Prediction of the number of defects in image sensors by vm using equipment QC data

Toshiya Okazaki, Kosuke Okusa, Kyo Yoshida

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper describes methods and evaluation results of predicting the number of defects in image sensors. We used regression tree and stepwise AIC for variable selection and generalized linear model for regression, instead of partial least squares (PLS) regression. The results showed improvement in prediction performance in comparison with the conventional method. By this, we could predict other countable values such as defects or dust particles.

Original languageEnglish
Title of host publication2018 International Symposium on Semiconductor Manufacturing, ISSM 2018 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538662687
DOIs
Publication statusPublished - Feb 22 2019
Event2018 International Symposium on Semiconductor Manufacturing, ISSM 2018 - Tokyo, Japan
Duration: Dec 10 2018Dec 11 2018

Publication series

NameIEEE International Symposium on Semiconductor Manufacturing Conference Proceedings
Volume2018-December
ISSN (Print)1523-553X

Conference

Conference2018 International Symposium on Semiconductor Manufacturing, ISSM 2018
CountryJapan
CityTokyo
Period12/10/1812/11/18

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Engineering(all)
  • Industrial and Manufacturing Engineering
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Prediction of the number of defects in image sensors by vm using equipment QC data'. Together they form a unique fingerprint.

Cite this