P‐102: Explainable AI Approach for MOD Outliers in FAB to MOD Process

Display panels manufactured in the FAB process are assembled with various films and components in the module (MOD) process, and their final quality is a result of accumulated the quality of hundreds of preceding processes. Due to these limitations, it is hard to analyze the cause of MOD's outli...

Full description

Saved in:
Bibliographic Details
Published in:SID International Symposium Digest of technical papers 2023-06, Vol.54 (1), p.1564-1567
Main Authors: Park, Sujin, Choi, Hyunho, Yim, Taekyung, Kyung, Yunyoung, Jin, Younggil, Kim, Jaewon
Format: Article
Language:English
Subjects:
AnoGAN
Anomaly Detection
AutoML
Data analysis
Diplay
eXplainable AI
Explainable artificial intelligence
f- AnoGAN
Modules
OLED
SHAP
XAI
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Display panels manufactured in the FAB process are assembled with various films and components in the module (MOD) process, and their final quality is a result of accumulated the quality of hundreds of preceding processes. Due to these limitations, it is hard to analyze the cause of MOD's outliers relevant to the FAB process and much time and money are consumed to maintain the cause process to eliminate MOD outliers. Therefore, we propose a new explainable AI (XAI) approach to detect and improve the module process's outliers by finding the correlation between FAB and MOD in this paper. The proposed approach was verified for MOD outlier cases in the real world's OLED development and manufacturing process. As a result, it confirmed 96.8% of the average accuracy for classifications. This approach presents the FAB factor that caused MOD's outliers based on the correlation between the MOD and the FAB.
ISSN:0097-966X
2168-0159
DOI:10.1002/sdtp.16891