Abstract: In some embodiments, a computer-implemented method of controlling a semiconductor manufacturing process is provided. A computing system generates predicted metrology values for a current run and a next run by providing metrology forecast inputs to a metrology forecast model. The computing system generates an updated recipe for executing at least one semiconductor manufacturing process step using the predicted metrology values for the current run and the next run.

Abstract: A method is disclosed for generating a corrective film pattern for reducing wafer bow in a semiconductor wafer fabrication process. The method inputs to a neural network a wafer bow signature for a predetermined semiconductor fabrication step. The neural network generates from the input a corrective film pattern corresponding to the wafer bow signature. The neural network is trained with a training dataset of wafer shape transformations and corresponding corrective film patterns.

Abstract: A system may receive a training data set including a set of target wafer shape transformations corresponding to negatives of warpage signatures of semiconductor wafers, and a set of training corrective film patterns that reduce warpage signatures of semiconductor wafers. A system may provide a surrogate machine learning model that includes a forward model comprising a first neural network model configured to take as input a corrective film pattern and output a corresponding wafer shape transformation and an inverse model comprising a second neural network model configured to take as input a wafer shape transformation and output a corresponding corrective film pattern. A system may train a forward model using training corrective film patterns. A system may train an inverse model using target wafer shape transformations, output from a forward model, and by calculating a loss that includes a regularization penalty for film patterns outside a main distribution.

Abstract: A method is disclosed for generating a corrective film pattern for reducing wafer bow in a semiconductor wafer fabrication process. The method inputs to a neural network a wafer bow signature for a predetermined semiconductor fabrication step. The neural network generates from the input a corrective film pattern corresponding to the wafer bow signature. The neural network is trained with a training dataset of wafer shape transformations and corresponding corrective film patterns.

Abstract: A system for defining structures within an image is described. The system includes reading of an input file, preprocessing the input file while preserving metadata such as scale information and then detecting features of the input file. In one version the detection first uses an edge detector followed by identification of features using a Hough transform. The output of the process is identified elements within the image.