Process monitoring using variational autoencoder for high-dimensional nonlinear processes

Seulki Lee, Mingu Kwak, Kwok Leung Tsui, Seoung Bum Kim

Research output: Contribution to journalArticle

Abstract

In many industries, statistical process monitoring techniques play a key role in improving processes through variation reduction and defect prevention. Modern large-scale industrial processes require appropriate monitoring techniques that can efficiently address high-dimensional nonlinear processes. Such processes have been successfully monitored with several latent variable-based methods. However, because these monitoring methods use Hotelling's T 2 statistics in the reduced space, a normality assumption underlies the construction of these tools. This assumption has limited the use of latent variable-based monitoring charts in both nonlinear and nonnormal situations. In this study, we propose a variational autoencoder (VAE) as a monitoring method that can address both nonlinear and nonnormal situations in high-dimensional processes. VAE is appropriate for T 2 charts because it causes the reduced space to follow a multivariate normal distribution. The effectiveness and applicability of the proposed VAE-based chart were demonstrated through experiments on simulated data and real data from a thin-film-transistor liquid-crystal display process.

Original languageEnglish
Pages (from-to)13-27
Number of pages15
JournalEngineering Applications of Artificial Intelligence
Volume83
DOIs
Publication statusPublished - 2019 Aug 1

Fingerprint

Process monitoring
Monitoring
Normal distribution
Thin film transistors
Liquid crystal displays
Statistics
Defects
Industry
Experiments

Keywords

  • High-dimensional process
  • Multivariate control chart
  • Nonlinear process
  • Statistical process monitoring
  • Variational autoencoder

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Artificial Intelligence
  • Electrical and Electronic Engineering

Cite this

Process monitoring using variational autoencoder for high-dimensional nonlinear processes. / Lee, Seulki; Kwak, Mingu; Tsui, Kwok Leung; Kim, Seoung Bum.

In: Engineering Applications of Artificial Intelligence, Vol. 83, 01.08.2019, p. 13-27.

Research output: Contribution to journalArticle

@article{bb9f5ba180ec430390b8eba22d5d1138,
title = "Process monitoring using variational autoencoder for high-dimensional nonlinear processes",
abstract = "In many industries, statistical process monitoring techniques play a key role in improving processes through variation reduction and defect prevention. Modern large-scale industrial processes require appropriate monitoring techniques that can efficiently address high-dimensional nonlinear processes. Such processes have been successfully monitored with several latent variable-based methods. However, because these monitoring methods use Hotelling's T 2 statistics in the reduced space, a normality assumption underlies the construction of these tools. This assumption has limited the use of latent variable-based monitoring charts in both nonlinear and nonnormal situations. In this study, we propose a variational autoencoder (VAE) as a monitoring method that can address both nonlinear and nonnormal situations in high-dimensional processes. VAE is appropriate for T 2 charts because it causes the reduced space to follow a multivariate normal distribution. The effectiveness and applicability of the proposed VAE-based chart were demonstrated through experiments on simulated data and real data from a thin-film-transistor liquid-crystal display process.",
keywords = "High-dimensional process, Multivariate control chart, Nonlinear process, Statistical process monitoring, Variational autoencoder",
author = "Seulki Lee and Mingu Kwak and Tsui, {Kwok Leung} and Kim, {Seoung Bum}",
year = "2019",
month = "8",
day = "1",
doi = "10.1016/j.engappai.2019.04.013",
language = "English",
volume = "83",
pages = "13--27",
journal = "Engineering Applications of Artificial Intelligence",
issn = "0952-1976",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Process monitoring using variational autoencoder for high-dimensional nonlinear processes

AU - Lee, Seulki

AU - Kwak, Mingu

AU - Tsui, Kwok Leung

AU - Kim, Seoung Bum

PY - 2019/8/1

Y1 - 2019/8/1

N2 - In many industries, statistical process monitoring techniques play a key role in improving processes through variation reduction and defect prevention. Modern large-scale industrial processes require appropriate monitoring techniques that can efficiently address high-dimensional nonlinear processes. Such processes have been successfully monitored with several latent variable-based methods. However, because these monitoring methods use Hotelling's T 2 statistics in the reduced space, a normality assumption underlies the construction of these tools. This assumption has limited the use of latent variable-based monitoring charts in both nonlinear and nonnormal situations. In this study, we propose a variational autoencoder (VAE) as a monitoring method that can address both nonlinear and nonnormal situations in high-dimensional processes. VAE is appropriate for T 2 charts because it causes the reduced space to follow a multivariate normal distribution. The effectiveness and applicability of the proposed VAE-based chart were demonstrated through experiments on simulated data and real data from a thin-film-transistor liquid-crystal display process.

AB - In many industries, statistical process monitoring techniques play a key role in improving processes through variation reduction and defect prevention. Modern large-scale industrial processes require appropriate monitoring techniques that can efficiently address high-dimensional nonlinear processes. Such processes have been successfully monitored with several latent variable-based methods. However, because these monitoring methods use Hotelling's T 2 statistics in the reduced space, a normality assumption underlies the construction of these tools. This assumption has limited the use of latent variable-based monitoring charts in both nonlinear and nonnormal situations. In this study, we propose a variational autoencoder (VAE) as a monitoring method that can address both nonlinear and nonnormal situations in high-dimensional processes. VAE is appropriate for T 2 charts because it causes the reduced space to follow a multivariate normal distribution. The effectiveness and applicability of the proposed VAE-based chart were demonstrated through experiments on simulated data and real data from a thin-film-transistor liquid-crystal display process.

KW - High-dimensional process

KW - Multivariate control chart

KW - Nonlinear process

KW - Statistical process monitoring

KW - Variational autoencoder

UR - http://www.scopus.com/inward/record.url?scp=85065604788&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065604788&partnerID=8YFLogxK

U2 - 10.1016/j.engappai.2019.04.013

DO - 10.1016/j.engappai.2019.04.013

M3 - Article

VL - 83

SP - 13

EP - 27

JO - Engineering Applications of Artificial Intelligence

JF - Engineering Applications of Artificial Intelligence

SN - 0952-1976

ER -