Local Calibration of the MEPDG for Flexible Pavement Design

MEPDG;Calibration;Mechanistic-Empirical;Rutting;Fatigue

Muthadi, Naresh Reddy ; Y. Richard Kim, Committee Chair,Murthy N. Guddati, Committee Member,T. Matthew Evans, Committee Member,Muthadi, Naresh Reddy ; Y. Richard Kim ; Committee Chair ; Murthy N. Guddati ; Committee Member ; T. Matthew Evans ; Committee Member

University:North Carolina State University

关键词: MEPDG;    Calibration;    Mechanistic-Empirical;    Rutting;    Fatigue;   

Others  :  https://repository.lib.ncsu.edu/bitstream/handle/1840.16/486/etd.pdf?sequence=1&isAllowed=y

美国|英语

来源: null

【 摘 要 】

The 1993 American Association of State Highway and Transportation Officials (AASHTO) Guide for Design of Pavement Structures is a mere modification of the empirical methods found in its earlier versions that are based on regression equations relating simple material and traffic inputs. Although the various editions of the AASHTO design guide have served well for several decades, they contain too many limitations to be continued as the nation's primary pavement design procedures. The Mechanistic-Empirical Pavement Design Guide (MEPDG) procedure, on the other hand, provides the tools for evaluating the effect of variations in input data on pavement performance. The design method in the MEPDG is mechanistic because it uses stresses and strains in a pavement system calculated from the pavement response model to predict the performance of the pavement. The empirical nature of the design method stems from the fact that the pavement performance predicted from laboratory-developed performance models is adjusted based on the observed performance from the field to reflect the differences between predicted and actual field performance. The performance models used in the MEPDG are calibrated using limited national databases and, thus, it is necessary to calibrate these models for local highway agencies implementation by taking into account local materials, traffic information, and environmental conditions. Two distress models, permanent deformation and bottom-up fatigue cracking (hereafter referred to as alligator cracking), were employed for this effort. Fifty-three pavement sections were selected for the calibration and validation process:30 long-term pavement performance (LTPP) pavements, which include 16 new flexible pavement sections and 14 rehabilitated sections, and 23 North Carolina Department of Transportation (NCDOT) sections. All the necessary data were obtained from the LTPP and the NCDOT databases. To provide reasonable values in cases where data were missing, MEPDG defaults, NCDOT typical range of values, and engineering judgment were employed. Finally, an experimental matrix is developed to identify any bias resulting from the use of local materials and conditions. The NCDOT currently relies on a subjective rating or non-numeric rating system of the permanent deformation data, which presented difficulties in the conversion to the MEPDG format.The verification runs for the LTPP sections using the parameters developed during the national calibration effort under the NCHRP (National Cooperative Highway Research Program) 1-37A project showed promising results. Microsoft Excel Solver was used to fit the predicted rut depth values to the measured values by changing the coefficients in the permanent deformation models for hot-mix asphalt (HMA) and unbound materials. This process was employed for each of the permanent deformation models separately. For the alligator cracking model, the only possibility of reducing the standard error and bias is through the transfer function. Again, Microsoft Excel Solver was used to minimize the sum of the squared errors of the measured and predicted cracking by varying the C1 and C2 parameters of the transfer function. It was found that there is no significant difference between the local calibrated standard error and the global standard error for the HMA permanent deformation model as well as the alligator cracking model. Therefore, it was decided to keep both the models for a more robust calibration in the future that would increase the number of sections and include more detailed inputs (mostly Level 1 inputs).