Use of Pavement Management Information System for Verification of Mechanistic-Empirical Pavement Design Guide Performance Predictions

The performance models used in the Mechanistic-Empirical Pavement Design Guide (MEPDG) are nationally calibrated with design inputs and performance data obtained primarily from the national Long-Term Pavement Performance database. It is necessary to verify and calibrate MEPDG performance models for local highway agencies' implementation by taking into account local materials, traffic information, and environmental conditions. This paper discusses the existing pavement management information system (PMIS) with respect to the MEPDG and the accuracy of the nationally calibrated MEPDG prediction models for Iowa highway conditions. All the available PMIS data for Interstate and primary road systems in Iowa were retrieved from the Iowa Department of Transportation (DOT) PMIS. The retrieved databases were then compared and evaluated with respect to the input requirements and outputs for Version 1.0 of the MEPDG software. Using Iowa DOT's comprehensive PMIS database, researchers selected 16 types of pavement sections across Iowa (not used for national calibration in the NCHRP 1-37A study). A database of MEPDG inputs and the actual pavement performance measures for the selected pavement sites were prepared for verification. The accuracy of the MEPDG performance models for Iowa conditions was statistically evaluated. The verification testing showed promising results in terms of MEPDG's performance prediction accuracy for Iowa conditions. Recalibrating the MEPDG performance models for Iowa conditions is recommended to improve the accuracy of pavement performance predictions. Disciplines Civil and Environmental Engineering | Construction Engineering and Management Comments This article is from Transportation Research Record: Journal of the Transportation Research Board, 2153 (2010); 30-39, doi: 10.3141/2153-04. Posted with permission. Authors Sunghwan Kim, Halil Ceylan, Kasthurirangan Gopalakrishnan, and Omar G. Smadi This article is available at Iowa State University Digital Repository: http://lib.dr.iastate.edu/ccee_pubs/21 cal regression techniques relating simple material characterizations, traffic characterization, and measures of performance. In recognition of the limitations of the current AASHTO guide, the new Mechanistic–Empirical Pavement Design Guide (MEPDG) and its software were developed through NCHRP Project 1-37A (5). The mechanistic part of MEPDG is the application of the principles of engineering mechanics to calculate pavement responses (stresses, strains, and deflection) under loads for the predictions of the pavement performance history. The empirical nature of the MEPDG stems from the laboratory-developed pavement performance models being adjusted or calibrated to the observed performance measurements (distresses) from the actual pavements. The MEPDG’s mechanistic–empirical procedure will require an even greater effort to successfully implement a useful design procedure. Without calibration, the results of mechanistic calculations cannot be used to predict rutting, cracking, and faulting with any degree of confidence. The distress mechanisms are far more complex than can be practically modeled; therefore, the use of empirical factors and calibration is necessary to obtain realistic performance predictions. The MEPDG does not provide a design thickness as the end products; instead, it provides the pavement performance throughout its design life. The design thickness can be determined by modifying design inputs and obtaining the best performance with an iterative procedure. The performance models used in the MEPDG are nationally calibrated using design inputs and performance data largely from the national Long-Term Pavement Performance (LTPP) database. The LTPP database used for national (global) calibration of MEPDG includes no hot-mix asphalt (HMA) sections and only one portland cement concrete (PCC) pavement section in Iowa (5). Thus, it is necessary to calibrate MEPDG performance models for local highway agencies’ use by taking into account local materials, traffic information, and environmental conditions. The local calibration process involves three important steps: verification, calibration, and validation (6). Verification refers to assessing the accuracy of the nationally (globally) calibrated prediction models for local conditions. Calibration refers to the mathematical process through which the total error or difference between observed and predicted values of distress is minimized. Validation refers to the process to confirm that the calibrated model can produce robust and accurate predictions for cases other than those used for model calibration. The first step of the local calibration plan is to perform verification runs on the pavement sections using the nationally calibrated MEPDG performance models (6). The MEPDG (5) recommends Use of Pavement Management Information System for Verification of Mechanistic–Empirical Pavement Design Guide Performance Predictions Sunghwan Kim, Halil Ceylan, Kasthurirangan Gopalakrishnan, and Omar Smadi 30 The performance models used in the Mechanistic–Empirical Pavement Design Guide (MEPDG) are nationally calibrated with design inputs and performance data obtained primarily from the national Long-Term Pavement Performance database. It is necessary to verify and calibrate MEPDG performance models for local highway agencies’ implementation by taking into account local materials, traffic information, and environmental conditions. This paper discusses the existing pavement management information system (PMIS) with respect to the MEPDG and the accuracy of the nationally calibrated MEPDG prediction models for Iowa highway conditions. All the available PMIS data for Interstate and primary road systems in Iowa were retrieved from the Iowa Department of Transportation (DOT) PMIS. The retrieved databases were then compared and evaluated with respect to the input requirements and outputs for Version 1.0 of the MEPDG software. Using Iowa DOT’s comprehensive PMIS database, researchers selected 16 types of pavement sections across Iowa (not used for national calibration in the NCHRP 1-37A study). A database of MEPDG inputs and the actual pavement performance measures for the selected pavement sites were prepared for verification. The accuracy of the MEPDG performance models for Iowa conditions was statistically evaluated. The verification testing showed promising results in terms of MEPDG’s performance prediction accuracy for Iowa conditions. Recalibrating the MEPDG performance models for Iowa conditions is recommended to improve the accuracy of pavement performance predictions. The current AASHTO Design Guide is based on methods that have evolved from the AASHO Road Test (1958–1961) (1). Through a number of editions from the initial publication in 1962, the interim guide in 1972 (2) and later editions (3, 4), minor changes and improvements have been made. Nonetheless, these later modifications have not significantly altered the original methods, which are based on empiriS. Kim, 192 Town Engineering Building; H. Ceylan, 406 Town Engineering Building; and K. Gopalakrishnan, 353 Town Engineering Building, Department of Civil, Construction, and Environmental Engineering, Iowa State University, Ames, IA 50011-3232. O. Smadi, Center for Transportation Research and Education, Institute for Transportation, Iowa State University, Ames, IA 50010-8664. Corresponding author: S. Kim, [email protected]. Transportation Research Record: Journal of the Transportation Research Board, No. 2153, Transportation Research Board of the National Academies, Washington, D.C., 2010, pp. 30–39. DOI: 10.3141/2153-04 that a verification database be developed to confirm that the national calibration factors or functions of performance models are adequate and appropriate for the construction, materials, climate, traffic, and other local conditions. The input data types required for analysis using the MEPDG software range from simple data, such as the pavement design features and pavement geometrics, to detailed data obtained from destructive testing (e.g., HMA dynamic modulus and PCC elastic modulus), nondestructive testing (e.g., falling weight deflectometer testing), and drainage surveys. The performance measures projected from MEPDG include longitudinal cracking, rutting, fatigue cracking, and thermal cracking for HMA pavements, and jointed plain concrete pavement (JPCP) joint faulting, JPCP transverse cracking, and continuously reinforced concrete pavement (CRCP) punch-outs (with limited crack width calibration) for PCC pavements. International Roughness Index (IRI) is also projected for new and rehabilitated pavement systems. Many of this information actually measured can be obtained from the local agency’s pavement management information system (PMIS). However, it is also needed to systematically evaluate the existing PMIS with respect to the MEPDG input parameters and projected performance measure results for local calibration. SCOPE AND OBJECTIVES