Module 10 · Lesson 7 of 18
End-line and final inspection improvement
Learning objectives
- Assess end-line/final inspection performance with escape and false-reject rates.
- Use attribute agreement to calibrate inspectors against a shared standard.
- Redesign inspection as risk-based, not layered.
Problem and CTQ
Inspection can miss defects, create delay or give inconsistent disposition. Manage escape rate, false-reject rate and buyer AQL together, not one at a time.
DMAIC path
- Map end-line, final, repair and re-inspection flows.
- Validate attribute agreement (kappa) against a calibrated defect library.
- Stratify escape and false-reject rates by lot, buyer, style and inspector.
Improvement choices
- Calibrated defect library and inspector certification.
- Risk-based allocation of inspection minutes.
- Ergonomic station, lighting standard and digital traceability.
Control plan
- Escape rate, false reject, agreement/kappa, final pass, inspection minutes, repair loop time and lot rejection.
- Buyer/quality authority sign-off for any AQL or acceptance change.
Garment-factory example
A women's woven blouse programme shows kappa 0.42 between two inspectors on 'puckering'. A photo-based defect library and joint calibration raise kappa to 0.81 while escape rate falls from 0.9% to 0.3%.
Common mistakes
- Treating inspection as prevention.
- Changing AQL informally to hit a shipment date.
Knowledge check
Pick one answer per question. Explanations appear after you submit.
1. The healthiest inspection redesign is:
2. Kappa of 0.42 means:
Author: Sanjeewa Dehiwalage · Last reviewed: 2026-07-21