Local Calibration of MEPDG for Flexible Pavements in New Mexico

摘要

Local calibration of the mechanistic-empirical pavement design guide (MEPDG) is performed by determining the pavement-performance model coefficients to minimize the difference between the measured and predicted distresses of New Mexico department of transportation (NMDOT) pavements. A total of 24 New Mexico pavement sections, which have all the MEPDG inputs and quantitative-distress values required for MEPDG calibration, were used for calibration. Pavement-performance models such as rutting, alligator cracking, longitudinal cracking, and roughness models were calibrated by an error-minimization algorithm. In the calibration methodology, the target was fixed to reduce the sum of squared errors, defined by the square of the difference between predicted and measured distress, so that any bias was eliminated and precision was increased. The optimized calibration coefficients are: β_(r1) = 1.1, β_(r2) = 1.1, β_(r3) = 0.8, β_(GB)= 0.8, and β_(SG) = 1.2 for the rutting model; C_1 = 0.625, C_2 = 0.25, and C_3 = 6,000 for alligator cracking; C_1 = 3, C_2 = 0.3, and C_3 = 1,000 for longitudinal cracking; and site factor = 0.015 for roughness. The results show that these calibration coefficients reduce error in the MEPDG prediction and assist better design of flexible pavements using MEPDG in New Mexico.