An Innovative Lithography Process Window Decision Based On Aggregation of Multi Machine Learning Approaches

Automated image analysis and image classification system that based on machine learning have been developed and applied to the PWQ/FEM flow to enhance process stability and accuracy. Moving these task from artificial to automation has the added benefit of avoiding person-to-person inconsistencies in classification. In this paper, we approach a data flow consists of an automated machine learning-enabled process window analysis system that relies on CDSEM images taken on a Focus/Exposure Matrix wafer. Combined systematic data preprocessing and aggregation of multi machine learning approaches, we improved the accuracy of the process window analysis over 94%.