01
Problem
A subjective visual inspection method was needed to detect particulate, residue, hair, and fiber defects on medical-device components with consistent operator decisions.
02
Constraints
- Cleanliness defects were attribute-based and subject to interpretation.
- Operator-to-operator agreement had to be demonstrated.
- Lighting, magnification, and defect definitions required standardization.
- The method had to support regulated product acceptance decisions.
03
Engineering Decisions
- Built a controlled sample set with known acceptable and defective conditions.
- Defined standardized viewing conditions and defect criteria.
- Used attribute agreement analysis to quantify operator effectiveness.
- Created visual references to reduce ambiguity in borderline classifications.
04
Validation
- Executed attribute Gage R&R across multiple operators and trials.
- Compared operator decisions against known sample status.
- Reviewed agreement statistics and misclassification patterns before release.
05
Risk & Mitigation
- Operator fatigue could cause missed defects; mitigated through inspection conditions and workflow limits.
- False positives could increase scrap; mitigated through clear defect definitions and review logic.
- Ambiguous defect standards could reduce reproducibility; mitigated through visual reference materials.
06
Outcome
The inspection method was validated as a consistent attribute acceptance process for cleanliness-related defects.