ABSTRACT
This study was conducted to locally calibrate the models in the AASHTOWare Pavement ME Design software for flexible pavements in Kansas. Twenty-seven newly constructed asphalt pavement projects were selected to calibrate the models. The traditional split sampling method was followed in the calibration process. MEPDG-predicted distresses of road segments were compared with the measured values. Statistical analysis was performed using the Microsoft Excel statistical toolbox. The permanent deformation model was calibrated using iterative runs of AASHTOWare to determine optimal coefficients that minimized the bias. The top-down fatigue cracking model and the International Roughness Index (IRI) model was calibrated using the generalized reduced gradient nonlinear optimization technique in Microsoft Excel Solver. Results showed that the local calibration process significantly reduced the bias between measured and predicted distress and IRI. Furthermore, the standard error of estimated (Se) was also decreased after local calibration.
