Detection and classification of defect patterns on semiconductor wafers

Chih Hsuan Wang, Way Kuo, Halima Bensmail

Research output: Contribution to journalArticle

41 Citations (Scopus)

Abstract

The detection of process problems and parameter drift at an early stage is crucial to successful semiconductor manufacture. The defect patterns on the wafer can act as an important source of information for quality engineers allowing them to isolate production problems. Traditionally, defect recognition is performed by quality engineers using a scanning electron microscope. This manual approach is not only expensive and time consuming but also it leads to high misidentification levels. In this paper, an automatic approach consisting of a spatial filter, a classification module and an estimation module is proposed to validate both real and simulated data. Experimental results show that three types of typical defect patterns: (i) a linear scratch; (ii) a circular ring; and (iii) an elliptical zone can be successfully extracted and classified. A Gaussian EM algorithm is used to estimate the elliptic and linear patterns, and a spherical-shell algorithm is used to estimate ring patterns. Furthermore, both convex and nonconvex defect patterns can be simultaneously recognized via a hybrid clustering method. The proposed method has the potential to be applied to other industries.

Original languageEnglish
Pages (from-to)1059-1068
Number of pages10
JournalIIE Transactions (Institute of Industrial Engineers)
Volume38
Issue number12
DOIs
Publication statusPublished - 1 Dec 2006
Externally publishedYes

Fingerprint

Semiconductor materials
Defects
Engineers
Electron microscopes
Scanning
Semiconductors
Industry
Module

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering
  • Management Science and Operations Research

Cite this

Detection and classification of defect patterns on semiconductor wafers. / Wang, Chih Hsuan; Kuo, Way; Bensmail, Halima.

In: IIE Transactions (Institute of Industrial Engineers), Vol. 38, No. 12, 01.12.2006, p. 1059-1068.

Research output: Contribution to journalArticle

@article{2996666cdf4146c8ba114b09b8fb6ae1,
title = "Detection and classification of defect patterns on semiconductor wafers",
abstract = "The detection of process problems and parameter drift at an early stage is crucial to successful semiconductor manufacture. The defect patterns on the wafer can act as an important source of information for quality engineers allowing them to isolate production problems. Traditionally, defect recognition is performed by quality engineers using a scanning electron microscope. This manual approach is not only expensive and time consuming but also it leads to high misidentification levels. In this paper, an automatic approach consisting of a spatial filter, a classification module and an estimation module is proposed to validate both real and simulated data. Experimental results show that three types of typical defect patterns: (i) a linear scratch; (ii) a circular ring; and (iii) an elliptical zone can be successfully extracted and classified. A Gaussian EM algorithm is used to estimate the elliptic and linear patterns, and a spherical-shell algorithm is used to estimate ring patterns. Furthermore, both convex and nonconvex defect patterns can be simultaneously recognized via a hybrid clustering method. The proposed method has the potential to be applied to other industries.",
author = "Wang, {Chih Hsuan} and Way Kuo and Halima Bensmail",
year = "2006",
month = "12",
day = "1",
doi = "10.1080/07408170600733236",
language = "English",
volume = "38",
pages = "1059--1068",
journal = "IISE Transactions",
issn = "2472-5854",
publisher = "Taylor and Francis Ltd.",
number = "12",

}

TY - JOUR

T1 - Detection and classification of defect patterns on semiconductor wafers

AU - Wang, Chih Hsuan

AU - Kuo, Way

AU - Bensmail, Halima

PY - 2006/12/1

Y1 - 2006/12/1

N2 - The detection of process problems and parameter drift at an early stage is crucial to successful semiconductor manufacture. The defect patterns on the wafer can act as an important source of information for quality engineers allowing them to isolate production problems. Traditionally, defect recognition is performed by quality engineers using a scanning electron microscope. This manual approach is not only expensive and time consuming but also it leads to high misidentification levels. In this paper, an automatic approach consisting of a spatial filter, a classification module and an estimation module is proposed to validate both real and simulated data. Experimental results show that three types of typical defect patterns: (i) a linear scratch; (ii) a circular ring; and (iii) an elliptical zone can be successfully extracted and classified. A Gaussian EM algorithm is used to estimate the elliptic and linear patterns, and a spherical-shell algorithm is used to estimate ring patterns. Furthermore, both convex and nonconvex defect patterns can be simultaneously recognized via a hybrid clustering method. The proposed method has the potential to be applied to other industries.

AB - The detection of process problems and parameter drift at an early stage is crucial to successful semiconductor manufacture. The defect patterns on the wafer can act as an important source of information for quality engineers allowing them to isolate production problems. Traditionally, defect recognition is performed by quality engineers using a scanning electron microscope. This manual approach is not only expensive and time consuming but also it leads to high misidentification levels. In this paper, an automatic approach consisting of a spatial filter, a classification module and an estimation module is proposed to validate both real and simulated data. Experimental results show that three types of typical defect patterns: (i) a linear scratch; (ii) a circular ring; and (iii) an elliptical zone can be successfully extracted and classified. A Gaussian EM algorithm is used to estimate the elliptic and linear patterns, and a spherical-shell algorithm is used to estimate ring patterns. Furthermore, both convex and nonconvex defect patterns can be simultaneously recognized via a hybrid clustering method. The proposed method has the potential to be applied to other industries.

UR - http://www.scopus.com/inward/record.url?scp=33750132385&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33750132385&partnerID=8YFLogxK

U2 - 10.1080/07408170600733236

DO - 10.1080/07408170600733236

M3 - Article

VL - 38

SP - 1059

EP - 1068

JO - IISE Transactions

JF - IISE Transactions

SN - 2472-5854

IS - 12

ER -