Click for next page ( R2


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page R1
NATIONAL NCHRP REPORT 538 COOPERATIVE HIGHWAY RESEARCH PROGRAM Traffic Data Collection, Analysis, and Forecasting for Mechanistic Pavement Design

OCR for page R1
TRANSPORTATION RESEARCH BOARD EXECUTIVE COMMITTEE 2005 (Membership as of February 2005) OFFICERS Chair: Joseph H. Boardman, Commissioner, New York State DOT Vice Chair: Michael D. Meyer, Professor, School of Civil and Environmental Engineering, Georgia Institute of Technology Executive Director: Robert E. Skinner, Jr., Transportation Research Board MEMBERS MICHAEL W. BEHRENS, Executive Director, Texas DOT LARRY L. BROWN, SR., Executive Director, Mississippi DOT DEBORAH H. BUTLER, Vice President, Customer Service, Norfolk Southern Corporation and Subsidiaries, Atlanta, GA ANNE P. CANBY, President, Surface Transportation Policy Project, Washington, DC JOHN L. CRAIG, Director, Nebraska Department of Roads DOUGLAS G. DUNCAN, President and CEO, FedEx Freight, Memphis, TN NICHOLAS J. GARBER, Professor of Civil Engineering, University of Virginia, Charlottesville ANGELA GITTENS, Consultant, Miami, FL GENEVIEVE GIULIANO, Director, Metrans Transportation Center, and Professor, School of Policy, Planning, and Development, USC, Los Angeles BERNARD S. GROSECLOSE, JR., President and CEO, South Carolina State Ports Authority SUSAN HANSON, Landry University Professor of Geography, Graduate School of Geography, Clark University JAMES R. HERTWIG, President, CSX Intermodal, Jacksonville, FL GLORIA J. JEFF, Director, Michigan DOT ADIB K. KANAFANI, Cahill Professor of Civil Engineering, University of California, Berkeley HERBERT S. LEVINSON, Principal, Herbert S. Levinson Transportation Consultant, New Haven, CT SUE MCNEIL, Director and Professor, Urban Transportation Center, University of Illinois, Chicago MICHAEL MORRIS, Director of Transportation, North Central Texas Council of Governments CAROL A. MURRAY, Commissioner, New Hampshire DOT JOHN E. NJORD, Executive Director, Utah DOT PHILIP A. SHUCET, Commissioner, Virginia DOT MICHAEL S. TOWNES, President and CEO, Hampton Roads Transit, Hampton, VA C. MICHAEL WALTON, Ernest H. Cockrell Centennial Chair in Engineering, University of Texas, Austin LINDA S. WATSON, Executive Director, LYNX--Central Florida Regional Transportation Authority MARION C. BLAKEY, Federal Aviation Administrator, U.S.DOT (ex officio) SAMUEL G. BONASSO, Acting Administrator, Research and Special Programs Administration, U.S.DOT (ex officio) REBECCA M. BREWSTER, President and COO, American Transportation Research Institute, Smyrna, GA (ex officio) GEORGE BUGLIARELLO, Chancellor, Polytechnic University, and Foreign Secretary, National Academy of Engineering (ex officio) THOMAS H. COLLINS (Adm., U.S. Coast Guard), Commandant, U.S. Coast Guard (ex officio) JENNIFER L. DORN, Federal Transit Administrator, U.S.DOT (ex officio) JAMES J. EBERHARDT, Chief Scientist, Office of FreedomCAR and Vehicle Technologies, U.S. Department of Energy (ex officio) EDWARD R. HAMBERGER, President and CEO, Association of American Railroads (ex officio) JOHN C. HORSLEY, Executive Director, American Association of State Highway and Transportation Officials (ex officio) ROBERT D. JAMISON, Acting Administrator, Federal Railroad Administration, U.S.DOT (ex officio) EDWARD JOHNSON, Director, Applied Science Directorate, National Aeronautics and Space Administration (ex officio) RICK KOWALEWSKI, Deputy Director, Bureau of Transportation Statistics, U.S.DOT (ex officio) WILLIAM W. MILLAR, President, American Public Transportation Association (ex officio) MARY E. PETERS, Federal Highway Administrator, U.S.DOT (ex officio) SUZANNE RUDZINSKI, Director, Transportation and Regional Programs, U.S. Environmental Protection Agency (ex officio) JEFFREY W. RUNGE, National Highway Traffic Safety Administrator, U.S.DOT (ex officio) ANNETTE M. SANDBERG, Federal Motor Carrier Safety Administrator, U.S.DOT (ex officio) WILLIAM G. SCHUBERT, Maritime Administrator, U.S.DOT (ex officio) JEFFREY N. SHANE, Under Secretary for Policy, U.S.DOT (ex officio) CARL A. STROCK (Maj. Gen., U.S. Army), Chief of Engineers and Commanding General, U.S. Army Corps of Engineers (ex officio) NATIONAL COOPERATIVE HIGHWAY RESEARCH PROGRAM Transportation Research Board Executive Committee Subcommittee for NCHRP JOSEPH H. BOARDMAN, New York State DOT (Chair) MARY E. PETERS, Federal Highway Administration JOHN C. HORSLEY, American Association of State Highway ROBERT E. SKINNER, JR., Transportation Research Board and Transportation Officials MICHAEL S. TOWNES, Hampton Roads Transit, Hampton, VA MICHAEL D. MEYER, Georgia Institute of Technology C. MICHAEL WALTON, University of Texas, Austin

OCR for page R1
NATIONAL COOPERATIVE HIGHWAY RESEARCH PROGRAM NCHRP REPORT 538 Traffic Data Collection, Analysis, and Forecasting for Mechanistic Pavement Design CAMBRIDGE SYSTEMATICS, INC. Chevy Chase, MD WASHINGTON STATE TRANSPORTATION CENTER Seattle, WA CHAPARRAL SYSTEMS CORPORATION Santa Fe, NM S UBJECT A REAS Planning and Administration Pavement Design, Management, and Performance Research Sponsored by the American Association of State Highway and Transportation Officials in Cooperation with the Federal Highway Administration TRANSPORTATION RESEARCH BOARD WASHINGTON, D.C. 2005 www.TRB.org

OCR for page R1
NATIONAL COOPERATIVE HIGHWAY RESEARCH NCHRP REPORT 538 PROGRAM Systematic, well-designed research provides the most effective Project 1-39 approach to the solution of many problems facing highway administrators and engineers. Often, highway problems are of local ISSN 0077-5614 interest and can best be studied by highway departments ISBN 0-309-08823-2 individually or in cooperation with their state universities and Library of Congress Control Number 2005922233 others. However, the accelerating growth of highway transportation develops increasingly complex problems of wide interest to 2005 Transportation Research Board highway authorities. These problems are best studied through a coordinated program of cooperative research. Price $22.00 In recognition of these needs, the highway administrators of the American Association of State Highway and Transportation Officials initiated in 1962 an objective national highway research program employing modern scientific techniques. This program is supported on a continuing basis by funds from participating member states of the Association and it receives the full cooperation and support of the Federal Highway Administration, United States NOTICE Department of Transportation. The project that is the subject of this report was a part of the National Cooperative The Transportation Research Board of the National Academies Highway Research Program conducted by the Transportation Research Board with the was requested by the Association to administer the research approval of the Governing Board of the National Research Council. Such approval program because of the Board's recognized objectivity and reflects the Governing Board's judgment that the program concerned is of national understanding of modern research practices. The Board is uniquely importance and appropriate with respect to both the purposes and resources of the suited for this purpose as it maintains an extensive committee National Research Council. structure from which authorities on any highway transportation The members of the technical committee selected to monitor this project and to review subject may be drawn; it possesses avenues of communications and this report were chosen for recognized scholarly competence and with due cooperation with federal, state and local governmental agencies, consideration for the balance of disciplines appropriate to the project. The opinions and universities, and industry; its relationship to the National Research conclusions expressed or implied are those of the research agency that performed the Council is an insurance of objectivity; it maintains a full-time research, and, while they have been accepted as appropriate by the technical committee, research correlation staff of specialists in highway transportation they are not necessarily those of the Transportation Research Board, the National matters to bring the findings of research directly to those who are in Research Council, the American Association of State Highway and Transportation a position to use them. Officials, or the Federal Highway Administration, U.S. Department of Transportation. The program is developed on the basis of research needs Each report is reviewed and accepted for publication by the technical committee identified by chief administrators of the highway and transportation according to procedures established and monitored by the Transportation Research departments and by committees of AASHTO. Each year, specific Board Executive Committee and the Governing Board of the National Research areas of research needs to be included in the program are proposed Council. to the National Research Council and the Board by the American Association of State Highway and Transportation Officials. Research projects to fulfill these needs are defined by the Board, and qualified research agencies are selected from those that have submitted proposals. Administration and surveillance of research contracts are the responsibilities of the National Research Council and the Transportation Research Board. Published reports of the The needs for highway research are many, and the National NATIONAL COOPERATIVE HIGHWAY RESEARCH PROGRAM Cooperative Highway Research Program can make significant contributions to the solution of highway transportation problems of are available from: mutual concern to many responsible groups. The program, however, is intended to complement rather than to substitute for or Transportation Research Board duplicate other highway research programs. Business Office 500 Fifth Street, NW Washington, DC 20001 and can be ordered through the Internet at: Note: The Transportation Research Board of the National Academies, the National Research Council, the Federal Highway Administration, the American Association of State Highway and Transportation Officials, and the individual http://www.national-academies.org/trb/bookstore states participating in the National Cooperative Highway Research Program do not endorse products or manufacturers. Trade or manufacturers' names appear herein solely because they are considered essential to the object of this report. Printed in the United States of America

OCR for page R1
The National Academy of Sciences is a private, nonprofit, self-perpetuating society of distinguished schol- ars engaged in scientific and engineering research, dedicated to the furtherance of science and technology and to their use for the general welfare. On the authority of the charter granted to it by the Congress in 1863, the Academy has a mandate that requires it to advise the federal government on scientific and techni- cal matters. Dr. Bruce M. Alberts is president of the National Academy of Sciences. The National Academy of Engineering was established in 1964, under the charter of the National Acad- emy of Sciences, as a parallel organization of outstanding engineers. It is autonomous in its administration and in the selection of its members, sharing with the National Academy of Sciences the responsibility for advising the federal government. The National Academy of Engineering also sponsors engineering programs aimed at meeting national needs, encourages education and research, and recognizes the superior achieve- ments of engineers. Dr. William A. Wulf is president of the National Academy of Engineering. The Institute of Medicine was established in 1970 by the National Academy of Sciences to secure the services of eminent members of appropriate professions in the examination of policy matters pertaining to the health of the public. The Institute acts under the responsibility given to the National Academy of Sciences by its congressional charter to be an adviser to the federal government and, on its own initiative, to identify issues of medical care, research, and education. Dr. Harvey V. Fineberg is president of the Institute of Medicine. The National Research Council was organized by the National Academy of Sciences in 1916 to associate the broad community of science and technology with the Academy's purposes of furthering knowledge and advising the federal government. Functioning in accordance with general policies determined by the Acad- emy, the Council has become the principal operating agency of both the National Academy of Sciences and the National Academy of Engineering in providing services to the government, the public, and the scientific and engineering communities. The Council is administered jointly by both the Academies and the Institute of Medicine. Dr. Bruce M. Alberts and Dr. William A. Wulf are chair and vice chair, respectively, of the National Research Council. The Transportation Research Board is a division of the National Research Council, which serves the National Academy of Sciences and the National Academy of Engineering. The Board's mission is to promote innovation and progress in transportation through research. In an objective and interdisciplinary setting, the Board facilitates the sharing of information on transportation practice and policy by researchers and practitioners; stimulates research and offers research management services that promote technical excellence; provides expert advice on transportation policy and programs; and disseminates research results broadly and encourages their implementation. The Board's varied activities annually engage more than 5,000 engineers, scientists, and other transportation researchers and practitioners from the public and private sectors and academia, all of whom contribute their expertise in the public interest. The program is supported by state transportation departments, federal agencies including the component administrations of the U.S. Department of Transportation, and other organizations and individuals interested in the development of transportation. www.TRB.org www.national-academies.org

OCR for page R1
COOPERATIVE RESEARCH PROGRAMS STAFF FOR NCHRP REPORT 538 ROBERT J. REILLY, Director, Cooperative Research Programs CRAWFORD F. JENCKS, Manager, NCHRP AMIR N. HANNA, Senior Program Officer EILEEN P. DELANEY, Director of Publications BETH HATCH, Assistant Editor NEALE BAXTER, Contract Editor NCHRP PROJECT 1-39 PANEL Field of Design--Area of Pavements DANNY A. DAWOOD, Pennsylvania DOT (Chair) KENNETH W. FULTS, University of Texas at Austin (formerly Texas DOT) CHARLES K. CEROCKE, Nevada DOT HARSHAD DESAI, FHWA (formerly Florida DOT) RALPH A. GILLMANN, FHWA JERRY LEGG, Elkview, WV (formerly West Virginia DOT) TED SCOTT, Roadway Express, Inc., Alexandria, VA ANDREW WILLIAMS JR., Ohio DOT LARRY WISER, FHWA Liaison Representative STEPHEN F. MAHER, TRB Liaison Representative A. ROBERT RAAB, TRB Liaison Representative AUTHOR ACKNOWLEDGMENTS The "TrafLoad" software developed as part of NCHRP Project rison of CS with the assistance of Mr. Anant Pradhan of CS. Spec- 1-39 was implemented by Mr. Nate Clark, Dr. Dmitry Gurenich, and ifications for the software (Part 4 of this report) were prepared by Mr. Ronald Powell of Cambridge Systematics, Inc. (CS) and Dr. the Principal Investigator, Dr. Herbert Weinblatt of CS, with the Joe Wilkinson of Chaparral Systems, with the assistance of Ms. assistance of Dr. Harry Cohen. Parts 1 and 2 of the report were pre- Cindy Cornell-Martinez of Chaparral. The user's manual for the pared primarily by Dr. Weinblatt and Mr. Mark Hallenbeck of the software (Part 3 of this report) was prepared by Ms. Frances Har- Washington State Transportation Center (TRAC).

OCR for page R1
This report includes guidelines for collecting traffic data to be used in pavement FOREWORD design and software for analyzing traffic data and producing traffic data inputs required By Amir N. Hanna for mechanistic pavement analysis and design. The software--designated TrafLoad--is Senior Program Officer available to users online (http://trb.org/news/blurb_detail.asp?id=4403). The report also Transportation Research describes the actions required at both the state and national level to promote successful Board implementation of the software. The report is a useful resource for state personnel and oth- ers involved in planning and designing highway pavements. Traffic information is one of the key data elements required for the design and analy- sis of pavement structures. In the procedure used in the 1993 AASHTO Guide for Design of Pavement Structures, a mixed traffic stream of different axle loads and axle configura- tions is converted into a design traffic number by converting each expected axle load into an equivalent number of 18-kip single-axle loads, known as equivalent single-axle loads (ESALs). Equivalency factors are used to determine the number of ESALs for each axle load and axle configuration. These factors are based on the present serviceability index concept and depend on the pavement type and structure. Studies have shown that these factors also are influenced by pavement condition, distress type, failure mode, and other parameters. A more direct and rational approach to the analysis and design of pavement struc- tures involves procedures that use mechanistic-empirical principles to estimate the effects of actual traffic on pavement response and distress. This approach has been used to develop a guide for the mechanistic-empirical design of new and rehabilitated pave- ment structures as part of NCHRP Project 1-37A (currently available on-line at http://www.trb.org/mepdg/). Because of the constraints on resources available in state and local highway agencies for traffic data collection, the guide allows for various lev- els of traffic data collection and analysis. The mechanistic-based distress prediction models used in this guide require specific data for each axle type and axle-load group. Because these traffic data inputs differ from those currently used in pavement design and analysis, there was an apparent need for research to provide clear information on traffic data and forecasting and to provide guidance on selection and operation of the equipment needed for collecting these data. NCHRP Project 1-39 was conducted to address this need. Under NCHRP Project 1-39, "Traffic Data Collection, Analysis, and Forecasting for Mechanistic Pavement Design," Cambridge Systematics, Inc., was assigned the objectives of (1) developing guidelines for collecting and forecasting traffic data to formulate load spectra for use in procedures proposed in the guide for mechanistic- empirical design and (2) providing guidance on selecting, installing, and operating traffic data-collection equipment and handling traffic data. This report is concerned with the first objective; the latter objective was addressed in detail in an earlier agency report--published as NCHRP Report 509: Equipment for Traffic Load Data.

OCR for page R1
To accomplish the first objective, the researchers (1) prepared guidelines for col- lecting traffic data to be used in pavement design and (2) developed software-- designated TrafLoad--for analyzing traffic data and producing the traffic data inputs required for the mechanic-empirical design. The researchers also developed the following information: 1. Results from analysis of the effect of the length of the data-collection period on the accuracy of pavement damage factors developed from short-duration weigh- in-motion data collection. 2. A discussion of three technical issues relating to the design of software for ana- lyzing traffic: traffic ratios versus traffic factors, partial-day classification counts and truck traffic distribution factors, and simple averaging versus weighted averaging of traffic data. 3. A procedures manual documenting the algorithms used in the software. 4. A software user manual. 5. Recommendations for software improvements that could be made at a later time. 6. A discussion of the actions required, at both the state and national level, to pro- mote a successful implementation of the TrafLoad software. In addition, the researchers discussed procedures for forecasting traffic volumes and a procedure for estimating coefficients of variation for estimates of average annual daily traffic by vehicle class. TrafLoad--the software developed in this project--and related user and procedures manuals are available online (http://trb.org/news/blurb_ detail.asp?id=4403). The information contained in this report should be of interest to those involved in the planning and design of highway pavements. It will be particularly useful to agen- cies contemplating collection of traffic data for use in conjunction with the guide for the mechanistic-empirical design of new and rehabilitated pavement structures.

OCR for page R1
Contents PART 1 TRAFFIC DATA COLLECTION, ANALYSIS, AND FORECASTING FOR MECHANISTIC PAVEMENT DESIGN 1.0 Introduction ......................................................................................................................... 1-1 2.0 Implementation Needs ...................................................................................................... 1-3 2.1 National-Level Implementation Actions ................................................................. 1-3 2.2 State-Level Implementation Actions ........................................................................ 1-4 3.0 The Effect of Length of Collection Period for WIM Data ........................................... 1-7 3.1 Methodology ................................................................................................................ 1-7 3.2 Results ........................................................................................................................... 1-9 4.0 Three Technical Issues....................................................................................................... 1-12 4.1 Traffic Ratios versus Traffic Factors ......................................................................... 1-12 4.2 Partial-Day Classification Counts and Truck Traffic Distribution Factors......... 1-15 4.3 Simple versus Weighted Averages ........................................................................... 1-17 5.0 Areas for Future Work........................................................................................................ 1-20 5.1 Areas for Future Research.......................................................................................... 1-20 5.2 Potential Improvements to TrafLoad ....................................................................... 1-23 Glossary ......................................................................................................................................... 1-28 Levels of Classification Site ....................................................................................................... 1-29 Levels of WIM Site ...................................................................................................................... 1-30 Appendix A: Forecasting ............................................................................................................ 1-31 Appendix B: Coefficients of Variation .................................................................................... 1-51 PART 2 GUIDELINES FOR COLLECTING TRAFFIC DATA TO BE USED IN PAVEMENT DESIGN 1.0 Introduction ......................................................................................................................... 2-1 1.1 Traffic Data Requirements for the Pavement Design Guide Software................ 2-1 1.2 The Remainder of Part 2............................................................................................. 2-2 2.0 Weight Data.......................................................................................................................... 2-4 2.1 Sources of Estimation Error ....................................................................................... 2-6 2.2 Alternative Data-Collection Programs..................................................................... 2-8

OCR for page R1
2.3 Level 1 WIM Sites........................................................................................................ 2-12 2.4 Level 2 WIM Sites and TWRGs ................................................................................. 2-17 2.5 Level 3 WIM Sites........................................................................................................ 2-20 2.6 Weight Data Collection............................................................................................... 2-21 3.0 Vehicle Classification Data ............................................................................................... 2-23 3.1 Levels of Classification Site........................................................................................ 2-23 3.2 Data Produced for the Pavement Design Guide Software.................................... 2-24 3.3 Level 1 Classification Sites ......................................................................................... 2-25 3.4 Level 2 Classification Sites ......................................................................................... 2-34 3.5 Level 3 Classification Sites ......................................................................................... 2-37 3.6 Forecasts ....................................................................................................................... 2-39 4.0 Data Handling ..................................................................................................................... 2-45 4.1 Data Collection ............................................................................................................ 2-45 4.2 Data Analysis ............................................................................................................... 2-47 4.3 Administrative and Institutional Changes .............................................................. 2-52 Glossary ......................................................................................................................................... 2-54 Levels of Classification Site ....................................................................................................... 2-55 Levels of WIM Site ...................................................................................................................... 2-56

OCR for page R1
Part 1 Traffic Data Collection, Analysis, and Forecasting for Mechanistic Pavement Design

OCR for page R1
A.3 Changes in Vehicle Use The previous sections of this appendix present procedures for forecasting changes in the num- bers of SUTs and CTs operating on a given road. Such changes can be expected to have cor- responding effects on the stresses on a pavement over its lifetime--increases in the daily num- ber of trucks operating on the road generally result in increases in the daily stresses on the pavement. However, these changes are not the only ones that may affect these stresses. Daily (or annual) stresses on pavement may also be affected by changes in the following: 1. Commodity mix. A shift in the mix of commodities carried on the road toward (or away from) dense commodities will tend to increase (or decrease) pavement stresses. 2. Payload density. For an appreciable number of manufactured products, increasing use of protective packaging has caused shipment density to decline. If this trend continues, it will tend to reduce pavement stresses produced by trucks carrying these products. 3. Size and weight limits. Changes in truck size and weight limits can have significant effects, in either direction, on truck configurations used and the load spectra of these configurations. Each of these three types of change in vehicle use will affect the load spectra of affected vehi- cle classes, and changes in size and weight limits also can have significant effects on the num- ber of trucks in affected vehicle classes. Because the current version of the Pavement Design Guide software is not designed to use forecast changes in load spectra, TrafLoad has no capa- bility for developing such forecasts. However, the effects of these potential changes in vehicle use do warrant some further discussion. Of the three types of change, the last is potentially the most significant and also is the least pre- dictable. Possible changes in size and weight limits include the following: Increases in size limits for existing configurations (with no change in weight limits). For cube-limited shipments, such increases would allow increasing shipment sizes, resulting in heavier axle loads and increasing pavement stresses. For vehicles whose gross vehicle weight (GVW) is limited by the bridge formula, increased length limits may also allow increasing GVWs for weight-limited shipments, also resulting in heavier axle loads and increasing pavement stresses. Changes in axle-weight limits. Increases in axle-weight limits would increase pavement stresses; decreases would decrease these stresses. Elimination of the 80,000-pound cap on gross vehicle weight (GVW) that currently exists on most roads in the Interstate system. The effects of such a change would depend on the limits that would control GVW in the absence of the 80,000-pound cap. However, most proposals for eliminating this cap (and the limits that currently apply on most roads that do not have this cap) would result in converting some traffic from five-axle semi-trailer combinations to longer combination vehicles (LCVs) with seven to nine axles. On a per- vehicle basis, this change would result in increasing both payloads and pavement stresses, with the increase in payloads generally being greater than the increase in pavement stresses 1-49

OCR for page R1
per vehicle. The result generally would be a modest shift in VMT from Class 9 to Class 13 vehicles and a small reduction in total VMT of CTs that would more than balance the increase in pavement stresses per vehicle. We observe that the Pavement Design Guide software will be capable of analyzing some of the effects of changes in size and weight limits. For example, forecasts of AADT for Class 9 and Class 13 trucks could be adjusted to reflect the effects of a shift in usage that is expected to result from possible lifting of the 80,000-pound cap. However, such an analysis would be incomplete. The change in weight limits would have a significant effect on the load spectra of Class 13 trucks (and, most likely, a small effect on the load spectra of Class 9 trucks). Because the results of such a partial analysis could be misleading, it should not be used. More gener- ally, for the purpose of pavement design, forecasts of AADT by vehicle class should not be adjusted for expected shifts between vehicle classes. Partly for this reason, TrafLoad has not been designed to allow users to specify separate growth rates for each vehicle class (though it is designed to allow separate rates for SUTs and combinations). A more significant issue relating to size and weight limits stems from the difficulty that exists in forecasting likely changes in these limits and in determining whether such future changes are likely to increase or decrease total pavement stresses. Except in the special case of changes in size and weight limits that have already been enacted but that were not in effect during the most recent year for which historical data are available, it is not possible to predict with confi- dence whether such changes will increase or decrease pavement stresses. Accordingly, the inability of the current version of the Pavement Design Guide software to use forecast changes in load spectra is of little consequence, and addressing this limitation should be a low priority. 1-50

OCR for page R1
Appendix B: Coefficients of Variation The coefficient of variation (CV) is a statistic that is designed to measure the likely error in an estimate in percentage terms. The CV of a variable is obtained by dividing its standard devi- ation (s) by the estimate. As originally designed,1 the Pavement Design Guide software would have required estimated CVs for each value of annual average daily traffic (AADT) by vehicle class. Accordingly, a pre- liminary set of procedures for estimating the required CVs was developed; it was subse- quently redesigned so that no use was made of CVs. Accordingly, there is no current need to estimate CVs, and TrafLoad does not produce CVs. However, there is some possibility that a future version of the Pavement Design Guide software will require CVs. For this reason, a slightly edited version of the preliminary procedures for estimating the CVs is presented in this appendix. In the case of the AADT, there are several sources of error. One source, the use of factor groups in the factoring of Level 2 classification counts, generates enough data to permit estimation of the standard deviation and the associated CV from this source. However, other sources (dis- cussed below) do not provide such information. In order to avoid underestimating the over- all error in the AADT estimates, the proposed procedure combines a statistical estimate of the CV resulting from the use of factor groups (if used) with subjective estimates of the additional error due to other sources. The application of this procedure to Level 2 sites is described, in some detail, in the first section of this appendix. The estimation of CVs for Level 1 and Level 3 sites is discussed in the second and third sections. And a concluding section discusses an issue relating to how the CVs should be applied by the pavement-design software. B.1 Level 2 Sites Consider a Level 2 classification site for which AADT estimates are developed for each vehi- cle class (VC) using combined seasonal/day-of-week (DOW) factoring, and assume that the factor group to which it is assigned consists of n Level 1A sites. If factors are developed sep- arately for each of the Level 1A sites, they can be applied to a classification count from the Level 2 site to produce n sets of AADT estimates. These estimates, in turn, can be averaged to produce mean values of AADT for each VC. By construction, these mean values of AADT are identical to the AADT estimates produced by a combined seasonal/DOW factoring procedure. 1 ERES Consultants and FUGRO-BRE, Draft Report, prepared for NCHRP Project 1-37A, 2000, pp. 414415. 1-51

OCR for page R1
For each VC, differences between the original n estimates of AADT and the overall mean can be used to infer information about the major source of error in the overall mean, the "within- group" variance among the seasonal and DOW patterns occurring at different sites in the fac- tor group. The first subsection below presents a procedure for estimating the CV due to the major source of error. TrafLoad can be readily modified to perform all computations required by this pro- cedure. The second and third subsections present more qualitative discussions of other poten- tial sources of error. And the fourth subsection proposes a procedure that could be incorpo- rated into a future version of TrafLoad that would estimate overall CVs for AADT for all Level 2 sites. Within-Group Variance Consider a combined seasonal/DOW factor group2 consisting of n Level 1A sites, and con- sider one or more short-duration classification counts obtained at a Level 2 site for a particu- lar vehicle class. Let xi = the estimate of AADT for this vehicle class produced by applying factors obtained from site i (i = 1, ..., n) to the count(s) in question and = the mean of the xi (= the estimate produced by applying the seasonal/DOW factors x obtained from the entire group of Level 1A sites). can be used to compute the standard deviation, Then the differences between the xi and x : s, of x ( xi - x ) 2 s= i (B.1) n-1 The CV resulting from the use of factors that are derived using data from an entire factor group can be estimated as follows: ( x ij - x ) n 2 CV = i =1 (B.2) (n - 1)x 2 Similarly, assume that separate seasonal and DOW factors are being used, with the seasonal factors derived using data from n Level 1A sites and the DOW factors derived using data from m Level 1A sites (which may or may not overlap the first n sites). Then the above derivation yields a similar formula for estimating the CV due to the use of seasonal and DOW factors from these two groups: 2 The current version of TrafLoad uses separate seasonal and DOW factor groups. The development of CVs for these factor groups is presented at the end of this subsection. (See Equation B.3.) 1-52

OCR for page R1
( x ij - x ) n m 2 i = 1 j= 1 CV = (B.3) (nm - 1)x 2 It may be noted that the CVs produced by Equations B.2 and B.3 depend, in part, on the days and months on which the short-duration classification counts are obtained. For example, it is likely that, for many factor groups, there will be more variance in the Friday factors than in the Wednesday factors. This difference in variances will result in higher CVs for counts obtained on Fridays than for those obtained on Wednesdays. Other Sources of Error There are several sources of error in the AADT estimates developed for Level 2 sites in addition to those due to within-group variance. The additional sources of error include the following: 1. Equipment errors at the site in question or at the Level 1A sites used as a source of factors. 2. Random variation in the seasonal and DOW patterns of truck volumes at a Level 1A site that is atypical of other sites belonging to the same factor group. 3. The use of a single set of factors developed using data for several vehicle classes (e.g., Classes 813) to produce separate AADT estimates for each of the vehicle classes. 4. Some other minor approximations used in the factoring process. (For example, in an area in which the harvest season ends in early November, counts collected early in the month are likely to be higher than counts collected later in the month. Accordingly, use of a November/Tuesday factor is likely to produce an upward bias in AADT estimates obtained from counts collected during the first Tuesday in November.) 5. Deficiencies in the process of associating Level 2 sites with factor groups (discussed below). Of the above sources of error, the least significant is the third. This limitation in the factoring procedure does not affect the estimates of AADT of combination trucks as a group, but only the distribution of AADT among the individual classes of combinations. Thus, the errors (as in Classes 813) tend to cancel each other, though they may influence pavement design. There are two types of site at which the second source of error, random variation, may be an issue. One type consists of sites with relatively low truck volumes. Random variation in these volumes can produce patterns that are atypical of other sites belonging to the same factor group. For this reason, factoring generally is performed using data that come only from sites with relatively high truck volumes. Random variation can also be a problem for Level 1A sites at which a significant portion of total truck traffic is influenced by a small number of decision-makers. For example, if a sig- nificant portion of truck traffic at a particular site serves a nearby construction site, the sea- sonal volume pattern at the site may be very different from the patterns at other sites in the 1-53

OCR for page R1
state and also from the pattern at the site in the following year (when the construction project may be completed). The last source of error requires additional discussion. There are two issues: the degree of ambiguity in the assignment of Level 2 sites to factor groups and the representativeness of Level 1A sites in the factor group of the Level 2 site to be factored. The representativeness issue arises for Level 2 sites in the lower functional systems. As an example, assume there is a single "Rural Other" (RO) factor group that is used for all rural non-Interstate sites. Most or all of the Level 1A sites in this factor group are likely to be princi- pal arterials. Truck traffic at these sites is likely to include some traffic that both originates and terminates at relatively distant locations, as well as traffic that originates or terminates nearby-- two types of traffic that can have somewhat different seasonal and DOW patterns. Almost all truck traffic at sites in the lower functional systems, on the other hand, are likely to be locally generated. Accordingly, if the two types of truck traffic actually do have different seasonal or DOW patterns, the RO factors (obtained from principal arterials) will tend to produce some unknown bias in the AADT estimates produced for Level 2 sites on the lower systems.3 The issue of ambiguity in the assignment of Level 2 sites to factor groups is one that only arises for certain types of factor groups. If a factor group is defined to apply to all sites in one or more functional systems, there is no ambiguity in the process. On the other hand, it may be desir- able to use other information in developing the factor groups. Factor groups developed in this way may have lower variances than ones that are based entirely on a functional system, and so Equations B.2 and B.3 will produce lower CVs. However, some additional error may be cre- ated in the AADT estimates for individual Level 2 sites because of ambiguity in the definitions of the factor groups. As an example, consider a state with truck activity on the RO that has distinctly different sea- sonal patterns in the eastern and western parts of the state, with a blend of the two patterns in the middle. If the RO group is divided into an eastern RO group and a western RO group, some decision will be required as to where to place the boundary between the two groups. Because seasonal patterns at all Level 1A sites are known, it generally will be easy to make sure that all Level 1A sites are placed in the appropriate group. However, seasonal patterns at Level 2 sites generally are unknown, so it is likely that some of these sites will not be placed in the appropriate group. The resulting misassignment will produce a (generally small) addi- tional error in the AADT estimates that is not captured by Equations B.2 and B.3. Level 2B Sites Level 2B sites are Level 2 sites at which only a partial-day manual classification count is avail- able. Procedures for using a partial-day classification count to estimate full-day truck volumes 3 The alternative of creating a separate factor group for the lower functional systems may be considered. However, this alternative would increase data collection costs and, because of the random variation in truck volumes at sites with relatively low truck volumes, could produce even poorer estimates of AADT. 1-54

OCR for page R1
(by class) for the day of the count are presented in Part 2, Section 3.4. In addition to the errors discussed in the preceding subsection, AADT estimates developed for Level 2B sites incorpo- rate errors that result from the conversion of partial-day counts to estimates of full-day truck volumes. There are two types of site: 1. Sites that are dominated by business-day trucking and for which truck traffic distribution factors (TTDFs) are derived directly from the partial-day counts and 2. Other sites. As discussed in Part 2, the estimates of full-day truck volumes that were developed for the first type of site will almost always be conservative; for these sites, the probability of under- estimating truck volumes for the day in question is negligible. Thus, for these sites, the prob- ability that errors in the partial-day/full-day conversion process will contribute to underesti- mating AADT will be negligible. However, the Pavement Design Guide software requires error estimates for the AADT primarily as an indicator of the extent to which the AADT may have been underestimated. Because the errors in this conversion process are almost certain not to contribute to underestimates of AADT, they need not be reflected in CVs that are produced by the Pavement Design Guide software. The same cannot be said for the other type of site. For these sites, the conversion process may result in either underestimating or overestimating full-day truck volumes on the day in ques- tion, and the extent of any underestimation could be fairly significant. Hence, for these sites, this source of error cannot be ignored. CVs for Level 2 Sites The preceding discussion identified several reasons why the AADT estimates are likely to incorporate some errors that are not reflected in the CVs produced by Equations B.2 and B.3; the actual CVs generally will be slightly larger than those indicated by Equations B.2 and B.3. Accordingly, it is recommended that CVs for Level 2 sites incorporate small upward adjust- ments to the value produced by Equation B.2 (for combined seasonal/DOW factoring) or B.3 (for separate seasonal and DOW factoring). The suggested adjustments are 0.01 for all sites, An additional 0.03 for sites with ambiguous assignments to factor groups, An additional 0.02 for sites in functional systems that are unrepresented or significantly underrepresented among the Level 1A sites from which the seasonal and DOW factors are developed, and An additional 0.05 for Level 2B sites that states believe fit into some state-defined com- posite of the business-day and through-truck TOD patterns. These adjustments are all added to CVs produced by Equation B.2 or Equation B.3. 1-55

OCR for page R1
B.2 Level 1 Sites Level 1A Sites Level 1A sites are sites at which base-year values of AADT are obtained from classification counts collected at the site during all of, or most of, a 12-month period. The only errors in these values are those resulting from equipment malfunction and from the approximation proce- dures that are used to estimate truck volumes during any periods of time when the equip- ment was malfunctioning or out of service. For such sites, 0.01 is a suggested conservative value for the CVs. Level 1B Sites Level 1B sites are sites on the same road as an associated Level 1A site. AADT at Level 1B sites is estimated by obtaining a set of short-duration classification counts at the site and applying factors obtained from the associated Level 1A site. Errors in the AADT estimates at Level 1B sites can be caused by the following: equipment malfunction at either the site or the associated Level 1A site, some of the inherent limitations of the factoring process (such as the use of an average "November/Tuesday" factor for any Tuesday in November), and differences in the seasonal and DOW patterns at the two sites. The CVs at these sites are clearly higher than those at Level 1A sites. If the associated Level 1A site were reasonably close to the Level 1B site in question, it would appear reasonable to assume a CV of 0.02. If the two sites are more distant, with some intersections/interchanges with significant truck routes occurring between the two sites, the CV is likely to be somewhat larger. Accordingly, for Level 1B sites, 0.02 is a suggested default value for the CV, and the user should increase this value (typically to 0.03 or 0.04) when the associated Level 1A site is relatively distant or lies beyond intersections or inter- changes with one or more major truck routes. B.3 Level 3 Sites Level 3A Sites Level 3A sites are sites that are on the same road as an associated Level 1 or 2 site and for which a volume count exists (for a time period during which traffic was also being counted at the associated site) but for which classification counts do not exist. Traffic volumes at Level 3A sites are sufficiently different from those at their associated sites to warrant separate analyses of the Level 3A sites, but a Level 3A site and its associated site are presumed to have fairly similar percentages of vehicles in the major truck classes. Two sources of error affect AADT estimates developed for a Level 3A site but do not affect the estimates developed for the associated site: The ratio of the short-duration volume counts obtained at the two sites may not accurately reflect the ratio of AADT at the two sites and 1-56

OCR for page R1
The distribution of total traffic among vehicle classes is likely to differ somewhat at the two sites. Because of these differences, and particularly because of the second difference, CVs for Level 3A sites should exceed the CVs at the associated sites by 0.10 for associated Level 1 sites and by 0.08 for associated Level 2 sites.4 Level 3B Sites Level 3B sites are sites for which TrafLoad produces only an estimate of annual average daily truck traffic (AADTT) and identification of the site's Truck Traffic Classification (TTC) group. There are two kinds of Level 3B site. The first kind of Level 3B site is one for which a recent traffic count exists. For such sites, AADT is estimated by factoring the traffic count, and AADTT is estimated by multiplying AADT by a nonsite-specific estimate of percent trucks. As stated in Part 2, Section 3.5, this procedure should only be applied at sites with very low volumes of trucks and buses. Errors in the fac- toring process and in the estimate of percent trucks are likely to produce fairly high CVs. CVs of 0.30 should be used for these sites. The second kind of Level 3B site is on a planned new road. For these sites, estimates of total traffic and truck volumes are, at best, developed from travel demand models. Errors in the resulting estimates of AADTT are likely to be quite high. For these sites, use CVs of 1.0. B.4 Applying the CVs The factoring procedure presented in Part 2, Section 3.3 uses one set of factors for each Type 1 VC group. For purposes of discussion, assume that there is one VC group for all SUTs and a second group for all combination trucks (CTs). As a result, for any Level 2 site, the procedures for estimating CVs will produce the same value for the CV for the four SUT vehicle classes and a different value for the CV for the six CT classes. The same will be true for Level 3A sites whose associated site is a Level 2 site. For any other Level 3 sites, and for any Level 1 site, the above procedures produce a single value for the CV for all 10 vehicle classes. Consider the AADT estimates produced for the six CT classes for a Level 2 site. These errors will be correlated with each other (because of imperfections in the factors applied to the six classification counts), but the correlations will not be perfect (in part, because of differences in the seasonal and DOW patterns exhibited by the six classes). The Pavement Design Guide soft- 4 A lower increment is proposed for associated Level 2 sites because of the tendency of errors from inde- pendent sources to partially cancel each other. This tendency has very little effect for associated Level 1 sites (because Level 1 sites have very small CVs), but it is likely to have a moderate effect for associ- ated Level 2 sites. 1-57

OCR for page R1
ware, on the other hand, can be designed to assume that the errors are totally uncorrelated or that they are perfectly correlated. Because uncorrelated errors tend to cancel each other, the former option will tend to understate the effects of the partially correlated errors while the latter option will tend to overstate these effects. The goal of producing conservative pavement designs suggests that it would be preferable for the Pavement Design Guide software to adopt the latter option, i.e., to assume that the errors for all CT classes are perfectly correlated. Though the discussion here focused on Level 2 sites, this assumption appears to be appropriate for Level 1 and 3 sites as well. In the case of the SUT classes, the conclusion is somewhat different. In many areas, there is likely to be substantial correlation between the AADT errors for Vehicle Classes 6 and 7 (three- axle trucks and trucks with four or more axles). However, these two classes are likely to exhibit seasonal and DOW patterns that differ greatly from those exhibited by buses (Class 4) and by two-axle, six-tire trucks (Class 5). Hence, it would be preferable for the Pavement Design Guide software to use three independent estimates of AADT error for SUTs: one for Class 4, one for Class 5, and one for Classes 6 and 7. 1-58

OCR for page R1
Part 2 Guidelines for Collecting Traffic Data to Be Used in Pavement Design