A convolutional neural network for fault classification and diagnosis in semiconductor manufacturing processes

Ki Bum Lee, Sejune Cheon, Chang Ouk Kim

Research output: Contribution to journalArticle

31 Citations (Scopus)

Abstract

Many studies on the prediction of manufacturing results using sensor signals have been conducted in the field of fault detection and classification (FDC) for semiconductor manufacturing processes. However, fault diagnosis used to find clues as to root causes remains a challenging area. In particular, process monitoring using neural networks has been employed to only a limited extent because it is a black box model, making the relationships between input data and output results difficult to interpret in actual manufacturing settings, despite its high classification performance. In this paper, we propose a convolutional neural network (CNN) model, named FDC-CNN, in which a receptive field tailored to multivariate sensor signals slides along the time axis, to extract fault features. This approach enables the association of the output of the first convolutional layer with the structural meaning of the raw data, making it possible to locate the variable and time information that represents process faults. In an experiment on a chemical vapor deposition process, the proposed method outperformed other deep learning models.

Original languageEnglish
Pages (from-to)135-142
Number of pages8
JournalIEEE Transactions on Semiconductor Manufacturing
Volume30
Issue number2
DOIs
Publication statusPublished - 2017 May 1

Fingerprint

fault detection
manufacturing
Semiconductor materials
Neural networks
Fault detection
output
sensors
Process monitoring
Sensors
chutes
learning
Failure analysis
boxes
Chemical vapor deposition
vapor deposition
causes
predictions
Experiments
Deep learning

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Industrial and Manufacturing Engineering
  • Electrical and Electronic Engineering

Cite this

@article{988dd07948d64a91ad4c27e0d9ab4bac,
title = "A convolutional neural network for fault classification and diagnosis in semiconductor manufacturing processes",
abstract = "Many studies on the prediction of manufacturing results using sensor signals have been conducted in the field of fault detection and classification (FDC) for semiconductor manufacturing processes. However, fault diagnosis used to find clues as to root causes remains a challenging area. In particular, process monitoring using neural networks has been employed to only a limited extent because it is a black box model, making the relationships between input data and output results difficult to interpret in actual manufacturing settings, despite its high classification performance. In this paper, we propose a convolutional neural network (CNN) model, named FDC-CNN, in which a receptive field tailored to multivariate sensor signals slides along the time axis, to extract fault features. This approach enables the association of the output of the first convolutional layer with the structural meaning of the raw data, making it possible to locate the variable and time information that represents process faults. In an experiment on a chemical vapor deposition process, the proposed method outperformed other deep learning models.",
author = "Lee, {Ki Bum} and Sejune Cheon and Kim, {Chang Ouk}",
year = "2017",
month = "5",
day = "1",
doi = "10.1109/TSM.2017.2676245",
language = "English",
volume = "30",
pages = "135--142",
journal = "IEEE Transactions on Semiconductor Manufacturing",
issn = "0894-6507",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

A convolutional neural network for fault classification and diagnosis in semiconductor manufacturing processes. / Lee, Ki Bum; Cheon, Sejune; Kim, Chang Ouk.

In: IEEE Transactions on Semiconductor Manufacturing, Vol. 30, No. 2, 01.05.2017, p. 135-142.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A convolutional neural network for fault classification and diagnosis in semiconductor manufacturing processes

AU - Lee, Ki Bum

AU - Cheon, Sejune

AU - Kim, Chang Ouk

PY - 2017/5/1

Y1 - 2017/5/1

N2 - Many studies on the prediction of manufacturing results using sensor signals have been conducted in the field of fault detection and classification (FDC) for semiconductor manufacturing processes. However, fault diagnosis used to find clues as to root causes remains a challenging area. In particular, process monitoring using neural networks has been employed to only a limited extent because it is a black box model, making the relationships between input data and output results difficult to interpret in actual manufacturing settings, despite its high classification performance. In this paper, we propose a convolutional neural network (CNN) model, named FDC-CNN, in which a receptive field tailored to multivariate sensor signals slides along the time axis, to extract fault features. This approach enables the association of the output of the first convolutional layer with the structural meaning of the raw data, making it possible to locate the variable and time information that represents process faults. In an experiment on a chemical vapor deposition process, the proposed method outperformed other deep learning models.

AB - Many studies on the prediction of manufacturing results using sensor signals have been conducted in the field of fault detection and classification (FDC) for semiconductor manufacturing processes. However, fault diagnosis used to find clues as to root causes remains a challenging area. In particular, process monitoring using neural networks has been employed to only a limited extent because it is a black box model, making the relationships between input data and output results difficult to interpret in actual manufacturing settings, despite its high classification performance. In this paper, we propose a convolutional neural network (CNN) model, named FDC-CNN, in which a receptive field tailored to multivariate sensor signals slides along the time axis, to extract fault features. This approach enables the association of the output of the first convolutional layer with the structural meaning of the raw data, making it possible to locate the variable and time information that represents process faults. In an experiment on a chemical vapor deposition process, the proposed method outperformed other deep learning models.

UR - http://www.scopus.com/inward/record.url?scp=85018771617&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85018771617&partnerID=8YFLogxK

U2 - 10.1109/TSM.2017.2676245

DO - 10.1109/TSM.2017.2676245

M3 - Article

VL - 30

SP - 135

EP - 142

JO - IEEE Transactions on Semiconductor Manufacturing

JF - IEEE Transactions on Semiconductor Manufacturing

SN - 0894-6507

IS - 2

ER -