SlideShare a Scribd company logo
Copyright © JMP Statistical Discovery LLC. All Rights Reserved.
KOREA
Semiconductor Wafer Test
Fail Map Clustering and Data Mining
using JMP Add-In
곽병훈, SK하이닉스
✓ Semiconductor Wafer Test Failure Map Clustering and
Data Mining using JMP Add-In
1. Semiconductor Process
2. Good, Bad Grouping in conventional way
3. User Friendly Interface
4. Data Mining Process & Result
5. Example
6. Summary
Over 99% of the defects in wafer test are within fabrication
Wafer Test
Failure
Good, Bad grouping in Convention way (1)
Failure
Rate
(%)
Wafer Test Time
>10.0%
↑BAD
↓GOOD
<1.00%
✓ Final Data Mining
→ Good Gr. vs Bad Gr.
✓ Problem
→ Garbage in, Garbage out
Various Maps @Bad Gr.
Blue = Fail
Orange = Pass
This is the type
of wafer map
engineers want
to analyze.
* Goal
Find the real cause
✓ .
✓ .
✓ .
✓ .
✓ .
but,
Garbage in, Garbage out
There are various types
of failure map including
the map we are finding
→ Garbage in
Good, Bad grouping in Convention way (2)
A large loss of time = Inefficient
✓ Advantage
: an accurate classification
✓ Disadvantage
: Time Consumption
✓ Engineer finds and classifies map.
We suggest the solutions with this
problems in this presentation.
User Friendly Interface
* Clustering
-. sda symbol
-. EPM
-. RPM
-. WT Item (Fail Bit Count)
*sda = statistical defect analysis
User Friendly Interface
Set period
Select Fab.
Select product
Select defect
Select Failure
to analyze
Clustering
Mining
* Clustering
-. sda symbol **
-. EPM
-. RPM **
-. WT Item (Fail Bit Count)
**sda = statistical defect analysis
**RPM = Reliability Parameter Monitoring
* Clustering selection → Set period → Select Fab. → Select product → Cluster Analysis Click!! → Select defect category → Select Failure Maps to analyze
Bad
Good Good Exclude Exclude
Exclude Exclude
User Friendly Interface
Bad
Good Good Exclude Exclude
Exclude Exclude
Bad Gr.
Lot & Wafer List ✓ Good, Bad Portion
Lot and Wafer
Lot and Wafer
✓ Clustering Group Portion
Clustering
Group
Portion
Good,
Bad
Portion
Fail
Rate
Fail
Rate
Fail
Rate
Fail
Rate
Fail
Rate
Fail
Rate
Fail
Rate
Fail
Rate
Fail
Rate
Fail
Rate
Fail
Rate
Fail
Rate
Fail
Rate
Fail
Rate
✓ Good, Bad group fail rate comparison
Classification using Hierarchical Clustering
Set period
Select Fab.
Select product
Select defect
Select Fail
to analyze
Good Good Exclude Exclude Exclude
Exclude
Exclude Bad
Bad
Bad
Bad
Bad Bad Bad
Good
✓ Clustering Group
: Engineers can select certain Area of a Wafer and
classify that area as a good group or bad group
Good Gr. Bad Gr.
VS
✓ Select Conditions ✓ Final Mining Group
* Good, Bad and Exclude Multiple choices possible
✓ Designating the good or bad groups
directly by looking at the clustered map
group
✓ Mining with final Good, Bad Groups
✓ Classifying failure maps in accurate and
fast.
➔ Dramatically improve engineer time loss
✓ Mining (Inline Legend)
Select Oper. Select Legend.
Data Mining Process Interface
* Select Inline → Select Start Data, End Data → Select Oper. → Select Legend. → Click “연계분석”
Data Mining Result Interface
Rank Summary by
Algorithm Mining Result
Cause Process
Ranking
Various mining algorithms
OOOO Step Legend
OOOO Step Legend
OOOO Step Legend
* Algorithms
-. Decision Tree
-. Randomforest
-. ANOVA
✓ Mining (EPM)
Data Mining Process Interface
aaa para.
aaa
para.
* Select EPM → Select Start Data, End Data → Click “연계분석”
Clustering & Data Mining Example (1)
✓ Select Conditions
Good Good Exclude Bad Bad
Bad
✓ Clustering Result ✓ Mining Result
Cause Process Rank Summary by Algorithm
OOOO Step Legend
OOOO Step Legend
OOOO Step Legend
OOOO Step Legend
OOOO Step Legend
OOOO Step Legend
A
EQ.
B
EQ.
C
EQ.
D
EQ.
D
EQ.
C
EQ.
B
EQ.
A
EQ.
OOOO Step Process Run Time
✓ Result
Click!!
Focused on specific EQ (D EQ) and process run time.
→ These could be most likely the causes.
Clustering & Data Mining Example (2)
✓ Mining Result (Data Visualization : Parallel Plot)
OOO
Step
OOO
Step
OOO
Step
OOO
Step
OOO
Step
OOO
Step
OOO
Step
OOO
Step
OOO
Step
OOO
Step
OOO
Step
OOO
Step
OOO
Step
By
Process
Step
Legend
Good group scattered.
Bad group scattered.
The path of Bad group overlapped with one equipment. →This could be most likely the cause.
a EQ
b EQ
c EQ
d EQ
* Association Analysis is also possible.
Data Visualization Analysis
Clustering & Data Mining Example (3)
○ Delayed step process detection between Good and Bad group using multivariate.
Dual axis occurrence
Good
Bad
Shift
Bad Gr.
Good Gr.
A Step
Process run time
B Step
Process run time
C Step
Process run time
D Step
Process run time
Process
Flow
Process
Flow
Good
Gr.
Bad
Gr.
+1 Day +1 Day +1 Day
+1 Day +8 Day +1 Day
↙Cause of defect
A Step
Process run time
B Step
Process run time
C Step
Process run time
D Step
Process run time
A
Step
Process
run
time
B Step
Process run time
A
Step
Process
run
time
C Step
Process run time
D Step
Process run time
A
Step
Process
run
time
A
Step
Process
run
time
A
Step
Process
run
time
B Step
Process run time
B Step
Process run time
Process run time
Process
run
time
○ Delay step detection between Good and Bad group using multivariate.
If there is a problem with
delay in B step
Data Visualization Analysis
WF Count Human [min] AI [min]
100 6 0.03
1000 60 0.16
3000 180 5
Clustering Process Time [ Human vs AI ]
Human
[min]
AI
[min]
AI
180min
5min
Human
Wafer Count
✓ Dramatically reduces time consumption for engineer
Summary
1. Classifying failure maps in faster and more accurate way
2. User Friendly Interface.
3. Reducing engineer’s time spends on repetitive work tasks dramatically
4. Anyone can use JMP Add-In.
Copyright © JMP Statistical Discovery LLC. All rights reserved.
KOREA
Thank you

More Related Content

Similar to 3-1: Semiconductor Wafer Test Fail Map Clustering and Data Mining using JMP Add-In

TCO19 Japan Introduction to Marathon Matches
TCO19 Japan Introduction to Marathon MatchesTCO19 Japan Introduction to Marathon Matches
TCO19 Japan Introduction to Marathon Matches
tomerun
 
Journey of Migrating Millions of Queries on The Cloud
Journey of Migrating Millions of Queries on The CloudJourney of Migrating Millions of Queries on The Cloud
Journey of Migrating Millions of Queries on The Cloud
takezoe
 
Andrew rusling 21 experiments to increase velocity
Andrew rusling 21 experiments to increase velocityAndrew rusling 21 experiments to increase velocity
Andrew rusling 21 experiments to increase velocity
Scrum Australia Pty Ltd
 
21 experiments to increase velocity
21 experiments to increase velocity21 experiments to increase velocity
21 experiments to increase velocity
Andrew Rusling
 
Demographics andweblogtargeting
Demographics andweblogtargetingDemographics andweblogtargeting
Demographics andweblogtargeting
Doug Chang
 
GC FAQ
GC FAQGC FAQ
Managing Uncertainty in Value-based SE
Managing Uncertainty in Value-based SEManaging Uncertainty in Value-based SE
Managing Uncertainty in Value-based SE
CS, NcState
 
Basic Design of Experiments Using the Custom DOE Platform
Basic Design of Experiments Using the Custom DOE PlatformBasic Design of Experiments Using the Custom DOE Platform
Basic Design of Experiments Using the Custom DOE Platform
JMP software from SAS
 
An overview of gradient descent optimization algorithms.pdf
An overview of gradient descent optimization algorithms.pdfAn overview of gradient descent optimization algorithms.pdf
An overview of gradient descent optimization algorithms.pdf
vudinhphuong96
 
Improving throughput with the Theory of Constraints and Queuing Theory
Improving throughput with the Theory of Constraints and Queuing TheoryImproving throughput with the Theory of Constraints and Queuing Theory
Improving throughput with the Theory of Constraints and Queuing Theory
Andrew Rusling
 
Rubykaigi2010mrkn bigdecimal
Rubykaigi2010mrkn bigdecimalRubykaigi2010mrkn bigdecimal
Rubykaigi2010mrkn bigdecimal
Kenta Murata
 
SSD Failures in the Field.pptx
SSD Failures in the Field.pptxSSD Failures in the Field.pptx
SSD Failures in the Field.pptx
Po-Chuan Chen
 
2014 toronto-torbug
2014 toronto-torbug2014 toronto-torbug
2014 toronto-torbug
c.titus.brown
 
21 Experiments to Increase Throughput
21 Experiments to Increase Throughput21 Experiments to Increase Throughput
21 Experiments to Increase Throughput
Andrew Rusling
 
Vladimir Primakov - Qa management in big agile teams
Vladimir Primakov - Qa management in big agile teamsVladimir Primakov - Qa management in big agile teams
Vladimir Primakov - Qa management in big agile teams
Ievgenii Katsan
 
The anatomy of a cascading failure
The anatomy of a cascading failureThe anatomy of a cascading failure
The anatomy of a cascading failure
Rares Musina
 
SLO DRIVEN DEVELOPMENT, ALON NATIV, Tomorrow.io
SLO DRIVEN DEVELOPMENT, ALON NATIV, Tomorrow.ioSLO DRIVEN DEVELOPMENT, ALON NATIV, Tomorrow.io
SLO DRIVEN DEVELOPMENT, ALON NATIV, Tomorrow.io
DevOpsDays Tel Aviv
 
Road to A/B testing - Alexey Vasiliev (ENG) | Ruby Meditation 25
Road to A/B testing - Alexey Vasiliev (ENG) | Ruby Meditation 25Road to A/B testing - Alexey Vasiliev (ENG) | Ruby Meditation 25
Road to A/B testing - Alexey Vasiliev (ENG) | Ruby Meditation 25
Ruby Meditation
 
Tuhin Mitra: How I Automate My Negative Tests
Tuhin Mitra: How I Automate My Negative TestsTuhin Mitra: How I Automate My Negative Tests
Tuhin Mitra: How I Automate My Negative Tests
Anna Royzman
 
Designing and Running Performance Experiments
Designing and Running Performance ExperimentsDesigning and Running Performance Experiments
Designing and Running Performance Experiments
J On The Beach
 

Similar to 3-1: Semiconductor Wafer Test Fail Map Clustering and Data Mining using JMP Add-In (20)

TCO19 Japan Introduction to Marathon Matches
TCO19 Japan Introduction to Marathon MatchesTCO19 Japan Introduction to Marathon Matches
TCO19 Japan Introduction to Marathon Matches
 
Journey of Migrating Millions of Queries on The Cloud
Journey of Migrating Millions of Queries on The CloudJourney of Migrating Millions of Queries on The Cloud
Journey of Migrating Millions of Queries on The Cloud
 
Andrew rusling 21 experiments to increase velocity
Andrew rusling 21 experiments to increase velocityAndrew rusling 21 experiments to increase velocity
Andrew rusling 21 experiments to increase velocity
 
21 experiments to increase velocity
21 experiments to increase velocity21 experiments to increase velocity
21 experiments to increase velocity
 
Demographics andweblogtargeting
Demographics andweblogtargetingDemographics andweblogtargeting
Demographics andweblogtargeting
 
GC FAQ
GC FAQGC FAQ
GC FAQ
 
Managing Uncertainty in Value-based SE
Managing Uncertainty in Value-based SEManaging Uncertainty in Value-based SE
Managing Uncertainty in Value-based SE
 
Basic Design of Experiments Using the Custom DOE Platform
Basic Design of Experiments Using the Custom DOE PlatformBasic Design of Experiments Using the Custom DOE Platform
Basic Design of Experiments Using the Custom DOE Platform
 
An overview of gradient descent optimization algorithms.pdf
An overview of gradient descent optimization algorithms.pdfAn overview of gradient descent optimization algorithms.pdf
An overview of gradient descent optimization algorithms.pdf
 
Improving throughput with the Theory of Constraints and Queuing Theory
Improving throughput with the Theory of Constraints and Queuing TheoryImproving throughput with the Theory of Constraints and Queuing Theory
Improving throughput with the Theory of Constraints and Queuing Theory
 
Rubykaigi2010mrkn bigdecimal
Rubykaigi2010mrkn bigdecimalRubykaigi2010mrkn bigdecimal
Rubykaigi2010mrkn bigdecimal
 
SSD Failures in the Field.pptx
SSD Failures in the Field.pptxSSD Failures in the Field.pptx
SSD Failures in the Field.pptx
 
2014 toronto-torbug
2014 toronto-torbug2014 toronto-torbug
2014 toronto-torbug
 
21 Experiments to Increase Throughput
21 Experiments to Increase Throughput21 Experiments to Increase Throughput
21 Experiments to Increase Throughput
 
Vladimir Primakov - Qa management in big agile teams
Vladimir Primakov - Qa management in big agile teamsVladimir Primakov - Qa management in big agile teams
Vladimir Primakov - Qa management in big agile teams
 
The anatomy of a cascading failure
The anatomy of a cascading failureThe anatomy of a cascading failure
The anatomy of a cascading failure
 
SLO DRIVEN DEVELOPMENT, ALON NATIV, Tomorrow.io
SLO DRIVEN DEVELOPMENT, ALON NATIV, Tomorrow.ioSLO DRIVEN DEVELOPMENT, ALON NATIV, Tomorrow.io
SLO DRIVEN DEVELOPMENT, ALON NATIV, Tomorrow.io
 
Road to A/B testing - Alexey Vasiliev (ENG) | Ruby Meditation 25
Road to A/B testing - Alexey Vasiliev (ENG) | Ruby Meditation 25Road to A/B testing - Alexey Vasiliev (ENG) | Ruby Meditation 25
Road to A/B testing - Alexey Vasiliev (ENG) | Ruby Meditation 25
 
Tuhin Mitra: How I Automate My Negative Tests
Tuhin Mitra: How I Automate My Negative TestsTuhin Mitra: How I Automate My Negative Tests
Tuhin Mitra: How I Automate My Negative Tests
 
Designing and Running Performance Experiments
Designing and Running Performance ExperimentsDesigning and Running Performance Experiments
Designing and Running Performance Experiments
 

More from JMP Discovery Summit Korea 2023

Keynote 2: 엔지니어링 효율성 개선
Keynote 2: 엔지니어링 효율성 개선Keynote 2: 엔지니어링 효율성 개선
Keynote 2: 엔지니어링 효율성 개선
JMP Discovery Summit Korea 2023
 
Keynote 3: 실험 설계의 강화 - 단순함에서 정교함으로
Keynote 3: 실험 설계의 강화 - 단순함에서 정교함으로Keynote 3: 실험 설계의 강화 - 단순함에서 정교함으로
Keynote 3: 실험 설계의 강화 - 단순함에서 정교함으로
JMP Discovery Summit Korea 2023
 
Keynote 4: IES 통계 분석 능력의 강화 - JMP와 Python의 시너지
Keynote 4: IES 통계 분석 능력의 강화 - JMP와 Python의 시너지Keynote 4: IES 통계 분석 능력의 강화 - JMP와 Python의 시너지
Keynote 4: IES 통계 분석 능력의 강화 - JMP와 Python의 시너지
JMP Discovery Summit Korea 2023
 
1-1: 바이오 의약품 공정 개발 및 특성 분석을 위한 DOE 연구 사례
1-1: 바이오 의약품 공정 개발 및 특성 분석을 위한 DOE 연구 사례1-1: 바이오 의약품 공정 개발 및 특성 분석을 위한 DOE 연구 사례
1-1: 바이오 의약품 공정 개발 및 특성 분석을 위한 DOE 연구 사례
JMP Discovery Summit Korea 2023
 
1-2: QbD 기반 의약품 개발의 통계 분석 및 JMP 활용 사례
1-2: QbD 기반 의약품 개발의 통계 분석 및 JMP 활용 사례1-2: QbD 기반 의약품 개발의 통계 분석 및 JMP 활용 사례
1-2: QbD 기반 의약품 개발의 통계 분석 및 JMP 활용 사례
JMP Discovery Summit Korea 2023
 
2-1: 석유 화학 산업에서의 JMP 활용 사례
2-1: 석유 화학 산업에서의 JMP 활용 사례2-1: 석유 화학 산업에서의 JMP 활용 사례
2-1: 석유 화학 산업에서의 JMP 활용 사례
JMP Discovery Summit Korea 2023
 
2-2: 다양한 무고장 조건에서 필드 차량 클레임 분석 자동화
2-2: 다양한 무고장 조건에서 필드 차량 클레임 분석 자동화2-2: 다양한 무고장 조건에서 필드 차량 클레임 분석 자동화
2-2: 다양한 무고장 조건에서 필드 차량 클레임 분석 자동화
JMP Discovery Summit Korea 2023
 
2-3: 데이터 분석 기반의 B2B 영업
2-3: 데이터 분석 기반의 B2B 영업2-3: 데이터 분석 기반의 B2B 영업
2-3: 데이터 분석 기반의 B2B 영업
JMP Discovery Summit Korea 2023
 
3-2: Smart Proactive Analysis Material
3-2: Smart Proactive Analysis Material3-2: Smart Proactive Analysis Material
3-2: Smart Proactive Analysis Material
JMP Discovery Summit Korea 2023
 
3-3: 분석 콘텐츠 재사용성 향상을 위한 AppStore 생태계 구축
3-3: 분석 콘텐츠 재사용성 향상을 위한 AppStore 생태계 구축3-3: 분석 콘텐츠 재사용성 향상을 위한 AppStore 생태계 구축
3-3: 분석 콘텐츠 재사용성 향상을 위한 AppStore 생태계 구축
JMP Discovery Summit Korea 2023
 
4-1: JMP/Python 에코시스템 성공 사례
4-1: JMP/Python 에코시스템 성공 사례4-1: JMP/Python 에코시스템 성공 사례
4-1: JMP/Python 에코시스템 성공 사례
JMP Discovery Summit Korea 2023
 
4-2: 네거티브 디자인 개념을 활용한 군함 초기 설계에 적용되는 디자인 영역 탐색
4-2: 네거티브 디자인 개념을 활용한 군함 초기 설계에 적용되는 디자인 영역 탐색4-2: 네거티브 디자인 개념을 활용한 군함 초기 설계에 적용되는 디자인 영역 탐색
4-2: 네거티브 디자인 개념을 활용한 군함 초기 설계에 적용되는 디자인 영역 탐색
JMP Discovery Summit Korea 2023
 
4-3: 제조 계측 장비 비정형 데이터 파일 기반으로 데이터 변환 및 관리도 분석 서비스
4-3: 제조 계측 장비 비정형 데이터 파일 기반으로 데이터 변환 및 관리도 분석 서비스4-3: 제조 계측 장비 비정형 데이터 파일 기반으로 데이터 변환 및 관리도 분석 서비스
4-3: 제조 계측 장비 비정형 데이터 파일 기반으로 데이터 변환 및 관리도 분석 서비스
JMP Discovery Summit Korea 2023
 

More from JMP Discovery Summit Korea 2023 (13)

Keynote 2: 엔지니어링 효율성 개선
Keynote 2: 엔지니어링 효율성 개선Keynote 2: 엔지니어링 효율성 개선
Keynote 2: 엔지니어링 효율성 개선
 
Keynote 3: 실험 설계의 강화 - 단순함에서 정교함으로
Keynote 3: 실험 설계의 강화 - 단순함에서 정교함으로Keynote 3: 실험 설계의 강화 - 단순함에서 정교함으로
Keynote 3: 실험 설계의 강화 - 단순함에서 정교함으로
 
Keynote 4: IES 통계 분석 능력의 강화 - JMP와 Python의 시너지
Keynote 4: IES 통계 분석 능력의 강화 - JMP와 Python의 시너지Keynote 4: IES 통계 분석 능력의 강화 - JMP와 Python의 시너지
Keynote 4: IES 통계 분석 능력의 강화 - JMP와 Python의 시너지
 
1-1: 바이오 의약품 공정 개발 및 특성 분석을 위한 DOE 연구 사례
1-1: 바이오 의약품 공정 개발 및 특성 분석을 위한 DOE 연구 사례1-1: 바이오 의약품 공정 개발 및 특성 분석을 위한 DOE 연구 사례
1-1: 바이오 의약품 공정 개발 및 특성 분석을 위한 DOE 연구 사례
 
1-2: QbD 기반 의약품 개발의 통계 분석 및 JMP 활용 사례
1-2: QbD 기반 의약품 개발의 통계 분석 및 JMP 활용 사례1-2: QbD 기반 의약품 개발의 통계 분석 및 JMP 활용 사례
1-2: QbD 기반 의약품 개발의 통계 분석 및 JMP 활용 사례
 
2-1: 석유 화학 산업에서의 JMP 활용 사례
2-1: 석유 화학 산업에서의 JMP 활용 사례2-1: 석유 화학 산업에서의 JMP 활용 사례
2-1: 석유 화학 산업에서의 JMP 활용 사례
 
2-2: 다양한 무고장 조건에서 필드 차량 클레임 분석 자동화
2-2: 다양한 무고장 조건에서 필드 차량 클레임 분석 자동화2-2: 다양한 무고장 조건에서 필드 차량 클레임 분석 자동화
2-2: 다양한 무고장 조건에서 필드 차량 클레임 분석 자동화
 
2-3: 데이터 분석 기반의 B2B 영업
2-3: 데이터 분석 기반의 B2B 영업2-3: 데이터 분석 기반의 B2B 영업
2-3: 데이터 분석 기반의 B2B 영업
 
3-2: Smart Proactive Analysis Material
3-2: Smart Proactive Analysis Material3-2: Smart Proactive Analysis Material
3-2: Smart Proactive Analysis Material
 
3-3: 분석 콘텐츠 재사용성 향상을 위한 AppStore 생태계 구축
3-3: 분석 콘텐츠 재사용성 향상을 위한 AppStore 생태계 구축3-3: 분석 콘텐츠 재사용성 향상을 위한 AppStore 생태계 구축
3-3: 분석 콘텐츠 재사용성 향상을 위한 AppStore 생태계 구축
 
4-1: JMP/Python 에코시스템 성공 사례
4-1: JMP/Python 에코시스템 성공 사례4-1: JMP/Python 에코시스템 성공 사례
4-1: JMP/Python 에코시스템 성공 사례
 
4-2: 네거티브 디자인 개념을 활용한 군함 초기 설계에 적용되는 디자인 영역 탐색
4-2: 네거티브 디자인 개념을 활용한 군함 초기 설계에 적용되는 디자인 영역 탐색4-2: 네거티브 디자인 개념을 활용한 군함 초기 설계에 적용되는 디자인 영역 탐색
4-2: 네거티브 디자인 개념을 활용한 군함 초기 설계에 적용되는 디자인 영역 탐색
 
4-3: 제조 계측 장비 비정형 데이터 파일 기반으로 데이터 변환 및 관리도 분석 서비스
4-3: 제조 계측 장비 비정형 데이터 파일 기반으로 데이터 변환 및 관리도 분석 서비스4-3: 제조 계측 장비 비정형 데이터 파일 기반으로 데이터 변환 및 관리도 분석 서비스
4-3: 제조 계측 장비 비정형 데이터 파일 기반으로 데이터 변환 및 관리도 분석 서비스
 

Recently uploaded

Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
University of Maribor
 
International Conference on NLP, Artificial Intelligence, Machine Learning an...
International Conference on NLP, Artificial Intelligence, Machine Learning an...International Conference on NLP, Artificial Intelligence, Machine Learning an...
International Conference on NLP, Artificial Intelligence, Machine Learning an...
gerogepatton
 
Harnessing WebAssembly for Real-time Stateless Streaming Pipelines
Harnessing WebAssembly for Real-time Stateless Streaming PipelinesHarnessing WebAssembly for Real-time Stateless Streaming Pipelines
Harnessing WebAssembly for Real-time Stateless Streaming Pipelines
Christina Lin
 
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMSA SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
IJNSA Journal
 
Embedded machine learning-based road conditions and driving behavior monitoring
Embedded machine learning-based road conditions and driving behavior monitoringEmbedded machine learning-based road conditions and driving behavior monitoring
Embedded machine learning-based road conditions and driving behavior monitoring
IJECEIAES
 
Textile Chemical Processing and Dyeing.pdf
Textile Chemical Processing and Dyeing.pdfTextile Chemical Processing and Dyeing.pdf
Textile Chemical Processing and Dyeing.pdf
NazakatAliKhoso2
 
2008 BUILDING CONSTRUCTION Illustrated - Ching Chapter 02 The Building.pdf
2008 BUILDING CONSTRUCTION Illustrated - Ching Chapter 02 The Building.pdf2008 BUILDING CONSTRUCTION Illustrated - Ching Chapter 02 The Building.pdf
2008 BUILDING CONSTRUCTION Illustrated - Ching Chapter 02 The Building.pdf
Yasser Mahgoub
 
IEEE Aerospace and Electronic Systems Society as a Graduate Student Member
IEEE Aerospace and Electronic Systems Society as a Graduate Student MemberIEEE Aerospace and Electronic Systems Society as a Graduate Student Member
IEEE Aerospace and Electronic Systems Society as a Graduate Student Member
VICTOR MAESTRE RAMIREZ
 
The Python for beginners. This is an advance computer language.
The Python for beginners. This is an advance computer language.The Python for beginners. This is an advance computer language.
The Python for beginners. This is an advance computer language.
sachin chaurasia
 
Unit-III-ELECTROCHEMICAL STORAGE DEVICES.ppt
Unit-III-ELECTROCHEMICAL STORAGE DEVICES.pptUnit-III-ELECTROCHEMICAL STORAGE DEVICES.ppt
Unit-III-ELECTROCHEMICAL STORAGE DEVICES.ppt
KrishnaveniKrishnara1
 
A review on techniques and modelling methodologies used for checking electrom...
A review on techniques and modelling methodologies used for checking electrom...A review on techniques and modelling methodologies used for checking electrom...
A review on techniques and modelling methodologies used for checking electrom...
nooriasukmaningtyas
 
Eric Nizeyimana's document 2006 from gicumbi to ttc nyamata handball play
Eric Nizeyimana's document 2006 from gicumbi to ttc nyamata handball playEric Nizeyimana's document 2006 from gicumbi to ttc nyamata handball play
Eric Nizeyimana's document 2006 from gicumbi to ttc nyamata handball play
enizeyimana36
 
Generative AI leverages algorithms to create various forms of content
Generative AI leverages algorithms to create various forms of contentGenerative AI leverages algorithms to create various forms of content
Generative AI leverages algorithms to create various forms of content
Hitesh Mohapatra
 
Recycled Concrete Aggregate in Construction Part II
Recycled Concrete Aggregate in Construction Part IIRecycled Concrete Aggregate in Construction Part II
Recycled Concrete Aggregate in Construction Part II
Aditya Rajan Patra
 
Modelagem de um CSTR com reação endotermica.pdf
Modelagem de um CSTR com reação endotermica.pdfModelagem de um CSTR com reação endotermica.pdf
Modelagem de um CSTR com reação endotermica.pdf
camseq
 
Question paper of renewable energy sources
Question paper of renewable energy sourcesQuestion paper of renewable energy sources
Question paper of renewable energy sources
mahammadsalmanmech
 
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressionsKuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
Victor Morales
 
ML Based Model for NIDS MSc Updated Presentation.v2.pptx
ML Based Model for NIDS MSc Updated Presentation.v2.pptxML Based Model for NIDS MSc Updated Presentation.v2.pptx
ML Based Model for NIDS MSc Updated Presentation.v2.pptx
JamalHussainArman
 
ISPM 15 Heat Treated Wood Stamps and why your shipping must have one
ISPM 15 Heat Treated Wood Stamps and why your shipping must have oneISPM 15 Heat Treated Wood Stamps and why your shipping must have one
ISPM 15 Heat Treated Wood Stamps and why your shipping must have one
Las Vegas Warehouse
 
22CYT12-Unit-V-E Waste and its Management.ppt
22CYT12-Unit-V-E Waste and its Management.ppt22CYT12-Unit-V-E Waste and its Management.ppt
22CYT12-Unit-V-E Waste and its Management.ppt
KrishnaveniKrishnara1
 

Recently uploaded (20)

Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
 
International Conference on NLP, Artificial Intelligence, Machine Learning an...
International Conference on NLP, Artificial Intelligence, Machine Learning an...International Conference on NLP, Artificial Intelligence, Machine Learning an...
International Conference on NLP, Artificial Intelligence, Machine Learning an...
 
Harnessing WebAssembly for Real-time Stateless Streaming Pipelines
Harnessing WebAssembly for Real-time Stateless Streaming PipelinesHarnessing WebAssembly for Real-time Stateless Streaming Pipelines
Harnessing WebAssembly for Real-time Stateless Streaming Pipelines
 
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMSA SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
 
Embedded machine learning-based road conditions and driving behavior monitoring
Embedded machine learning-based road conditions and driving behavior monitoringEmbedded machine learning-based road conditions and driving behavior monitoring
Embedded machine learning-based road conditions and driving behavior monitoring
 
Textile Chemical Processing and Dyeing.pdf
Textile Chemical Processing and Dyeing.pdfTextile Chemical Processing and Dyeing.pdf
Textile Chemical Processing and Dyeing.pdf
 
2008 BUILDING CONSTRUCTION Illustrated - Ching Chapter 02 The Building.pdf
2008 BUILDING CONSTRUCTION Illustrated - Ching Chapter 02 The Building.pdf2008 BUILDING CONSTRUCTION Illustrated - Ching Chapter 02 The Building.pdf
2008 BUILDING CONSTRUCTION Illustrated - Ching Chapter 02 The Building.pdf
 
IEEE Aerospace and Electronic Systems Society as a Graduate Student Member
IEEE Aerospace and Electronic Systems Society as a Graduate Student MemberIEEE Aerospace and Electronic Systems Society as a Graduate Student Member
IEEE Aerospace and Electronic Systems Society as a Graduate Student Member
 
The Python for beginners. This is an advance computer language.
The Python for beginners. This is an advance computer language.The Python for beginners. This is an advance computer language.
The Python for beginners. This is an advance computer language.
 
Unit-III-ELECTROCHEMICAL STORAGE DEVICES.ppt
Unit-III-ELECTROCHEMICAL STORAGE DEVICES.pptUnit-III-ELECTROCHEMICAL STORAGE DEVICES.ppt
Unit-III-ELECTROCHEMICAL STORAGE DEVICES.ppt
 
A review on techniques and modelling methodologies used for checking electrom...
A review on techniques and modelling methodologies used for checking electrom...A review on techniques and modelling methodologies used for checking electrom...
A review on techniques and modelling methodologies used for checking electrom...
 
Eric Nizeyimana's document 2006 from gicumbi to ttc nyamata handball play
Eric Nizeyimana's document 2006 from gicumbi to ttc nyamata handball playEric Nizeyimana's document 2006 from gicumbi to ttc nyamata handball play
Eric Nizeyimana's document 2006 from gicumbi to ttc nyamata handball play
 
Generative AI leverages algorithms to create various forms of content
Generative AI leverages algorithms to create various forms of contentGenerative AI leverages algorithms to create various forms of content
Generative AI leverages algorithms to create various forms of content
 
Recycled Concrete Aggregate in Construction Part II
Recycled Concrete Aggregate in Construction Part IIRecycled Concrete Aggregate in Construction Part II
Recycled Concrete Aggregate in Construction Part II
 
Modelagem de um CSTR com reação endotermica.pdf
Modelagem de um CSTR com reação endotermica.pdfModelagem de um CSTR com reação endotermica.pdf
Modelagem de um CSTR com reação endotermica.pdf
 
Question paper of renewable energy sources
Question paper of renewable energy sourcesQuestion paper of renewable energy sources
Question paper of renewable energy sources
 
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressionsKuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
 
ML Based Model for NIDS MSc Updated Presentation.v2.pptx
ML Based Model for NIDS MSc Updated Presentation.v2.pptxML Based Model for NIDS MSc Updated Presentation.v2.pptx
ML Based Model for NIDS MSc Updated Presentation.v2.pptx
 
ISPM 15 Heat Treated Wood Stamps and why your shipping must have one
ISPM 15 Heat Treated Wood Stamps and why your shipping must have oneISPM 15 Heat Treated Wood Stamps and why your shipping must have one
ISPM 15 Heat Treated Wood Stamps and why your shipping must have one
 
22CYT12-Unit-V-E Waste and its Management.ppt
22CYT12-Unit-V-E Waste and its Management.ppt22CYT12-Unit-V-E Waste and its Management.ppt
22CYT12-Unit-V-E Waste and its Management.ppt
 

3-1: Semiconductor Wafer Test Fail Map Clustering and Data Mining using JMP Add-In

  • 1. Copyright © JMP Statistical Discovery LLC. All Rights Reserved. KOREA
  • 2. Semiconductor Wafer Test Fail Map Clustering and Data Mining using JMP Add-In 곽병훈, SK하이닉스
  • 3. ✓ Semiconductor Wafer Test Failure Map Clustering and Data Mining using JMP Add-In 1. Semiconductor Process 2. Good, Bad Grouping in conventional way 3. User Friendly Interface 4. Data Mining Process & Result 5. Example 6. Summary
  • 4. Over 99% of the defects in wafer test are within fabrication Wafer Test Failure
  • 5. Good, Bad grouping in Convention way (1) Failure Rate (%) Wafer Test Time >10.0% ↑BAD ↓GOOD <1.00% ✓ Final Data Mining → Good Gr. vs Bad Gr. ✓ Problem → Garbage in, Garbage out Various Maps @Bad Gr. Blue = Fail Orange = Pass This is the type of wafer map engineers want to analyze. * Goal Find the real cause ✓ . ✓ . ✓ . ✓ . ✓ . but, Garbage in, Garbage out There are various types of failure map including the map we are finding → Garbage in
  • 6. Good, Bad grouping in Convention way (2) A large loss of time = Inefficient ✓ Advantage : an accurate classification ✓ Disadvantage : Time Consumption ✓ Engineer finds and classifies map. We suggest the solutions with this problems in this presentation.
  • 7. User Friendly Interface * Clustering -. sda symbol -. EPM -. RPM -. WT Item (Fail Bit Count) *sda = statistical defect analysis
  • 8. User Friendly Interface Set period Select Fab. Select product Select defect Select Failure to analyze Clustering Mining * Clustering -. sda symbol ** -. EPM -. RPM ** -. WT Item (Fail Bit Count) **sda = statistical defect analysis **RPM = Reliability Parameter Monitoring * Clustering selection → Set period → Select Fab. → Select product → Cluster Analysis Click!! → Select defect category → Select Failure Maps to analyze Bad Good Good Exclude Exclude Exclude Exclude
  • 9. User Friendly Interface Bad Good Good Exclude Exclude Exclude Exclude Bad Gr. Lot & Wafer List ✓ Good, Bad Portion Lot and Wafer Lot and Wafer ✓ Clustering Group Portion Clustering Group Portion Good, Bad Portion Fail Rate Fail Rate Fail Rate Fail Rate Fail Rate Fail Rate Fail Rate Fail Rate Fail Rate Fail Rate Fail Rate Fail Rate Fail Rate Fail Rate ✓ Good, Bad group fail rate comparison
  • 10. Classification using Hierarchical Clustering Set period Select Fab. Select product Select defect Select Fail to analyze Good Good Exclude Exclude Exclude Exclude Exclude Bad Bad Bad Bad Bad Bad Bad Good ✓ Clustering Group : Engineers can select certain Area of a Wafer and classify that area as a good group or bad group Good Gr. Bad Gr. VS ✓ Select Conditions ✓ Final Mining Group * Good, Bad and Exclude Multiple choices possible ✓ Designating the good or bad groups directly by looking at the clustered map group ✓ Mining with final Good, Bad Groups ✓ Classifying failure maps in accurate and fast. ➔ Dramatically improve engineer time loss
  • 11. ✓ Mining (Inline Legend) Select Oper. Select Legend. Data Mining Process Interface * Select Inline → Select Start Data, End Data → Select Oper. → Select Legend. → Click “연계분석”
  • 12. Data Mining Result Interface Rank Summary by Algorithm Mining Result Cause Process Ranking Various mining algorithms OOOO Step Legend OOOO Step Legend OOOO Step Legend * Algorithms -. Decision Tree -. Randomforest -. ANOVA
  • 13. ✓ Mining (EPM) Data Mining Process Interface aaa para. aaa para. * Select EPM → Select Start Data, End Data → Click “연계분석”
  • 14. Clustering & Data Mining Example (1) ✓ Select Conditions Good Good Exclude Bad Bad Bad ✓ Clustering Result ✓ Mining Result Cause Process Rank Summary by Algorithm OOOO Step Legend OOOO Step Legend OOOO Step Legend OOOO Step Legend OOOO Step Legend OOOO Step Legend A EQ. B EQ. C EQ. D EQ. D EQ. C EQ. B EQ. A EQ. OOOO Step Process Run Time ✓ Result Click!! Focused on specific EQ (D EQ) and process run time. → These could be most likely the causes.
  • 15. Clustering & Data Mining Example (2) ✓ Mining Result (Data Visualization : Parallel Plot) OOO Step OOO Step OOO Step OOO Step OOO Step OOO Step OOO Step OOO Step OOO Step OOO Step OOO Step OOO Step OOO Step By Process Step Legend Good group scattered. Bad group scattered. The path of Bad group overlapped with one equipment. →This could be most likely the cause. a EQ b EQ c EQ d EQ * Association Analysis is also possible. Data Visualization Analysis
  • 16. Clustering & Data Mining Example (3) ○ Delayed step process detection between Good and Bad group using multivariate. Dual axis occurrence Good Bad Shift Bad Gr. Good Gr. A Step Process run time B Step Process run time C Step Process run time D Step Process run time Process Flow Process Flow Good Gr. Bad Gr. +1 Day +1 Day +1 Day +1 Day +8 Day +1 Day ↙Cause of defect A Step Process run time B Step Process run time C Step Process run time D Step Process run time A Step Process run time B Step Process run time A Step Process run time C Step Process run time D Step Process run time A Step Process run time A Step Process run time A Step Process run time B Step Process run time B Step Process run time Process run time Process run time ○ Delay step detection between Good and Bad group using multivariate. If there is a problem with delay in B step Data Visualization Analysis
  • 17. WF Count Human [min] AI [min] 100 6 0.03 1000 60 0.16 3000 180 5 Clustering Process Time [ Human vs AI ] Human [min] AI [min] AI 180min 5min Human Wafer Count ✓ Dramatically reduces time consumption for engineer
  • 18. Summary 1. Classifying failure maps in faster and more accurate way 2. User Friendly Interface. 3. Reducing engineer’s time spends on repetitive work tasks dramatically 4. Anyone can use JMP Add-In.
  • 19. Copyright © JMP Statistical Discovery LLC. All rights reserved. KOREA Thank you