The standard method for defect disposition and verification of repair success in the mask shop is through the utilization of the aerial imaging platform, AIMS<sup>TM</sup>. The CD (Critical Dimension) deviation of the defective or repaired region as well as the pattern shift can be calculated by comparing the measured aerial images of this region to that of a reference. Through this analysis it can be determined if the defect or repaired region will be printed on the wafer under the illumination conditions of the scanner. The analysis of the measured aerial images from the AIMS<sup>TM</sup> are commonly performed manually using the analysis software available on the system or with the help of an analysis software called RV (Repair Verification). Because the process is manual, it is not standardized and is subject to operator variations. This method of manual aerial image analysis is time consuming, dependent on the skill level of the operator and significantly contributes to the overall mask manufacturing process flow. AutoAnalysis (AA), the first application available for the FAVOR® platform, provides fully automated analysis of AIMS<sup>TM</sup> aerial images  and runs in parallel to the measurement of the aerial images. In this paper, we investigate the initial AutoAnalysis performance compared to the conventional method using RV and its application to a production environment. The evaluation is based on the defect CD of three pattern types: contact holes, dense line and spaces and peripheral structure. The defect analysis results for different patterns and illumination conditions will be correlated and challenges in transitioning to the new approach will be discussed.