog also incorporates the knowledge from procedures that were urgently developed in 1986, for the design of pavements during the "Autovía" (freeway) construction program (1984-1991). The catalog is for the design of all pavement types for all categories of heavy traffic, and provides the design engineers with a range of solutions for different heavy traffic and subgrade bearing capacity categories. Final solutions are selected based on technical and

A-20

economic reasons, such as type and availability of local materials. The table of the content of the Spanish pavement design catalog is given in table A-4 for illustration. Table A-4. Table of contents of the Spanish pavement design catalog. Chapter

Content

Page

Presentation

9

1

Object

11

2

Scope of Application

13

3

Factors of Dimensioning

15

3.1 Heavy Traffic

15

3.2 Subgrade

16

3.3 Materials of the Pavements

19

4

Catalog of Pavement Sections

27

5

Shoulders

31

6

5.1 Categories of Heavy Traffic T0 and T1

31

5.2 Category T2

31

5.3 Categories T3 and T4

32

Joints of Vibrated or Compacted Concrete

35

6.1 Longitudinal Joints

35

6.2 Transverse Joints

37

Appendix I

Schemes of Solutions of Pavements with Mainline and shoulders for Each Type of Section in the Catalog

39

Appendix II

Definitions

197

A-21

Although the Spanish pavement design catalog is empirically based, sections have been checked using analytical methods to estimate stresses and strains under loads, and their fatigue strength. Asphalt pavements were modeled as multi-layer systems using a mechanistic approach that took factors such as material elastic parameters and fatigue laws into account. The rigid pavements were verified using the PCA (1984) procedure. The failure criteria used in these analyses were fatigue cracking for asphalt pavement and cracking and faulting for concrete pavements. Faulting was only considered for undowelled jointed PCC pavements. As further mitigation against failure, dowels, as well as lean concrete bases, are required for concrete pavements to avoid faulting failures. Also, very stiff asphalt mixes (approximately 6,000 MPa [870 ksi] at 25 oC) that are highly resistant to rutting are used for flexible pavements. It should be mentioned that these stiff mixes for flexible pavements were introduced prior to 1975 through new specifications, to withstand much heavier loads (130 kN [29,224 lb] legal single axle loads and 210 kN [47,208 lb] legal tandem axle loads), and tire pressures in excess of 0.9 MPa [130.5 psi], than are experienced in the United States. Table A-5 shows the five categories (T0 to T4) that are specified in the catalog for classifying traffic loading for pavement design. The classification is on the basis of the average daily truck traffic (ADTT) in the design lane during the first year of service. As shown in table A-6, there are three categories (E1 to E3) of subgrade support established A-22

in the catalog based on the CBR value. There are additional specifications for subgrade grading, plasticity of granular materials, and the minimum binder percentage or compressive strength for soil cement and lime treated layers. The design catalog includes tables that provide layer thicknesses for the different traffic categories. Examples are shown in tables A-7, A-8 and A-9. The thicknesses shown in the tables are minimum requirements; slightly thicker layers are constructed to assure compliance at all points. Design features such as dowel presence, joint spacing and shoulder design are determined by the traffic load classification.

Table A-5. Categories of truck traffic in Spanish catalog.

ADTT T0

> 2.000

T1

2.000 - 800

T2

800 - 200

T3

200 - 50

T4

< 50

A-23

Table A-6. Subgrade categories in Spanish catalog.

E1

5 < CBR < 10

E2

10 < CBR < 20

E3

20 < CBR

Table A-7. Examples of AC pavement structures from Spanish catalog (thicknesses in cm). Traffic

T1

Section no. Asphalt concrete

121 122 123 124 131 132 133 134 30

25

25

CTB

30

25

22

22

Soil-cement

20

Crushed gravel

20

Uncrushed gravel Subgrade

15

25

20

20

20 25

25

E1

E2

A-24

15

E3

20

Table A-8. Minimum layer thicknesses provided in Spanish catalog for vibrated plain concrete pavements (in cm).

Heavy traffic

T0

T1

T2

Vibrated concrete

28

25

23

Lean concrete or CTB

15

15

15

Uncrushed gravel

20

-

20

-

20

-

-

Subgrade

E2

E3

E2

E3

E1

E2

E3

Table A-9. Minimum layer thicknesses provided in Spanish catalog for pavements with roller compacted concrete (in cm).

Heavy traffic

T0

T1

T2

Asphalt concrete

10

10

8

RCC

25

22

20

Soil-cement

20

20

20

Uncrushed gravel

-

-

-

-

20

-

-

Subgrade

E2

E3

E2

E3

E1

E2

E3

A-25

The practicality of the Spanish catalog after 20 years of use has been proven, and it is fully accepted by government officials, the asphalt and cement industries, consulting engineers and contractors. Consequently, apart from the national catalog for highway pavement design issued by the government, there are other catalogs available for pavement design for other facilities. Two examples are the Manual of Concrete Pavements for Low-Volume Roads published by the Spanish Institute of Cement and Its Applications (IECA) and the Guidelines for the Design and Construction of Port Pavements of the State Ports Board. Professor Carlos Kraemer of the Polythenic University of Madrid, Spain, who contributed to the development of all the three catalogs cited in this section, points to several advantages that have been experienced in Spain by the catalog design approach doing the following: 1. Establishing uniform design criteria that are based on standard specified materials of known quality. 2. Providing engineers (including those who are not specialized in pavement design), contractors, and politicians, with a practical tool that reduces the risk of misunderstandings and errors inherent in the communication of pavement design solutions. 3. Improving the average performance of pavements by eliminating, through experience, under- and overdesigned solutions.

A-26

4. Allowing the designer to focus on the choice of the most appropriate solution in each specific case on the basis of the availability of materials and cost. 5. Reducing the number of alternative pavement structures that contractors have to become familiar with. 6. Simplifying supervision of design, construction quality control, assessment of performance and, in short, maintenance management. 7. Allowing the progressive optimization of the design process through performance observations. 8. Protecting the civil engineers from politicians, who are sometimes more interested in cheap, short life solutions (slightly longer than their election term), but that may essentially be non-economical. 9. Some of the noted disadvantages of using a catalog are as follows: 1. There is a risk to introducing a routine design process. Pavement design is sometimes eventually considered as a solved problem. Hence, fewer engineers are motivated to become specialist in the area. 2. It may be more difficult to introduce new materials, even with standard structural designs that are open to initiatives (for instance, high modulus asphalt mixes used instead of conventional ones).

A-27

3. The catalog may be less adaptable to regional or local conditions, such as extreme climates, local marginal materials, particular local aggregates (e.g. volcanic materials that may not be in catalog). 4. There can be a certain lack or flexibility to adapt to budget limitations.

Austria The Austrian specifications for the structural design of pavements provides thicknesses based on the design traffic level and the bearing capacity of the subgrade. The catalog allows for a minimum bearing capacity of 35 MPa (5,075 psi) for the subgrade as determined by plate load tests. The traffic level is determined from the daily traffic load volume (DTLV), a distribution factor and truck factor. The catalog requires the application of necessary subgrade strengthening measures, including soil stabilization or replacement with better materials, in order to obtain the appropriate bearing capacity for all soils. For each traffic class and the specified minimum subgrade bearing capacity, the catalog provides designs for the following five construction types: 1. Flexible pavement consisting of bituminous surfacing and road base, on unbound base of natural gravel or crushed material. 2. Flexible pavement consisting of bituminous surfacing and road base, on an unbound base of wet-mix aggregate.

A-28

3. Semi-rigid (composite) pavement type consisting of a bituminous layer on a cement treated base. 4. Concrete pavement on an unbound subbase. 5. Concrete pavement on a cement treated subbase.

The most appropriate choice is made based on an economic evaluation. The design thicknesses are presented in the design tables for rigid pavements and for flexible and composite pavements, respectively, presented in figures A-7 and A-8. The catalog is based on a design procedure that involved the use of a variety of closed-form equations and an equation derived from correlation studies, to determine the permissible number of load applications to failure, N, for each of the pavement types. For the flexible pavement types, for a 20-year design life and by applying multi-layer elastic theory using the BISAR computer program and taking seasonal effects into account, Miner's hypothesis was used to determine the required pavement structure for each of the traffic loading and subgrade bearing capacity combinations. Similarly, for rigid pavements and for a 30-year design life, using Eisenmann's method for computing stresses in pavements and Westergaard's closed-form solution for computing stresses at the slab edge due to traffic loading, the required pavement structure was determined for the various traffic loading and subgrade bearing capacity combinations. Thermal curling stresses were taken into account in the computations.

A-29

Minimum subgrade elastic modulus

A-30

Figure A-7. Design table for rigid pavements in Austria (thicknesses in mm).

A-31

Figure A-8. Design table for flexible and composite pavements in Austria (thicknesses in mm). A-32

The pavement thicknesses provided in the catalog are reported to be valid only for normal conditions (e.g. free flowing traffic, usual subsoil, climate and favorable drainage conditions), and for pavements that do not involve engineer structures. Special provisions, including the use of specific mix design procedures for road materials, are required for road sections that experience predominantly slow moving commercial traffic or high shear stresses. Also, the catalog provides specific guidelines for pavements on bridges and in tunnels.

France There are several catalogs that are used for pavement design in France. The current national catalog was developed in a collaborative effort by three governmental agencies (SETRA, LCPC, and CETE), with the participation of representatives from the French ministry in charge of transportation, the highway sector, and the highway engineering profession. A semi-private tollway company called SCETAUROUTE, published the third edition of its catalog for superhighways (or freeways) in March 1994. Regional and local catalogs are also available. An example is the catalog entitled Design Strategies for Pavement Structures of Paris Roadways, developed by the West and East (LROP and LREP) Greater Paris Region, Road Research Laboratories, in collaboration with the Department of Highways of the City of Paris. These examples provide evidence that a national catalog can serve as basis for the development of specialized catalogs for specific regions, locales, or highway types. A-33

The first national catalog issued in 1977 for both AC and PCC pavements, was developed by the Service of Technical Studies of Roads and Superhighways (SETRA) and the Central Laboratory for Bridges and Pavements (LCPC). Initially, this catalog proposed the design of rigid pavements using the California technique (non-dowelled short slabs). This catalog was updated in 1988 by SETRA, LCPC, and the Southeast Center for Technical Studies of Equipment (CETE), to include continuously reinforced pavements, doweled pavements, and non-erodible foundations (stabilized layers). The current SETRA-LCPC-CETE catalog is a comprehensive document that consists of the following: •

a catalog of standard structures for new pavements,

•

a technical guide for establishing cross-sections of pavements, and

•

a users guide.

A number of computer programs are also available with the catalog that can be used for economic comparisons of different pavement structures and material quantities for the different techniques. The catalog of standard structures provides guidelines for the design of 26 types of pavements structures, including flexible, semi-rigid (composite), and seven concrete pavement types. Typical examples for flexible and rigid pavements are shown in figures A-9 and A-10. The design lives for the flexible, composite, and rigid pavements are 15, 20 and 25 years, respectively. In general, due to better construction control, the

A-34

thicknesses provided for freeways are 2-3 cm thinner than those of the primary and secondary roads.

Figure A-9. Typical structure for flexible pavement in the French catalog.

A-35

Figure A-10. Typical structure for a CRCP on lean concrete base in the French catalog. The technical guide for establishing transverse pavement cross-sections summarizes certain directives concerning the geometry, construction and constitution of various elements of the pavement. Examples of the recommendations provided for the pavement cross-section, site vegetation, and drainage are shown in figures A-11 and A-12, respectively. For all the pavement structures, in order to use the catalog, it is necessary to determine the traffic level and subgrade or platform support using procedures explained in the user's guide. Four categories of traffic loading (T0 to T3), corresponding to the initial average daily truck traffic (ADTT) count in the design lane for trucks with weights A-36

greater than or equal to 50 kN (5 tonnes), are defined in the catalog. An annual traffic growth rate of 7% during the design life is assumed. The subgrade is classified into three "platforms" (PF1 to PF3), with corresponding subgrade modulus values that range from 200 to 1,200 MPa (29 to 174 ksi). Charts with guidelines for frost protection of the pavement are the only adjustments made for climatic conditions. All the other catalogs, such as the SCETAUROUTE catalog for superhighways and the regional catalog for Greater Paris, are based on the same basic principles introduced in the national catalog. In fact, the SCETAUROUTE catalog includes several references to the SETRA-LCPC catalog. The table of contents for SCETAUROUTE pavement design catalog is given in table A-10. Examples of charts from the SCETAUROUTE catalog for flexible and concrete pavement design are shown in figures A-13 and A-14, respectively. For each type of structure, the thickness of the pavement layers can be determined from a chart using the two parameters Ti (traffic loading) and PFi (subgrade).

A-37

Figure A-11. Recommendations for pavement cross-section in the French catalog.

Figure A-12. Example of site vegetation and drainage guidelines in the French catalog.

A-38

Table A-10. Table of Contents of SCETAUROUTE pavement design catalog. Chapter

Content

Page

Introduction

1

1

Methodology

3

2

Traffic

7

3

Support platform of pavement subgrade

13

4

Techniques of pavement construction

21

5

Pavement structures

53

6

Transverse cross section

71

7

Maintenance scenarios

85

8

Verification of freeze-thaw

89

9

Pavement structures for (traffic) areas

95

10

Final surface finishing

99

Appendix 1

101

Appendix 2

103

Appendix 3

105

Appendix 4

107

Bibliography

111

A-39

Figure A-13. An example of flexible pavement design in SCETAUROUTE catalog (thicknesses in cm). A-40

Figure A-14. An example of rigid pavement design in SCETAUROUTE catalog (thicknesses in cm). A-41

An annual traffic growth rate of 4% is assumed in the SCETAUROUTE catalog. For a different annual growth rate, the catalog recommends changes in the dimensions of the pavement structure. In an apparent reference to the rate of 7% used in the national catalog, the SCETAUROUTE catalog notes that the use of a 7% annual traffic growth rate instead of 4%, will lead to an overestimation of the base thickness by 1 cm (0.4 in). For a given traffic and subgrade bearing capacity, and the condition that the required surface maintenance is carried out, the different structures proposed in the catalog will more or less guarantee equivalent life spans.

Swiss Catalog The Swiss pavement design standard consists of technical specifications for the subgrade and a catalog of pavement structural designs. The Swiss pavement design catalog is also based on thickness determination for classifications of the traffic loading and subgrade properties. Tables A-11 and A-12, respectively, show the classification for traffic loading and subgrade support provided in the Swiss catalog.

A-42

Table A-11 Classification of traffic in Swiss catalog.

Traffic Class

Equivalent daily traffic weight (TF)

T1 very light

10 to 30

T2 light

> 30 to 100

T3 medium

> 100 to 300

T4 heavy

> 300 to 1000

T5 very heavy

> 1000 to 3000

Table A-12. Classification of subgrade in Swiss catalog. ME1 (kN/m2)

CBR (%)

k (MN/m3)

6000 to 15,000

3 to 6

15 to 30

S2 medium

> 15,000 to 30,000

> 6 to 12

> 30 to 60

S3 strong

> 30,000 to 60,000

> 12 to 25

> 60 to 100

> 60,000

> 25

> 100

Subgrade class S1 weak

S4 very strong

A-43

The catalog contains matrices of designs for several flexible and concrete pavement types. Figure A-15 shows an example of the design matrix for AC pavement on hot-mix asphalt and aggregate base, and figure A-16 gives an example of the design matrix for PCC pavement on aggregate subbase. The catalog also provides specifications for layer stabilization, minimum layer thickness for freeze/thaw conditions, base and subgrade uniformity, and layer compaction.

A-44

Figure A-17. Swiss design matrix for AC pavement on hot-mix asphalt and aggregate subbase. A-43

Figure A-18. Swiss design matrix for PCC pavement on stabilized subbase.

A-44

Appendix B MINUTES OF NCHRP PROJECT 1-32 CONSENSUS MEETING January 22-26, 1996, Chicago In Attendance: 1. Resource group members (note, where ever the following text refers to resource group members, it means only the following individuals): Sohila Bemanian James Brown Ray Brown Bill Cape Max Grogg Wouter Gulden Marlin Knutson Roger Larson David Lippert James Mack Dick Moore Mark McDaniel Dave Newcomb Linda Pierce Chuck Van Dusen Duane Young *

Robert Packard substituted for Mr. Knutson for a portion of the meeting. Mark Snyder and Dario Pedigo were unable to attend.

2. European Expert: Lorenzo Dominichini, Italy (assisted by Francesca La Torre) 3. Research team (ERES/BRE): Michael Darter Harold Von Quintus Brian Killingsworth Jane Jiang Wayne Seiler Jerry Daleiden

Emmanuel Owusu-Antwi Ken McGhee

Overview NCHRP project 1-32 consensus meeting on pavement design features was held in Chicago from 1 pm on January 22 to noon on January 26, 1996. The objective of this meeting was to reach consensus on key pavement design site conditions and design features recommended for inclusion in the pavement design catalog. Participants were sent a copy of the draft design catalog prior to the meeting for review. A copy of the meeting agenda is provided in the appendix. A brief summary of workshop activities is as follows. Monday afternoon: introductions, objectives of NCHRP Project 1-32 and of this meeting, overview of the draft catalog, European experience presentation, consensus building process, and an open discussion on the catalog development. Tuesday: more discussion on catalog development, the entire group discussed and took consensus ballots on some general issues concerning both flexible and rigid pavements including site conditions, the pavement design process, and general design criteria (e.g. design reliability).

B-1

Wednesday: the group was split into a flexible pavement group and a rigid pavement group. Each group discussed, built consensus on pavement design features and took ballots on many recommendations. Thursday morning: the entire group met to summarize each group’s activities and discuss the progress of the consensus building on the previous day. Then the group discussed and took ballots on the initial and terminal serviceability to be used in the catalog. After that, the group broke into the same two groups again to continue discussions and balloting on pavement design feature recommendations. The flexible pavement group finished at noon, so some members joined the rigid pavement group in the discussion and consensus building on PCC pavement design features during the afternoon. Friday morning: the entire group again summarized Thursday’s progress. A presentation was made on the knowledge-based expert system (KBES) development plan, and the group had a discussion on the usage, development, and implementation of the KBES. After that, terminal serviceability and slab on grade option for PCC pavement design were further discussed and balloted. In conclusion, the entire group had an open discussion on the future development and other general aspects of the catalog. The consensus building process included the following steps for each design recommendation. First, a member of the research team briefly presented a specific design recommendation (i.e., subgrade treatment, design reliability). Second, the consensus group freely discussed the recommendation. Third, if it was apparent that most members agreed with the recommendation, a consensus ballot was completed by each member (see appendix for consensus ballots). Fourth, the ballot results were entered into a personal computer (Excel spreadsheet) and a frequency distribution was projected onto a screen for all to see. Fifth, if a consensus was reached (meaning no one disagreed with the recommendation), the group moved on to the next recommendation. Sixth, if one or more participants disagreed as indicated by a 40 or lower rating, their reason was read to the group (anyone who disagreed had to write the reason on the ballot), and additional discussion was held until either a consensus was achieved or it became apparent that it was impossible to achieve a consensus at the time. Nearly always, the further discussion resulted in either a modification of the recommendation, or the attachment of notes to the recommendation indicating the concerns of the consensus group. When agreement could not be reached, the recommendation was brought up at a later time and discussed and reballoted, which usually resulted in a consensus at that time. A summary of all ballots for each recommendation voted, histograms of the voting results, and all comments provided by the resource group members are given in the appendix. Overall, the consensus meeting was very successful and a consensus was reached on almost all the issues, often after various revisions were made. A virtual wealth of knowledge and experience existed in the consensus group and many good ideas were brought out in the discussions. The group generally appeared to have a positive feeling about the achievements of the consensus group meeting. Monday January 22, 1:00-5:00pm Introductions

B-2

Mike Darter welcomed everyone, and each attendee and the research team introduced themselves. He then presented the NCHRP Project 1-32 Research Problem Statement, objectives, the work tasks, and the objective of this meeting. This objective was: for the resource group to reach decisions about the pavement catalog that best reflects the thinking of all group members (i.e., reach consensus). The resource group will have an extremely important role in the final set of recommendations given in the pavement catalog. European Catalogs: Uses and Development Professor Lorenzo Domenichini of Italy made a presentation on the concept, usage and development of European catalogs. A brief summary follows. A catalog is a “hard” version of an expert system and tries to represent simplistically a very complex issue. A catalog is: • A design tool providing guidelines and rules of good practice. • Not a tailored design for all possible conditions. • A research tool promoting the achievement of better quality in pavement construction and reduction in the scatter of the results in the pavement performance studies. A pavement catalog aims to: • Provide guidelines for the standardization of pavement structures in traffic areas. • Provide a simple means to present an agencies policy on pavement management. • Introduce good practices in pavement design and construction (more than just pavement thickness; i.e. subdrainage). The development of pavement design catalogs dates back to 1932 in France, where the first catalog appears to have been proposed, however, most were developed in the 1960's and 70's. Pavement design catalogs developed in Europe are shown in table 1. Table 1. European pavement design catalog development.

• • • •

Country France

Year of first release 1971

Revisions Rev A 1977; Rev B 1988

Germany

1986

Rev A 1989; Rev B 1990

Spain

1963

Rev A 1976; Rev B 1990

Belgium

1983-1985

Swiss

1989

Italy

1993

European catalogs use number of heavy vehicles as the traffic input as opposed to equivalent single axles used in the US because they feel the transition is "unreliable" and "tedious". European catalogs include semi-rigid pavements (composites). Allow variable thickness across transverse cross-sections (especially France). Graphical representation of thickness designs and cross-sections.

B-3

• •

German catalog includes solution for bus lanes and bus stops. Belgium includes overloaded design structures as well as "normal" traffic structures.

Spanish Catalog: Introduces roller-compacted portland cement concrete over a soil cement with a bituminous wearing course to meet smoothness.

B-4

Swiss Catalog: Includes section on splitting out the wearing course from the bituminous base. Also includes recommendations for staged construction for asphalt pavements. Italian Catalog: • Developed over a 3-year period using a working group approach where 5 different groups studied specific items regarding pavement design. • Climatic regions not differentiated in this catalog (below 1000 meters above sea level). The catalog included checks by mechanistic analysis. Therefore, designed to most severe climatic conditions. • One concept that Europeans (especially the French) have moved forward is the concept of developing local catalogs for specific regions and cities within a national system. • Also some catalogs refer to standardizing rehabilitation techniques for pre-existing systems (i.e. reconstruction). • France has developed a catalog for low volume roads and residential areas. They have introduced some new ways to present pavement design in terms of the functional criteria (as opposed to structural criteria). • Initial French Catalog - 15,000 copies distributed over 3 years, very successful. • Helped clarify the difficulties in designing pavement. • Helped in promoting research and continuing pavement design process (engineers did not stop with catalog provided designs). • European catalogs assume that designs between flexible and rigid pavement are equivalent about the structural characteristics (i.e. structural design life), however are not necessarily equivalent in terms of maintenance. Questions: (to Lorenzo Domenichini) • What about equivalency between pavement types in terms of maintenance? European catalogs are based on a structural solution. They do not include functional conditions, such as wearing courses, so differences in future maintenance/rehabilitation may exist. • What about modes of failure, are they discussed? Yes in notes to cells. • Validation of catalogs? Revised every few years, but after 25 years they are still very much on the stage. • Why a catalog, as opposed to other procedures? Actual design process knowledge is held in a limited number of people; a catalog helps to disseminate information. Catalog makes people realize the difficulty in pavement design. It increases interest in pavement design and in maintenance and rehabilitation. • What about the potential loss of pavement expertise if a catalog is used by an agency? Domenichini indicated just the opposite occurred in Europe, where catalogs increased interest in pavement design. Consensus Building Ken McGhee made a presentation on the consensus building concept, guidelines, and ground rules to be used in this meeting. Consensus Building • All can support; none oppose.

B-5

• • • •

Not a majority vote. Mutual understanding of views. Reached fairly and openly. Win-win situation.

Guidelines: • Talk, but listen • Encourage others • Share information • Do not bargain or trade support • Differences = strength • Disagree without being disagreeable • Silence is not consensus • Chaos is also not consensus Most People Will Support If: 1. Opportunity to express views. 2. They think views were heard and understood. 3. Feel views were considered by entire group. Ground Rules I. Topic Recommendations by Research Team (5-10 min.) II. Open Discussion by Entire Group (15 min. max.) A. New Ideas B. Proposed Changes C. Objections (Limit Individual Discourse 5 min.) III. Seek Group Consensus (20 min.) A. Consider Ideas, Objections, Changes B. Ballot - Degree of Consensus Description and Discussion of Draft Catalog Mike Darter and Harold Von Quintus made a brief presentation on the content, basis, and development of the draft catalog. The group had a very comprehensive discussion on different aspects of the catalog and its development. Group Discussion: • Negative aspects: Cookbook; Positive aspects: Captures knowledge of outgoing experts. • Future of the draft catalog. It is desirable to reconvene consensus group with final product. • Concerned about the catalog being “written in stone.” How to change it? • Catalog should compliment but not replace current knowledge. • Catalog will help standardize practice and help inexperienced engineers. • "Policy" concerns; people may be forced to use catalog. • Need to "stupid proof" the catalog. • Usage for low volume roads may be good because of lack of engineering knowledge.

B-6

•

Catalog should be intermediate step and not final product. Actual best use could be Internetbased knowledge-based expert system. Could be a tool by which information could be shared.

B-7

January 23, Tuesday, Entire Group Continued Discussion on Catalog Concept • • • •

• • • • • • • •

Concern about whether or not this catalog is applicable to local agencies. This catalog is aimed at state highway agencies as per problem statement/guidance by NCHRP. But will local agencies use it, when it may not be applicable? Possible that local agencies will default to the lowest level which is still too high for local roads. Implementation statement (of KBES) should also include information regarding implementation of the paper catalog. Concern that the catalog could be used to optimize design. The research team answered that the catalog produced under NCHRP 1-32 is NOT recommended for use in optimizing a project level design. That is not the purpose of this research study. Actually, the ranges given for design features like layer thicknesses make this impossible. The purpose of this catalog is to provide a “knowledge base” of recommendations achieved through analytical methods and the consensus of experts that would provide guidance to highway agencies on good practice. Any agency wishing to use the catalog for project level design must tailor it to their specific materials, climate, soils, traffic and policies. Another idea would be to identify to designers the optimum type of design in terms of initial cost & LCC. Answer, this catalog will not provide life-cycle cost procedures. Customizing should be separated into two levels: Site (project) State adaptability Washington State catalog is used for determining an approximate design so that an appropriate amount of money can be set aside at the network level. After that, project level designs are conducted using normal design procedures. Also can be used to help simplify pavement management in that only a few structural sections would be tracked for performance. A section on “How To Implement” or tailor to a SHA is needed in the catalog. Will stage construction will be given in catalog? None will be provided.

Basis for Development of Design Catalog The 1993 AASHTO design guide is recommended as the basis for initial development of the structural sections in the catalog. They would then be checked using mechanistic-empirical models. Other design features would be based on the best of state practice and experience from the consensus group. Group Discussion: The question of exactly what was to be the basis of the structural designs included in the catalog was discussed at length. After much discussion, the resource group balloted and reached consensus on the following proposed design catalog developmental process (see appendix page B-1 for a summary of balloting results and comments by participants): • 1993 AASHTO design guide will be used to develop the initial structural designs. These designs would then be checked and other key design features added using the following. • FHWA pavement manuals • Available mechanistic/empirical performance models

B-8

• • • •

State highway agencies (SHAs) pavement design practices LTPP and other pavement performance databases Design guidelines from industry Performance modeling/limiting criteria

There was considerable discussion as to the equivalence of flexible and rigid pavement designs as produced by the AASHTO Guide. Nothing is provided in the AASHTO Guide on this topic. The following statement was prepared by the research team concerning this topic based on the discussion and will be placed in the catalog (after review by the Resource Group). Equivalence Of Pavement Designs What type of design equivalency does the AASHTO Design procedure provide between different pavement types (i.e., full-depth AC, AC with conventional aggregate base, JPCP, JRCP, AC with treated base, CRCP)? The AASHTO design procedure, used correctly with proper inputs, provides pavement structures that carry a specified amount of mixed traffic loadings between an initial serviceability level and a terminal serviceability level, at a specified level of design reliability. The design is based on data from one climate, thus, the procedure is not directly applicable to other climatic regions of the country, however, that is a different issue. Since the procedure is based on full-scale field testing of flexible and rigid pavements over a 2 year period only, the method does not include “aging” effects beyond two years. “Aging” is defined herein as any process that causes damage (reduction in serviceability) to a pavement other than traffic load. The effects of aging is mostly a durability issue and relates heavily to materials selection, mixture design, and the subgrade. Some examples include the following for all pavement types: frost heave, swelling soils, settlements of foundations, and disintegration of any pavement layer from freeze-thaw effects. For flexible pavements specifically: hardening of asphalt binder resulting in thermal or shrinkage related cracking, reflection cracking from treated bases, and stripping of asphalt resulting in increased potential for rutting. For rigid pavements specifically: D cracking, reactive aggregates, and incompressibles that result in joint spalling, and corrosion of steel that results in various problems. As “aging” damage occurs, traffic loading may result in a more rapid deterioration because of the existing fractures or softening or disintegration of materials and other effects (dynamic loads). None of the AASHO Road Test pavements received any maintenance or rehabilitation during the time they were considered in test. The application of maintenance or rehabilitation may, therefore, extend the design life of any pavement designed by the AASHTO procedure. Therefore, two different types of pavements (either two different flexible pavements, two different rigid pavements, or one flexible and one rigid pavement) designed for the same mixed traffic would not necessarily perform the same (same trend of serviceability) if one pavement was trafficked over a 2 year time period and the other over a 20 year time period. The pavement trafficked over a 20 year time period may develop a lot of “aging” damage to the pavement that could reduce the serviceability of the pavement, causing it to reach a terminal level long before the design traffic was applied. Maintenance and/or rehabilitation may be needed to extend the pavements life until it carries the design traffic.

B-9

Therefore, even if a flexible and a rigid pavement, or for that matter two different types of flexible or two different types of rigid pavements, could be designed to carry the same amount of mixed traffic loadings using the AASHTO Guide, they would not necessarily perform the same over the a given time period or require the same amount of maintenance or rehabilitation due to the differences in “aging” of the pavements. This catalog provides structural sections that are expected to carry a specified amount of mixed traffic that has been projected to occur over a given design period. Differing amounts of maintenance and rehabilitation may be required to reach the end of the design analysis period. Site Conditions Traffic Predicted 18-kip ESALs and the corresponding ranges in the draft catalog were proposed to be used for pavement structural design because this is the input for AASHTO, is the most widely use traffic input for design, and incorporates many traffic variables (lane distribution, axle type, axle load distribution, growth rates over time, etc.). Due to the different equivalence factors for flexible and rigid pavements, the catalog will refer to flexible pavement ESALs and rigid pavement ESALs. Group Discussion • Someone suggested that we put limiting axle weight (load) on low type designs. Recommend that we should add a note about low type roads with overloads should use different traffic class. • High end on traffic may be too low (i.e. 40-year design). Suggest 100 million for flexible and 150 million rigid. • Someone asked if high end rigid ESALs is actually correct because it was noted that most of the concrete pavements at AASHO road test did not fail. • Design Life vs. Total Life. Maintenance treatments will still be required due to durability failures. • Flexible ESALs vs. Rigid ESALs. Should we use another class that combines (i.e. load class, commercial truck traffic, ADT, AADT (present))? • Concern that two pavement types in a traffic class are equal. However, they may not be equal in terms of overall life cycle cost. Catalog is misleading in this regard. • Add statements regarding maintenance and rehabilitation to reach terminal serviceability levels. "Designs are based solely on structural issues and not durability or functional issues". "Cross sections may require different maintenance with time". Consensus was reached to use predicted cumulative 18-kip (80 kN) ESAL for pavement structural design, and a higher range of 100-150 million ESALs will be added to the catalog structural design charts. See summary of balloting results and comments in the appendix, page B-3. Subgrade The research team proposed to use a laboratory determined resilient modulus (MR) value of the subgrade for flexible pavement design and elastic k-value of the subgrade for rigid pavement design, as is utilized in the AASHTO design guide. Group Discussion

B-10

•

What about the granular subgrade material? Answer: granular is included in the “strong” category for each pavement type. • Should put * on the structural design resulting from any minimum value required. • k-value: can be determined from AASHTO/ASTM procedures plate load testing, backcalculation, or correlations with soils tests or classification. Some recommendations will be provided based on NCHRP 1-30 results. • The k value shown in the catalog represents the subgrade support value only, not the base or subbase. It is also the seasonally adjusted k value (the same is true for resilient modulus). • Is there a connection between the Mr and k-values? Approximate only. • MR can be determined from laboratory testing, backcalculation, and soil test/classification correlations. A composite MR may be needed where different layers of soil type exists when using lab or soil class or soil test results. Consensus was achieved on this recommendation, see appendix page B-2 for summary of balloting results. Climate It was proposed to use the LTPP four climatic regions as general climatic zones to differentiate different designs and design features in the catalog. After discussion, the resource group balloted and reached a consensus on this recommendation. See appendix page B-4 for the summary of balloting results and comments. General Design Criteria Design Reliability The recommended design reliability levels the increase with increasing ESALs were presented. After discussion, the resource group balloted on the reliability levels shown in the draft catalog, however, the design reliability of the lowest traffic level would be set at 75% for both pavement types. Consensus was not reached because one person disagreed with this scheme and suggested to use only 3 levels of design reliability. The group also discussed other possible modes or methods such as defining reliability as a function of ADT, functional class, or heavy truck traffic volume. This item was reballoted at a later time and a consensus was reached where the lowest traffic level would use R = 75 percent, and the other levels as recommended in the draft catalog. See appendix page B-5 for summary of balloting results and comments. Local Conditions Subgrade Treatment The draft catalog includes some text and recommendations as to the treatment of the subgrade where poor soil conditions exist, however, it is incomplete. Discussion of group • Subgrade preparation should be included so that a proper working platform can be established. • Uniformity of grade along job is very important. Must be able to provide information/ guidelines in transition areas (i.e. cut/fills). Someone mentioned that when they repair existing pavements, many times these repairs are required over a cut/fill transition. • For swelling soils and frost susceptible soils minimal discussion and guidance should be provided with numerous references to existing information on these topics. Treatment for swelling soils should be level up overlays.

B-11

• • •

Discuss how to recognize these conditions and the associated consequences. These things affect performance, but cannot deal with increasing pavement structure. Provide some guidance on recognizing subgrade problems and their consequence. There is an assumed quality subgrade in design. If not achieved then performance problem. Do not change slab thickness based on subgrade treatment.

The group reached consensus on the need to provide guidelines on improving the subgrade under certain conditions, which include subgrade treatment/modification, compaction, uniformity, frost susceptible soil, swelling soil, etc. This section of the draft catalog will be rewritten using the above comments and others provided in the summary of balloting results in the appendix, page B-6. Subsurface Drainage Cross-sections of flexible and rigid pavements containing a permeable drainage layer were included in the draft catalog. Also, a section on determining the need for subsurface drainage and other details is provided in the draft catalog. Considerable discussion occurred on this topic throughout the workshop which showed it to be highly controversial. The overall feedback was that the “jury is still out” on the costs, benefits, and risks of permeable bases for flexible and rigid pavements and that we should not include them in the catalog as a recommended section at this time. Discussion: • This is a project level decision. • Complex issue; however does the catalog provide minimum guidelines or does the catalog not include any comments at all? • Do not want to recommend subdrainage in flat areas when water has no place to go. Just want to try to keep water out. • Concern over stability of untreated permeable aggregate base courses under PCC pavements. Some cracking of slabs has resulted. Slab cracking has resulted on some projects with treated bases immediately after construction due to increased friction and inadequate sawing. One project on I80 having a lime treated subgrade just beneath the permeable layer has pumped the layer full of fines and caused punchouts in the CRCP. • Questions about drains: • When do we use drains? • How effective and what conditions are they cost effective? • A permeable base under a CRCP does not perform well. The group achieved consensus on the following guidelines of developing the drainage section (see summary of balloting results and comments in appendix, page B-7): • • •

Take the cross-sections with permeable bases out of the catalog. Place some general recommendations in the drainage section in the catalog. The jury is still out on the benefits and liabilities of a permeable base drainage system. If an agency elects to provide subdrainage, provide a fully functional drainage systems as opposed to just providing drainage. Discuss advantages and disadvantages of each approach.

B-12

January 24, Wednesday, Asphalt Session Initial Serviceability A value of 4.5 was used to develop the draft catalog and is recommended for all pavement types. Discussion. Are these values used to determine the initial designs appropriate with current construction specifications? Very smooth pavements are being achieved where smoothness specifications area being used. Should they have been a function of pavement type? Consensus was reached by the flexible pavement group on using 4.5 as the initial serviceability for both types of pavements. Note, that this topic was discussed with both groups together on Thursday and the results of those ballots are presented later. Terminal Serviceability A value of 2.5 was used to develop the draft catalog and is recommended for both pavement types (with the exception that 3.0 was used for the higher traffic level flexible pavements). Discussion. For terminal serviceability, you must determine what you are talking about. Terminal serviceability would require a major rehabilitation. • • •

Comment that we should look at total loss in serviceability as opposed to initial/terminal levels. Should not go with ∆PSI > 2.0. Presently the draft catalog includes levels where the drop is 1.5 or 2.0 Harold mentioned the smoothness competitions in relation to the comparison between initial serviceability levels. Should they be the same?

Consensus was reached by the flexible pavement group showing that all agreed with the terminal serviceability of 2.5 for both types of pavements. Note, that this topic was discussed with both groups together on Thursday and the results of those ballots are presented later. Cross Sections • Full-depth pavements Shaded Cells The flexible pavement portion of the catalog includes some cells in the design matrix as shaded to indicate that these designs might not be practical. Discussion • Washington catalog has shaded cells based upon functionality not necessarily best practice (i.e. 75% reliability on high traffic roads vs. full-depth on weak (unstabilized) subgrade). A consensus ballot provided the following agreement for each type of AC pavement (see appendix pages B-8, 9,10, 11 and 12): Show cells of good practice (#s); Show cells of questionable design (#s with shade); and Show cells of bad practice (no #s).

B-13

• • • * * * * * * * * * *

Suggestion that all options (cross-sections) for a given traffic level be on one page. Check titles - take out non-frost/frost. Take out stabilization sections and just mention that subgrade should be at least the minimum effective Mr (3,000 psi). Change in cross-sections from lime-treated subgrade to stabilized subgrade. Show stabilization of subgrade for 3000 psi and 5000 psi subgrades (not 9000 and 14000 psi). Concerns regarding subbase materials; all should be shown as "unbound granular" subbase. Someone asked for explanation of ranges provided in cells or elimination of same. An explanation will be given. The ranges in the flexible catalog were based on the range of traffic loading for the cell. Subgrade Mr was held at the mean for the cell. Weak subgrades (resilient modului of 3000 and 5000 psi) need subbase cover (at least 4 in). Eliminate subbase for CTB sections with stabilized subgrade. Define shaded areas as marginal and provide an explanation of why they are considered so. Limit subbase to 8 in max. for CTB sections without stabilized subgrade. Group agreed that range of values should only be used in one layer only (HMAC). Permeable bases should not be shown as structural sections (see consensus ballot page B-13).

Materials Characterization • No breakout of HMAC layers (i.e. wearing, binder, base) • Material characteristics of HMAC - Generalize regarding stability, moisture damage, etc. - State that "premier mix design and construction practices should be specified" • Binder selection - No guidelines on binder selection required - State that: "Proper binder selection is required for existing climate" and, "detailed discussion is beyond the scope of this report" • Structural Layer Coefficients - Unbound base (crushed stone) 0.14 - Subbase (unbound granular material) 0.10 - ATB (include cold mix) 0.25 - CTB 0.22 - HMAC 0.42 • Seasonal Adjustments - Note that none have been made because the subgrade resilient modulus is defined as the effective seasonally adjusted value. Consensus ballots were taken for materials characterization for dense graded AC asphalt stabilized base, unbound granular base/subbase, and cement treated bases (see appendix pages B-14, 15, and 16). Mechanistic Checks - Roughness - Fatigue; Limit tensile strain at bottom of HMAC or ASB or tensile stress at bottom of CTB - Deformation; Limit subgrade vertical compressive strains - Deflection; Limit surface deflection A consensus ballot was taken on mechanistic checks as shown in the appendix, page B-17.

B-14

January 24, Wednesday—Concrete Pavement Session PCC Pavement Cross Sections Recommendations for specific cross sections were provided in the draft catalog. Extensive discussion on this topic occurred as follows. • • • • • •

Must define edge support better (three levels of edge support defined).. Base layer normally goes beyond slab edge, a widened base. Show drainage right outside of shoulder. Trapezoidal slabs are normally designed for outside lane and then reduced toward the inner lane. Trapezoidal slabs: increase 1 in at outer lane and decrease 1 in at the inner lane. A crowned pavement cannot be constructed as smoothness as a uniform cross slope pavement across traffic lanes.

Based on the discussion, 6 cross sections were identified to be shown in the catalog: 1. Conventional lanes with no edge support, i.e. Gravel or AC shoulder. 2. Conventional lanes with medium edge support, e.g. tied shoulder with longitudinal joint not monilithically placed (shoulder is added after traffic lanes placed). 3. Conventional with super edge support, e.g integral tied shoulder (traffic lane and shoulder included in same placement), or widened slab, or integral curbs. 4. Trapezoidal with no edge support, i.e. AC shoulder. 5. Trapezoidal with medium edge support, e.g. tied shoulder. 6. Trapezoidal with super edge support, e.g integral tied shoulder, widened slab, integral curbs. For trapezoidal sections, discuss widening for future traffic. Show some cross-sections from states for illustration. Check thin slabs for longitudinal cracking when widened slab is utilized. Consensus was reached on the first ballot as shown in the appendix on cross sections (see many comments on ballots also, appendix page B-18). PCC Pavement Design Features PCC pavement structural design- General design inputs The following design inputs were proposed in draft catalog: • P0 = 4.5; Pt = 2.5; ( will be discussed with AC group) • Overall standard deviation: 0.39 • Mean 28-day, 3rd point flexural strength: 650 psi (method for adjustment of design for other values will be provided in catalog) • PCC slab elastic modulus: 4,000,000 psi • Cd: use 1.0 since no drainage system will be shown with the cross section • J-value: Use AASHTO recommended values and modify for the three edge support conditions. The first balloting did not reach consensus due to some disagreement on selecting J-values (page B-19). After discussions and comparing with PCA recommendations, J-values under different situations were proposed as shown in table 2. The group reached consensus on this recommendation (page B-20).

B-15

Table 2. Load transfer coefficient, J-value Pavement Type

Joint load transfer

Edge support

J-value

JPCP/JRCP

Doweled

Super

2.7

Medium

3.0

None

3.2

Super

3.7; if ESAL include various drainage options: Have trapezoidal section in subgrade. I think trapezoidal CRCP section should at least be looked at. 4. As modified during the discussions. Include discussion on when "thinning up" shoulder be considered. 5. I think lean concrete base shoulder should be considered in edge support sections. 6. Need good discussion on cross section details/ options this is a policy issue - state needs freedom to use what works in their state. 7. We covered a lot of terms. Hope you got them all. 8. As modified by Darter’s comments.

B-47

SUMMARY OF BALLOTING RESULTS Discussion Topic: Item: Issue:

Design Inputs for PCCP Structural Design General design inputs; Vote No. 1 Use the proposed input values

Distribution of Rating 8

Number of Raters

7 6

Average = 62 StDev = 21

5 4 3 2 1

80-90

70-80

90-100

Rating Scale

60-70

50-60

40-50

30-40

20-30

10-20

0-10

0

Comments: 1. Believe that CRCP *widened lane should be 2.5; Tied = 2.8; Based on what Mike says, AC = 3.1; even poorly tied shoulders significantly reduced punch out. 2. Question split for doweled vs. undoweled pavements; CRCP use of 2.6 only if steel reinforcement design improved and very good base support provided. 3. Disagree on J-factor for undoweled joints, J-value should be lowered for low traffic situations.

B-48

SUMMARY OF BALLOTING RESULTS

Discussion Topic: Design Inputs for PCCP Structural Design – Vote No. 2 Item: Issue:

General Design Inputs; Vote No. 2 Use the proposed input values (include adjustments of J Values)

Comments: 1. Make more compatible w/AASHTO Low volume Road Chapter. 2. Other model results have evaluated relative to thickness.

B-49

SUMMARY OF BALLOTING RESULTS Discussion Topic: Item: Issue:

PCCP Structural Design Minimum slab thicknesses Use no minimum slab thicknesses for JPCP/JRCP; 8 in for CRCP

Distribution of Rating 8

Number of Raters

7 6

Average = 59 StDev = 25

5 4 3 2 1

80-90

70-80

90-100

Rating Scale

60-70

50-60

40-50

30-40

20-30

10-20

0-10

0

Comments: 1. Check for tensile strain at bottom of concrete slab. 2. Models should be used to check thicknesses to ensure adequacy. FEM analysis may be required. 3. 6” min needed-especially if on grade; need to add note of cautions if allowed. 4. No min for JPCP & JRCP; min for CRCP = 8"

B-50

SUMMARY OF BALLOTING RESULTS Discussion Topic: Item: Issue:

PCCP Structural Design Base Usage; Vote No. 1 Design PCC slab on grade as an optional design

Distribution of Rating 8

Number of Raters

7 6

Average = 65 StDev = 22

5 4 3 2 1

80-90

70-80

90-100

Rating Scale

60-70

50-60

40-50

30-40

20-30

10-20

0-10

0

Comments: 1. Don't recommend deleting subbase. 2. Should require base at all traffic levels shown; need to add note of caution if allowed. 3. Add box for “on subgrade” for 0.75-1.5 & 1.5-3.0 ESAL’s.

B-51

SUMMARY OF BALLOTING RESULTS Discussion Topic: Item: Issue:

PCCP Structural Design Base usage; Vote No. 2 Design PCC slab on grade as an optional design for ESALs < 3 million

Distribution of Rating 8

Number of Raters

7 6 5

Average = 58 StDev = 27

4 3 2 1

90-100

80-90

70-80

60-70

50-60

40-50

30-40

20-30

10-20

0-10

0

Rating Scale

Comments: 1. I don't believe that this is a nationwide good practice. 2. Show concrete on subgrade as marginal for strong subgrade. 3. Eliminate weak; medium-marginal; acceptable strong; could extend strong on grade up to 6 million.

B-52

SUMMARY OF BALLOTING RESULTS Discussion Topic: Item: Issue:

PCCP Structural Design Base usage; Vote No. 3 Design PCC slab on grade as an optional design for ESALs < 3 million;

Distribution of Rating 8

Number of Raters

7 6 5

Average = 67 StDev = 12

4 3 2 1

90-100

80-90

70-80

60-70

50-60

40-50

30-40

20-30

10-20

0-10

0

Rating Scale

eliminate weak and put marginal on medium and strong soil Comments: 1. Show all cells, but weakest subgrade at 0.75-1.5 ESALs " marginal" 2. Should include mechanistic check- change thickness as needed; comments on marginal design / grading limits in some cases weak subgrade out. 3. Add caution note about potential for vol. change in subgrade to cause failure. 4. Must provide cautiously-mechanistic analysis and any other models to verify cracking and faulting. 5. Only medium support should be shaded; also no minimum thickness requirement should be included. No arbitrary artificial constraints on AC or PCC arbitrarily. 6. I would like a cautionary note on heavy “over loaded” trucks or a mechanistic check using such a truck. 7. As long as limitations have been addressed. 8. With write up of why this practice is marginal. 9. Mechanistic check on thickness- thickened edge 7-5-7. 10. If k=300 psi, regardless of concrete thickness, the “design is not marginal—it is better— thickness is not the issue. 11. Both medium & strong subgrade should be shaded as marginal – list items that have made this design successful, such as wintering B-53

SUMMARY OF BALLOTING RESULTS Discussion Topic: Item:

PCCP Structural Design Keep permeable bases as structural design alternatives

Distribution of Rating 8

Number of Raters

7 6

Average = 51 StDev = 28

5 4 3 2 1

90-100

80-90

70-80

60-70

50-60

40-50

30-40

20-30

10-20

0-10

0

Rating Scale

Comments: 1. Some early failure may be first indication of major problem w/ permeable base. Do we need 10 yr of construction & 100's of miles before questioning design? 2. As original w/ no base option. 3. Cement stabilized should be recognized in a different category than either unbond or asphalt bound bases. Structural value needs to be assigned to cement base.

B-54

SUMMARY OF BALLOTING RESULTS Discussion Topic: Item: Issue:

PCCP Structural Design Base thickness range Use 4-6 in for base thickness range

Distribution of Rating 8

Number of Raters

7 6

Average = 77 StDev = 8

5 4 3 2 1

90-100

80-90

70-80

60-70

50-60

40-50

30-40

20-30

10-20

0-10

0

Rating Scale

Comments: 1. Stabilized base 4-6 in; unstabilized 6 in. less if not used for working platform; 2. For aggregate base: if working platform, 6" minimum ; Non- may be