Denoised Residual Trace Analysis for Monitoring Semiconductor Process Faults

Research output: Contribution to journalArticle

Abstract

The detection of wafer faults in early process steps through monitoring and analyzing multivariate process trace data contribute to wafer yield improvements. Standard classification algorithms have been generally used for fault detection and classification (FDC). However, this approach can cause information loss while extracting statistical features from the trace data and cannot consider class imbalance situations where much fewer faulty wafers are generated than normal wafers. In addition, the approach does not consider normal wafer-to-wafer (W2W) variations and sensor noise inherent in the trace data. These drawbacks significantly degrade FDC performance. This paper proposes a method that builds an FDC model only with trace data of normal wafers in which W2W variations and sensor noise exist. The one-class FDC method detects the occurrence of abnormal trace patterns that cause wafer faults by removing W2W variations and sensor noise from raw traces by using denoising autoencoders, and this method finds the fault-introducing process parameters with the occurrence times. In experiments using the trace data of etch and chemical vapor deposition processes, the proposed method exhibited 1% and 6% higher performance than the best-performing method among comparison methods in terms of the geometric mean of the normal and fault detection accuracies.

Original languageEnglish
Article number8713521
Pages (from-to)293-301
Number of pages9
JournalIEEE Transactions on Semiconductor Manufacturing
Volume32
Issue number3
DOIs
Publication statusPublished - 2019 Aug

Fingerprint

Trace analysis
Fault detection
fault detection
wafers
Semiconductor materials
Monitoring
Sensors
sensors
occurrences
Chemical vapor deposition
causes
vapor deposition
Experiments

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{37a4e4f0b43d4cffb3eb27070e7fcab7,
title = "Denoised Residual Trace Analysis for Monitoring Semiconductor Process Faults",
abstract = "The detection of wafer faults in early process steps through monitoring and analyzing multivariate process trace data contribute to wafer yield improvements. Standard classification algorithms have been generally used for fault detection and classification (FDC). However, this approach can cause information loss while extracting statistical features from the trace data and cannot consider class imbalance situations where much fewer faulty wafers are generated than normal wafers. In addition, the approach does not consider normal wafer-to-wafer (W2W) variations and sensor noise inherent in the trace data. These drawbacks significantly degrade FDC performance. This paper proposes a method that builds an FDC model only with trace data of normal wafers in which W2W variations and sensor noise exist. The one-class FDC method detects the occurrence of abnormal trace patterns that cause wafer faults by removing W2W variations and sensor noise from raw traces by using denoising autoencoders, and this method finds the fault-introducing process parameters with the occurrence times. In experiments using the trace data of etch and chemical vapor deposition processes, the proposed method exhibited 1{\%} and 6{\%} higher performance than the best-performing method among comparison methods in terms of the geometric mean of the normal and fault detection accuracies.",
author = "Jaeyeon Jang and Min, {Byung Wook} and Kim, {Chang Ouk}",
year = "2019",
month = "8",
doi = "10.1109/TSM.2019.2916374",
language = "English",
volume = "32",
pages = "293--301",
journal = "IEEE Transactions on Semiconductor Manufacturing",
issn = "0894-6507",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "3",

}

Denoised Residual Trace Analysis for Monitoring Semiconductor Process Faults. / Jang, Jaeyeon; Min, Byung Wook; Kim, Chang Ouk.

In: IEEE Transactions on Semiconductor Manufacturing, Vol. 32, No. 3, 8713521, 08.2019, p. 293-301.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Denoised Residual Trace Analysis for Monitoring Semiconductor Process Faults

AU - Jang, Jaeyeon

AU - Min, Byung Wook

AU - Kim, Chang Ouk

PY - 2019/8

Y1 - 2019/8

N2 - The detection of wafer faults in early process steps through monitoring and analyzing multivariate process trace data contribute to wafer yield improvements. Standard classification algorithms have been generally used for fault detection and classification (FDC). However, this approach can cause information loss while extracting statistical features from the trace data and cannot consider class imbalance situations where much fewer faulty wafers are generated than normal wafers. In addition, the approach does not consider normal wafer-to-wafer (W2W) variations and sensor noise inherent in the trace data. These drawbacks significantly degrade FDC performance. This paper proposes a method that builds an FDC model only with trace data of normal wafers in which W2W variations and sensor noise exist. The one-class FDC method detects the occurrence of abnormal trace patterns that cause wafer faults by removing W2W variations and sensor noise from raw traces by using denoising autoencoders, and this method finds the fault-introducing process parameters with the occurrence times. In experiments using the trace data of etch and chemical vapor deposition processes, the proposed method exhibited 1% and 6% higher performance than the best-performing method among comparison methods in terms of the geometric mean of the normal and fault detection accuracies.

AB - The detection of wafer faults in early process steps through monitoring and analyzing multivariate process trace data contribute to wafer yield improvements. Standard classification algorithms have been generally used for fault detection and classification (FDC). However, this approach can cause information loss while extracting statistical features from the trace data and cannot consider class imbalance situations where much fewer faulty wafers are generated than normal wafers. In addition, the approach does not consider normal wafer-to-wafer (W2W) variations and sensor noise inherent in the trace data. These drawbacks significantly degrade FDC performance. This paper proposes a method that builds an FDC model only with trace data of normal wafers in which W2W variations and sensor noise exist. The one-class FDC method detects the occurrence of abnormal trace patterns that cause wafer faults by removing W2W variations and sensor noise from raw traces by using denoising autoencoders, and this method finds the fault-introducing process parameters with the occurrence times. In experiments using the trace data of etch and chemical vapor deposition processes, the proposed method exhibited 1% and 6% higher performance than the best-performing method among comparison methods in terms of the geometric mean of the normal and fault detection accuracies.

UR - http://www.scopus.com/inward/record.url?scp=85069768450&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069768450&partnerID=8YFLogxK

U2 - 10.1109/TSM.2019.2916374

DO - 10.1109/TSM.2019.2916374

M3 - Article

AN - SCOPUS:85069768450

VL - 32

SP - 293

EP - 301

JO - IEEE Transactions on Semiconductor Manufacturing

JF - IEEE Transactions on Semiconductor Manufacturing

SN - 0894-6507

IS - 3

M1 - 8713521

ER -