Bin2Vec: A better wafer bin map coloring scheme for comprehensible visualization and effective bad wafer classification

Junhong Kim, Hyungseok Kim, Jaesun Park, Kyounghyun Mo, Pilsung Kang

Research output: Contribution to journalArticle

Abstract

A wafer bin map (WBM), which is the result of an electrical die-sorting test, provides information on which bins failed what tests, and plays an important role in finding defective wafer patterns in semiconductor manufacturing. Current wafer inspection based on WBM has two problems: good/bad WBM classification is performed by engineers and the bin code coloring scheme does not reflect the relationship between bin codes. To solve these problems, we propose a neural network-based bin coloring method called Bin2Vec to make similar bin codes are represented by similar colors. We also build a convolutional neural network-based WBM classification model to reduce the variations in the decisions made by engineers with different expertise by learning the company-wide historical WBM classification results. Based on a real dataset with a total of 27,701 WBMs, our WBM classification model significantly outperformed benchmarked machine learning models. In addition, the visualization results of the proposed Bin2Vec method makes it easier to discover meaningful WBM patterns compared with the random RGB coloring scheme. We expect the proposed framework to improve both efficiencies by automating the bad wafer classification process and effectiveness by assigning similar bin codes and their corresponding colors on the WBM.

Original languageEnglish
Article number597
JournalApplied Sciences (Switzerland)
Volume9
Issue number3
DOIs
Publication statusPublished - 2019 Feb 11

Fingerprint

Bins
Coloring
Visualization
wafers
engineers
color
machine learning
Color
Neural networks
Engineers
classifying
learning
inspection
manufacturing
Sorting
Learning systems
Inspection

Keywords

  • Bad wafer classification
  • Bin2Vec
  • Convolution neural network
  • Wafer bin map (WBM)
  • Word2Vec

ASJC Scopus subject areas

  • Materials Science(all)
  • Instrumentation
  • Engineering(all)
  • Process Chemistry and Technology
  • Computer Science Applications
  • Fluid Flow and Transfer Processes

Cite this

Bin2Vec : A better wafer bin map coloring scheme for comprehensible visualization and effective bad wafer classification. / Kim, Junhong; Kim, Hyungseok; Park, Jaesun; Mo, Kyounghyun; Kang, Pilsung.

In: Applied Sciences (Switzerland), Vol. 9, No. 3, 597, 11.02.2019.

Research output: Contribution to journalArticle

@article{8bd7c5a9503b43d99057c70e23248f65,
title = "Bin2Vec: A better wafer bin map coloring scheme for comprehensible visualization and effective bad wafer classification",
abstract = "A wafer bin map (WBM), which is the result of an electrical die-sorting test, provides information on which bins failed what tests, and plays an important role in finding defective wafer patterns in semiconductor manufacturing. Current wafer inspection based on WBM has two problems: good/bad WBM classification is performed by engineers and the bin code coloring scheme does not reflect the relationship between bin codes. To solve these problems, we propose a neural network-based bin coloring method called Bin2Vec to make similar bin codes are represented by similar colors. We also build a convolutional neural network-based WBM classification model to reduce the variations in the decisions made by engineers with different expertise by learning the company-wide historical WBM classification results. Based on a real dataset with a total of 27,701 WBMs, our WBM classification model significantly outperformed benchmarked machine learning models. In addition, the visualization results of the proposed Bin2Vec method makes it easier to discover meaningful WBM patterns compared with the random RGB coloring scheme. We expect the proposed framework to improve both efficiencies by automating the bad wafer classification process and effectiveness by assigning similar bin codes and their corresponding colors on the WBM.",
keywords = "Bad wafer classification, Bin2Vec, Convolution neural network, Wafer bin map (WBM), Word2Vec",
author = "Junhong Kim and Hyungseok Kim and Jaesun Park and Kyounghyun Mo and Pilsung Kang",
year = "2019",
month = "2",
day = "11",
doi = "10.3390/app9030597",
language = "English",
volume = "9",
journal = "Applied Sciences (Switzerland)",
issn = "2076-3417",
publisher = "Multidisciplinary Digital Publishing Institute",
number = "3",

}

TY - JOUR

T1 - Bin2Vec

T2 - A better wafer bin map coloring scheme for comprehensible visualization and effective bad wafer classification

AU - Kim, Junhong

AU - Kim, Hyungseok

AU - Park, Jaesun

AU - Mo, Kyounghyun

AU - Kang, Pilsung

PY - 2019/2/11

Y1 - 2019/2/11

N2 - A wafer bin map (WBM), which is the result of an electrical die-sorting test, provides information on which bins failed what tests, and plays an important role in finding defective wafer patterns in semiconductor manufacturing. Current wafer inspection based on WBM has two problems: good/bad WBM classification is performed by engineers and the bin code coloring scheme does not reflect the relationship between bin codes. To solve these problems, we propose a neural network-based bin coloring method called Bin2Vec to make similar bin codes are represented by similar colors. We also build a convolutional neural network-based WBM classification model to reduce the variations in the decisions made by engineers with different expertise by learning the company-wide historical WBM classification results. Based on a real dataset with a total of 27,701 WBMs, our WBM classification model significantly outperformed benchmarked machine learning models. In addition, the visualization results of the proposed Bin2Vec method makes it easier to discover meaningful WBM patterns compared with the random RGB coloring scheme. We expect the proposed framework to improve both efficiencies by automating the bad wafer classification process and effectiveness by assigning similar bin codes and their corresponding colors on the WBM.

AB - A wafer bin map (WBM), which is the result of an electrical die-sorting test, provides information on which bins failed what tests, and plays an important role in finding defective wafer patterns in semiconductor manufacturing. Current wafer inspection based on WBM has two problems: good/bad WBM classification is performed by engineers and the bin code coloring scheme does not reflect the relationship between bin codes. To solve these problems, we propose a neural network-based bin coloring method called Bin2Vec to make similar bin codes are represented by similar colors. We also build a convolutional neural network-based WBM classification model to reduce the variations in the decisions made by engineers with different expertise by learning the company-wide historical WBM classification results. Based on a real dataset with a total of 27,701 WBMs, our WBM classification model significantly outperformed benchmarked machine learning models. In addition, the visualization results of the proposed Bin2Vec method makes it easier to discover meaningful WBM patterns compared with the random RGB coloring scheme. We expect the proposed framework to improve both efficiencies by automating the bad wafer classification process and effectiveness by assigning similar bin codes and their corresponding colors on the WBM.

KW - Bad wafer classification

KW - Bin2Vec

KW - Convolution neural network

KW - Wafer bin map (WBM)

KW - Word2Vec

UR - http://www.scopus.com/inward/record.url?scp=85061355773&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061355773&partnerID=8YFLogxK

U2 - 10.3390/app9030597

DO - 10.3390/app9030597

M3 - Article

AN - SCOPUS:85061355773

VL - 9

JO - Applied Sciences (Switzerland)

JF - Applied Sciences (Switzerland)

SN - 2076-3417

IS - 3

M1 - 597

ER -