Calibration of the Flexible Pavement Distress Prediction Models in the Mechanistic Empirical Pavement Design Guide (MEPDG) for North Carolina

JADOUN, FADI MUNIR. Calibration of the Flexible Pavement Distress Prediction Models in the Mechanistic Empirical Pavement Design Guide (MEPDG) for North Carolina. (Under the direction of Y. Richard Kim). Roadway pavement structures in the United States have been designed over the last several decades using empirical-based procedures that were developed using performance data measured in the 1950s at the AASHO road test in Ottawa, Illinois. Because of the significant changes in truck axle loads and configurations, truck tire pressure, construction practices and materials, in addition to climate and subgrade soil differences from one location to another, these empirical design procedures have become unsuitable for the current design of new and rehabilitated pavement structures. Due to the limitations of the empirical-based pavement design procedures, the AASHTO Joint Task Force on pavements initiated an effort in 1996 to develop an improved pavement design guide that employs existing state-of-thepractice mechanistic-based models and design procedures. The product of this initiative became available in 2004 in the form of software called the Mechanistic-Empirical Pavement Design Guide (MEPDG). The mechanistic component of the MEPDG calculates pavement critical responses based on layer material properties and traffic loading. The empirical component bridges the gap between laboratory and field performance. The performance prediction models in the MEPDG were calibrated and validated using performance data measured for hundreds of sections across the United States. However, these nationally calibrated models do not necessarily reflect local materials, local construction practices, and local traffic characteristics. This dissertation focuses on North Carolina pavement structures, for which the nationally calibrated models do not lead to accurate pavement designs, as was found from verification work. The MEPDG distress prediction models must be recalibrated using local materials, traffic, and environmental data. Realizing the huge impact of traffic on pavement performance, it is also necessary to characterize local traffic, to develop valid estimates of traffic volume, traffic growth factors, and truck classifications and axle loads as an integral part of a reliable pavement design. The work presented in this dissertation focuses on four major topics: 1) the permanent deformation (or so-called rutting) performance characterization of twelve asphalt mixtures commonly used in North Carolina; 2) recalibration of the flexible pavement distress prediction models in the MEPDG to reflect local materials and conditions; 3) the development of a GIS-based methodology to enable the extraction of local subgrade soils data from a national soils database; and 4) the unique characterization of local North Carolina traffic while considering the effect of traffic on pavement performance. For the rutting performance characterization, triaxial repeated load permanent deformation (TRLPD) confined tests are performed, and material-specific rutting model coefficients for twelve common asphalt mixtures are developed. In order to calibrate the rutting prediction model coefficients, two optimization approaches are evaluated; one uses the generalized reduced gradient (GRG) method, and the other uses a genetic algorithm (GA) optimization technique. For the subgrade materials, a GIS-based methodology is developed to take advantage of the NCHRP 9-23A national soils database. The method allows any road section in North Carolina to be superimposed accurately onto pre-developed soil maps. Regarding traffic characterization, the MEPDG is employed to develop damage factor regression models to estimate the relative damage caused by 140 axle type/load combinations. Damage factors are essential for proper clustering of multiple MEPDG traffic inputs. Materials and traffic information is used successfully to recalibrate the rutting and alligator cracking prediction models to better capture monitored performance. Issues related to the accuracy of monitored distresses, the necessity for forensic studies, unbound materials characterization, and other factors are acknowledged herein, and recommendations are made for improvements in these areas. The outcomes from this dissertation work are summarized to make available a guideline that benefits agencies performing MEPDG local calibration. Expected barriers and proposed methodologies for different stages of local calibration are discussed. Calibration of the Flexible Pavement Distress Prediction Models in the Mechanistic Empirical Pavement Design Guide (MEPDG) for North Carolina by Fadi Munir Jadoun A dissertation submitted to the Graduate Faculty of North Carolina State University in partial fulfillment of the requirements for the degree of Doctor of Philosophy