Reducing durably the cost of poor quality through intelligent defect analysis
Intelligent clustering of non-conformities to prioritize corrective actions and sustainably reduce the costs of poor quality.
Contexte
Airbus processes millions of non-conformities (NC) on purchased parts, often managed in silos by each production site. This fragmentation creates major inefficiencies and significantly increases costs of poor quality (COPQ). The critical challenge is to centralize analysis to identify root causes and prioritize strategic corrective actions.
Solution
We deployed a Machine Learning model based on an approach of automated clustering. The solution intelligently groups non-conformities by similar characteristics to predict their actual criticality. This tool allows Quality and Procurement teams to focus their efforts on the incidents with the greatest industrial impact.
Bénéfices
- Significant reduction in administrative workload related to managing disputes and anomalies
- Reduction in financial impact and operational impact of defects through criticality prediction
- Increased profitability through better resource allocation for high-stakes economic issues
"The major challenge was to unify non-compliance data from heterogeneous industrial sites. We used NLP to standardize this complex textual information. This step was crucial in enabling the clustering model to detect recurring patterns that were invisible during manual management."
Reduce your cost of poor quality (COPQ)
Optimize your anomaly management with predictive AI
Autres cas clients
