P‐59: In‐Fab Raman Spectroscopy for Defect Analysis of Random Failures

Abstract

Random defects, which account for more than 90% of electronic product manufacturing defects, are defects caused by foreign substances. Although the detection ability has been improved through various efforts using optical inspection and image features, the ultimate enhancement to improve quality and yield cannot be achieved just by detecting defects. In particular, there was a limit to classifying and specifying the failure source because the ingredients were not known, and it was often impossible to eliminate defects even after spending a lot of time. We developed a world first in‐fab Raman spectroscopy microscopy to be applied to the analysis of foreign substances in the entire area of the 8.5th generation glass substrate and applied it in real time between manufacturing. As a result, it is possible to detect defective sources in real time and to improve quality and yield through process control and ultimate improvement.

P‐167: Autoencoder with Variables Added Model for Manufacturing Process Anomaly Detection Using Multivariate Time‐series Data that Minimizes Data Information Loss

“…There are two ways to process multivariate time series data. A method of extracting features of data and converting them into a tabular form [1][2][3][4] and a method of encoding and combining univariate data into images [5][6][7][8]. Data processing is useful in that it removes noise and leaves only the information needed for modeling, but the loss of information in the data is unavoidable.…”