Semiconductor device geometry methods and systems

摘要

Systems and methods for predicting substrate geometry associated with a patterning process are described. input information including process information and/or geometry information for the pattern is received; Using a machine learning prediction model, the multidimensional output substrate geometry is predicted. The multidimensional output information includes pattern probability images. A probabilistic edge placement error band and/or a probabilistic failure rate may be predicted based on the pattern probabilistic images. The input information includes simulated aerial images, simulated resist images, target substrate dimensions, and/or data from a scanner associated with semiconductor device fabrication. For example, different aerial images may correspond to different heights in resist layers associated with the patterning process.