A Statistical Study on Highly Accurate Quality Prediction for High-mix Low-Volume Semiconductor Products

Abstract

Accurate prediction of product performance is very important in semiconductor manufacturing processes. Manufacturing plants with high-mix low-volume types face the problem of having to create many quality prediction models with a small sample size. In this high-mix-low-volume-type plant, the construction of a highly efficient and accurate prediction model for product performance is an important issue. In this study, we propose a quality prediction model based on the hierarchical Bayesian model that can predict quality with high accuracy even for a small number of samples.