Similarity Ratio Analysis for Early Stage Fault Detection with Optical Emission Spectrometer in Plasma Etching Process

A Similarity Ratio Analysis (SRA) method is proposed for early-stage Fault Detection (FD) in plasma etching processes using real-time Optical Emission Spectrometer (OES) data as input. The SRA method can help to realise a highly precise control system by detecting abnormal etch-rate faults in real-t...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:PloS one Ročník 9; číslo 4; s. e95679
Hlavní autori: Yang, Jie, McArdle, Conor, Daniels, Stephen
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: United States Public Library of Science 01.04.2014
Public Library of Science (PLoS)
Predmet:
Alarm systems
Algorithms
Analysis
Chemical reactions
Computer and Information Sciences
Control systems
Electronics industry
Emission analysis
Emissions
Engineering
Engineering and Technology
Etching
False alarms
Fault detection
Integrated circuit fabrication
Methods
Models, Theoretical
Physical Sciences
Plasma etching
Quality
Real time
Similarity
Time series
ISSN:1932-6203, 1932-6203
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:A Similarity Ratio Analysis (SRA) method is proposed for early-stage Fault Detection (FD) in plasma etching processes using real-time Optical Emission Spectrometer (OES) data as input. The SRA method can help to realise a highly precise control system by detecting abnormal etch-rate faults in real-time during an etching process. The method processes spectrum scans at successive time points and uses a windowing mechanism over the time series to alleviate problems with timing uncertainties due to process shift from one process run to another. A SRA library is first built to capture features of a healthy etching process. By comparing with the SRA library, a Similarity Ratio (SR) statistic is then calculated for each spectrum scan as the monitored process progresses. A fault detection mechanism, named 3-Warning-1-Alarm (3W1A), takes the SR values as inputs and triggers a system alarm when certain conditions are satisfied. This design reduces the chance of false alarm, and provides a reliable fault reporting service. The SRA method is demonstrated on a real semiconductor manufacturing dataset. The effectiveness of SRA-based fault detection is evaluated using a time-series SR test and also using a post-process SR test. The time-series SR provides an early-stage fault detection service, so less energy and materials will be wasted by faulty processing. The post-process SR provides a fault detection service with higher reliability than the time-series SR, but with fault testing conducted only after each process run completes.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
Competing Interests: This work is sponsored by the Irish Research Council and Intel Ireland. This does not alter the authors’ adherence to PLOS ONE policies on sharing data and materials.
Conceived and designed the experiments: JY CM SD. Performed the experiments: JY. Analyzed the data: JY CM SD. Contributed reagents/materials/analysis tools: SD. Wrote the paper: JY. Revised and approved the manuscript: JY CM SD.
ISSN:1932-6203
1932-6203
DOI:10.1371/journal.pone.0095679