AI Vision Inspection Application and Assessment Guideline

Die Annotation des Normtextes:

CQI-38 provides a standardized, industry-focused approach for assessing and managing AI-based vision inspection systems in automotive manufacturing. The guidance helps OEMs, suppliers at all tiers, and vision inspection equipment manufacturers evaluate system capability, accuracy, and reliability while addressing the risks associated with dynamic, data-driven inspection processes.

Designed to supplement IATF 16949, CQI-38 focuses on AI systems that use image data to perform feature recognition, defect classification, and status evaluation rather than traditional rule-based inspection or precise dimensional measurement. The guidance takes a total lifecycle approach, covering system planning and manufacturing implementation through system and process acceptance, capability maintenance, and continual improvement.

CQI-38 also emphasizes risk-based validation and controlling changes to equipment, AI models, data, and operating environments. This approach helps organizations maintain reliable inspection performance, support customer-specific requirements, and strengthen their AI quality management capabilities.