AASHTO Mechanistic- Empirical Pavement Design Guide Parametric Study CFIRE CFIRE 03-24 March 2012 National Center for Freight & Infrastructure Research & Education Department of Civil and Environmental Engineering College of Engineering University of Wisconsin–Madison Authors: Ruipeng Li and Steven Cramer University of Wisconsin–Madison Principal Investigator: Steven Cramer University of Wisconsin–Madison 2 DISCLAIMER This research was funded by the National Center for Freight and Infrastructure Research and Education and the Wisconsin Highway Research Program of the Wisconsin Department of Transportation. The contents of this report reflect the views of the authors, who are responsible for the facts and the accuracy of the information presented herein. This document is disseminated under the sponsorship of the Department of Transportation, University Transportation Centers Program, in the interest of information exchange. The U.S. Government assumes no liability for the contents or use thereof. The contents do not necessarily reflect the official views of the National Center for Freight and Infrastructure Research and Education, the University of Wisconsin, the Wisconsin Department of Transportation, or the USDOT’s RITA at the time of publication. The United States Government assumes no liability for its contents or use thereof. This report does not constitute a standard, specification, or regulation. The United States Government does not endorse products or manufacturers. Trade and manufacturers names appear in this report only because they are considered essential to the object of the document. Wisconsin Highway Research Program AASHTO Mechanistic–Empirical Pavement Design Guide Parametric Study Institution: University of Wisconsin-Madison Principal Investigator: Steven Cramer Authors: Ruipeng Li and Steven Cramer Date: March 2012 ii Technical Report Documentation Page 1. Report No. 2. Government Accession No. 3. Recipient’s Catalog No. CFIRE 03-24 CFDA 20.701 4. Title and Subtitle 5. Report Date March 2012 AASHTO Mechanistic-Empirical Pavement Design Guide Parametric Study 6. Performing Organization Code 7. Author/s 8. Performing Organization Report No. Ruipeng Li and Steven Cramer, University of Wisconsin-Madison CFIRE 03-24 9. Performing Organization Name and Address 10. Work Unit No. (TRAIS) National Center for Freight and Infrastructure Research and Education (CFIRE) University of Wisconsin-Madison 11. Contract or Grant No. DTRT06-G-0020 1415 Engineering Drive, 2205 EH Madison, WI 53706 12. Sponsoring Organization Name and Address 13. Type of Report and Period Covered Research and Innovative Technology Administration Final Report [1/1/2010 – United States Department of Transportation 3/31/2012] 1200 New Jersey Avenue, SE 14. Sponsoring Agency Code Washington, DC 20590 15. Supplementary Notes Project completed for the Wisconsin DOT by CFIRE. 16. Abstract This study focuses on assessing the robustness of the AASHTO Mechanistic-Empirical Pavement Design Guide (MEPDG v 1.1) for rigid pavement design projects in Wisconsin. The primary tasks conducted in this study included performing sensitivity analysis on MEPDG’s inputs for jointed plain concrete pavement (JPCP) design, evaluating the practicality of each input parameter’s sensitivity, determining the effects of using different concrete materials and different hierarchical levels of inputs on predicted JPCP performances, and identifying the limitations of the current MEPDG. It was found that the coefficient of thermal expansion and modulus of rupture had a strong impact on MEPDG predictions and these outcomes appeared consistent with conventional wisdom. The adverse effects of the concrete unit weight and the positive benefits of the widened concrete slab were suspected to have been overestimated by the MEPDG. It was verified that MEPDG predictions were significantly different depending on the concrete materials and hierarchical levels of inputs chosen. Therefore, the pavement designer was recommended to select the proper design strategy. The limitations of MEPDG were illustrated in that it did not account for the effects brought by the supplementary cementitious materials on the pavement design and it erroneously restrained the permissible range for concrete modulus of rupture input. Although a significant number of material inputs are possible with level 1 implementation of the MEPDG, these inputs are often used in a narrow context and are largely independent from other inputs and material properties. It cannot be presumed that the change in one input will lead to a full set of parametric changes that such a change would actually induce in practice. 17. Key Words 18. Distribution Statement No restrictions. This report is available through the Transportation Research Pavement Design Guide, MEPDG, AASHTO, Information Services of the National Transportation Library. Concrete, Coefficient of Thermal Expansion 19. Security Classification (of this report) 20. Security Classification (of this page) 21. No. Of Pages 22. Price Unclassified Unclassified 19 -0- Form DOT F 1700.7 (8-72) Reproduction of form and completed page is authorized. iii Table of Contents Project Background .................................................................................................................................... 1 Research Task ............................................................................................................................................. 1 1. Software installation ............................................................................................................................. 1 2. Identification of Variables Required for Pavement Design .................................................................. 1 3. Sensitivity Analysis .............................................................................................................................. 2 3.1. Single-Variable Sensitivity Analysis ............................................................................................. 2 3.2. Effects on Pavement Thicknesses due to Concrete Materials ........................................................ 8 4. Effects on Pavement Thicknesses due to Different Hierarchical Strength Input Options .................. 11 5. MEPDG Software’s Limitations ......................................................................................................... 11 Summary .................................................................................................................................................... 12 Reference ................................................................................................................................................... 12 Appendix A Detailed Inputs Description for MEPDG .................................................................... 13 Appendix B Hierarchical Strength Inputs of PCC Layer for JPCP New Design ......................... 15 Appendix C Sample Inputs for Sensitivity Analysis ........................................................................ 17 Appendix D Curve Fitting Analysis ................................................................................................. 18 1 Project Background More economical and durable road construction has motivated the Wisconsin Department of Transportation (WisDOT) to consider implementing the new AASHTO Mechanistic-Empirical Pavement Design Guide (MEPDG). This guide is a computational tool for the new and rehabilitation pavement design utilizing a set of mechanistic and empirical models to predict the future performance of the pavements. In particular, major factors influencing the behavior and durability of a pavement, such as climate, traffic, and material characteristics are integrated within the considerations of the built-in models to obtain a precise prediction of performance and durability of pavements. The potential benefits of the MEPDG are to provide engineers and contractors with advanced tools and improvements in conservation, management, as well as significant economic savings. The objectives of the project included installing and operating the MEPDG software to identify all the required data inputs, to assess the most sensitive inputs, and to identify MEPDG’s limitations. The main parameters required by the MEPDG for the prediction of the new or rehabilitation design of Jointed Plain Concrete Pavement (JPCP), Continuously Reinforced Concrete Pavement (CRCP), and Portland Cement Concrete (PCC) overlay include information on traffic, climate, and pavement structure. Particularly for the pavement structure inputs, pavement layer design and material characteristics are specified. Detailed descriptions of each input part are included in Appendix A. Research Task 1. Software installation The MEPDG v1.1 installation software and the climatic files for the entire country were downloadable from the NCHRP website (NCHRP Design Guide, Mechanistic-Empirical Design of New & Rehabilitated Pavement Structures. http://onlinepubs.trb.org/onlinepubs/archive/mepdg/home.htm). The climatic files were installed in the same directory used for the MEPDG installation, allowing the MEPDG software to access the weather station data for pavement design. 2. Identification of Variables Required for Pavement Design Four categories of design inputs are required by MEPDG, which are general inputs, traffic inputs, climate inputs, and material characterization inputs. General inputs include site-specific information and analysis reliability information. Inputs within this part deal with pavement site conditions, selected design parameters for evaluation purpose, and the corresponding reliability for each design parameters. Traffic inputs include traffic volume information, wheel load information, and traffic adjustment factors. Climatic information from the selected weather station database and the water table depth at the construction site are required inputs for the climate section. The climatic factors that affect pavement design include temperature and moisture, causing PCC slabs to curl and warp. Material characterization inputs include basic material properties for computing pavement response and distress (Guide for Mechanistic-Empirical Design of New and Rehabilitated Pavement Structures. National Cooperative Highway Research Program, Transportation Research Board, National Research Council, 2004). This report is mainly focused on researching the influences of material property changes of the Portland Cement Concrete (PCC) Layer on predicted JPCP performances by MEPDG. Such material properties include basic ones for computing pavement responses such as PCC elastic modulus and Poisson’s ratio, additional ones for the distress/transfer functions such as PCC modulus of rupture and tensile strength, and others for climate model, such as index properties and thermal properties. PCC 2 modulus of rupture and modulus of elasticity are the key parameters for the MEPDG to calculate JPCP distress and response. Three hierarchical levels are permitted for obtaining these two parameters. The Level 1 option requires direct measurements of them at 7, 14, 28, and 90 day. The Level 2 option requires direct measurements on concrete compressive strength at 7, 14, 28, and 90 day and then it converts the compressive strength results to the corresponding modulus of elasticity and modulus of rupture results. The Level 3 option requires only one measurement or an estimation of the concrete 28- day compressive strength value or modulus of rupture value. It utilizes empirical relations to first convert the 28-day modulus of rupture result to the corresponding modulus of rupture results at 7, 14, and 90 day. These results are further converted to compressive strength values at the same dates. The same empirical relations used in level 2 option are used to acquire the corresponding modulus of elasticity results (Guide for Mechanistic-Empirical Design of New and Rehabilitated Pavement Structures. National Cooperative Highway Research Program, Transportation Research Board, National Research Council, 2004). Detailed descriptions of the three levels of input options are shown in Appendix B. 3. Sensitivity Analysis Three design parameters are proposed for JPCP design, which are transverse cracking, mean transverse faulting, and IRI. All of these design parameters must be satisfied at the specified reliability levels for a passed acceptable pavement design by MEPDG. Transverse cracking (unit, %) is defined as the ratio of the total number of transverse cracks across the breadth of slabs to the total number of slabs. Mean transverse faulting (unit, in) is defined as the average elevation difference between two adjacent slabs at the transverse joints for the entire pavement. IRI is the abbreviation for international roughness index, a parameter used to reflect the pavement surface smoothness condition, measured by in/mile. A high value of IRI denotes a highly deteriorated pavement surface (Guide for Mechanistic-Empirical Design of New and Rehabilitated Pavement Structures. National Cooperative Highway Research Program, Transportation Research Board, National Research Council, 2004). The sensitivity analysis conducted in the following section is to quantitatively relate the change of each individual input to the change of the design parameters. The inputs for the standard JPCP design are shown in Appendix C. 3.1. Single-Variable Sensitivity Analysis To quantify the sensitivity of each variable, ‘y’ was defined as the relative percentage change of a predicted performance (design parameter) due to the relative percentage change of an input. For example: ∆𝑐𝑟𝑎𝑐𝑘𝑖𝑛𝑔 𝑦𝑐𝑟𝑎𝑐𝑘𝑖𝑛𝑔 = 𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑐𝑟𝑎𝑐𝑘𝑖𝑛𝑔 ∆𝑓𝑎𝑢𝑙𝑡𝑖𝑛𝑔 𝑦𝑓𝑎𝑢𝑙𝑡𝑖𝑛𝑔 = 𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑓𝑎𝑢𝑙𝑡𝑖𝑛𝑔 where: ∆𝐼𝑅𝐼 𝑖𝑛𝑐𝑟𝑒𝑚𝑒𝑛𝑡 𝑦𝐼𝑅𝐼 = : cracking increment; 𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝐼𝑅𝐼 𝑖𝑛𝑐𝑟𝑒𝑚𝑒𝑛𝑡 : cracking of the standard design; ∆𝑐𝑟𝑎𝑐𝑘𝑖𝑛𝑔: faulting increment; 𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑐𝑟𝑎𝑐𝑘𝑖𝑛𝑔: faulting of the standard design; ∆𝑓𝑎𝑢𝑙𝑡𝑖𝑛𝑔 : the new predicted IRI increment minus the standard trial design’s IRI increment; 𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑓𝑎𝑢𝑙𝑡𝑖𝑛𝑔 : the standard trial design’s IRI increment; T∆𝐼h𝑅e 𝐼e 𝑖x𝑛p𝑐r𝑟es𝑒s𝑚io𝑒n𝑛 f𝑡or calculating IRI change was different because the PCC slab had an initial non-zero IRI v𝑖𝑛a𝑖l𝑡u𝑖e𝑎. 𝑙 𝐼𝑅𝐼 𝑖𝑛𝑐𝑟𝑒𝑚𝑒𝑛𝑡 3 The variable ‘x’ was defined as the relative percentage change of an input. For example, the x expression for the compressive strength was: ′ ′ ′ where: 𝑥𝑓𝑐=∆𝑓𝑐 𝑓𝑐 ′: change of the input ′; ′: the input ′ used in the standard JPCP trial design; ∆𝑓𝑐 𝑓𝑐 𝑓𝑐 Samp𝑓le𝑐 inputs for sensitivity analysis are shown in Appendix C. For each variable selected for sensitivity analysis, a trend line was constituted by the data points of x versus y. The lower limit of y is - 1, corresponding to the situation when the new predicted pavement performance (for example, cracking) is zero. The upper limit of y is unknown and could be a very large number. A typical x vs. y relation is shown in the Figure 1. It is observed that y increases rapidly when x is positive, and approaches -1 for negative x values. The shape shown in Figure 1 is analogous to the form of an exponent function, but it is possible that the shape of x versus y relation is close to a linear form for very insensitive inputs. Therefore, two forms of fitting curves were proposed to simulate each variable’s sensitivity, which were , a hybrid form with unknown values for A, B, C, and D, or , an exponent form with unknown values for A, B, and C. Based on the comparison of the two fitting 𝑥 𝑥 curve𝑦s,= 𝐴∙𝐵 +𝐶∙𝑥 +w𝐷as selected to simulate each variable’s sensitivity due to h𝑦ig=her𝐴 a∙c𝐵cur+ac𝐶y. (Details are shown in Appendix D) 𝑥 𝑦 = 𝐴∙𝐵 +𝐶 Figure 1 y vs. x relationship cracking CTE The sensitivity of a variable is defined as the change of the output due to a unit change of an input, which is expressed as the slope of the line crossing the origin, the point corresponding to the input and the output of the standard JPCP design. The example involving the definition of the sensitivity of CTE is shown in Figure 1, where line 1 is the tangent line crossing the origin. Sensitivity of CTE was