|The AASHTOWareTM Pavement ME Design software was adopted by the American Association of State Highway and Transportation Officials (AASHTO) for structural pavement designs. The performance prediction models in the software were only calibrated based on a national database of pavement sections in the U.S. and Canada. These models may not apply to local pavement designs due to insufficient adequacy. The National Center for Asphalt Technology (NCAT), equipped with a full-scale accelerated pavement Test Track and asphalt materials laboratory, supported this study on evaluation, local calibration, and validation of the rutting, bottom-up fatigue cracking, and IRI models. The NCAT database was developed with research-grade detail and accuracy and was locally-based regarding the information of materials, traffic, and climate, and field performance. In the process of local calibration, automation was used during software runs and data compiling to minimize human interaction with the computer. Considerable labor savings (100% reduction) and time savings (nearly 35% reduction) were gained. As for evaluation results, over-predictions by the nationally-calibrated rutting and bottom-up fatigue cracking model were seen for a majority of experimental sections, and local calibration reduced bias and standard error of the estimate. The IRI prediction by the nationally-calibrated model was only accurate between 35 in./mile to 65 in./mile, and local calibration insignificantly improved the IRI prediction accuracy. The improvement of model accuracy was adequately validated for the locally-calibrated rutting and bottom-up fatigue cracking model, but not for the locally-calibrated IRI model, using independent local datasets. The recommended calibration coefficients should be evaluated based on a local database if they are intended for other design conditions. The automation method is recommended for future calibration studies since benefits in saving time and labor cost.