Improvement of Defect Engineering Methods
What we mean
Early detection of production related quality issues and proactive responses ensure low yield loss.
How we do it
Through methods like sampling, control strategy, defect to yield correlation, defect reporting, defect cataloguing , statistical analysis of defect data in relation to all other production relevant data we can create the most effective approach for you.
- AI Applications
- Software Solutions
- Yield/Quality
- Consulting
- Data Integration
- Implementation
- Organization
- Data Analytics
- yield and reliability improvement
- defect engineering
- process control methodology
- toolmonitoring
- risk assessment
- workshops
- coaching
- moderation
Example 1
Chip based Defect to Yield Correlation
‚Contigency table‘ Methode / Berechnung von Kill Rate & Defect Related Loss
Requirements
quality of inspection recipes, defect filter: large defects, Adder, Defects of interest per inspection step, low parametric yield loss
Process step A
Kill Rate: 21%
Defect Related Loss: 3,1%
(All Binsorts)
Example 2
Yield correlation with intitial large detect filters
4 weeks data, one product, standard wafer, Large defects only, special defect related binsorts used, low yielding wafer excluded
Example 3
Large Defect Filter definition and optimization
Filter:
Volume > 35 & Grade > 50
New Filter:
Volume > 25 & Grade > 50
Initial definition of Large Defect filter by inspection tool raw parameters: Volume/ Grade (DF inspection), DSIZE (BF inspection)
Optimization of Large Defect filter by kill rate and defective related yield loss analysis