Wafer Map-based Defect Detection Using Convolutional Neural Networks

Abstract

The Electrical die sorting (EDS) test is performed to discriminate defective wafers for the purpose of improving the yield of the wafers during the semiconductor manufacturing process, and wafer maps are generated as a result. Semiconductor manufacturing process and equipment engineers use the patterns of the wafer map based on their knowledge to judge the defective wafer and estimate the cause. We use convolutional neural network which demonstrate good performance in the image classification. The convolutional neural network is used as a classification model of which the image of wafer map itself as input and whether the image is good or bad as output. While previous studies have used hand-crafted features for wafer map-based fault detection, the methodology used in this study is that the convolutional neural network learns the features useful for classification, it has the advantage of integrating knowledge. We show that the proposed classifier has better prediction accuracy than the conventional machine learning based techniques such as multilayer perceptron and random forest empirically by experiments on the data collected in the actual semiconductor manufacturing process.

Development of a Wafer Defect Pattern Classifier Using Polar Coordinate System Transformed Inputs and Convolutional Neural Networks

“…For this reason, many semiconductor companies are continuously researching to automate defect pattern classification using machine learning and deep learning techniques. Existing studies show that efforts have been made to improve the classification accuracy of defect patterns by using WBM images as training data for CNNs [5][6][7]. Upon analyzing each defect pattern on the WBM image used here, we can see that there are positional and geometric features in the distribution of defect dies for each defect pattern class.…”

“…Conventional classification methods, such as SVM, were applied in the beginning [10]. More recently, researchers have employed deep learning methods, mostly based on convolutional neural network (CNN) models [5][6][7]. CNNs possess translational invariance, which ensures that the absolute position of any image does not affect the classification performance of the model.…”

Hot Strip Mill Gearbox Monitoring and Diagnosis Based on Convolutional Neural Networks Using the Pseudo-Labeling Method

“…We will be examining related research on equipment abnormality detection using CNN after confirming the application cases of autoencoder, a deep learning method that is unsupervised [10][11][12]. We will be examining the latest research trends combining LSTMs with autoencoders, both of which are specialized for the analysis of time series data [13][14][15]. Perera and Brage [16] proposed a method of classifying data using the expectation-maximization technique and a Gaussian mixture model (GMM) and then monitoring performance using an autoencoder (AE, an unsupervised learning algorithm).…”