FdQAutonomous Quality Inspection in Flexible Production Environments

Helena Marquez

FdQAutonomous Quality Inspection in Flexible Production Environments

Author(s)	Dr. Helena Marquez
Country	United States
Abstract	Flexible production environments increasingly require quality-inspection systems capable of operating across changing products, variable lot sizes, frequent changeovers, evolving defect classes, and non-stationary manufacturing conditions. Conventional manual inspection and fixed-rule machine-vision systems can provide acceptable performance under stable conditions but become difficult to maintain when product geometry, surface appearance, lighting, process parameters, or defect distributions change. This study develops an autonomous quality-inspection framework for flexible manufacturing based on adaptive machine vision, industrial anomaly detection, uncertainty estimation, active learning, human-supervised decision escalation, and continuous model monitoring. Because no proprietary production dataset or factory experiment was supplied, the quantitative analysis is explicitly presented as a simulation-based methodological study. Four hypothetical inspection architectures are compared: manual/rule-based inspection, fixed artificial-intelligence inspection, adaptive multi-product inspection, and autonomous human-supervised quality inspection. Five dimensions are evaluated: defect-detection capability, product-change adaptability, inspection throughput, data efficiency, and decision reliability. The simulated Composite Flexible Inspection Score increases from 54 for manual/rule-based inspection to 71 for fixed AI inspection, 84 for adaptive multi-product inspection, and 92 for autonomous human-supervised inspection. The integrated architecture performs most strongly because it combines automated decision-making with confidence estimation, selective operator review, continuous model adaptation, and production-system feedback. The numerical results are illustrative rather than empirical. The paper argues that autonomous inspection should not mean eliminating human quality expertise; instead, autonomy should allow routine inspections to be executed automatically while uncertain, novel, or high-risk cases are escalated to qualified personnel. The proposed Flexible Autonomous Quality Inspection Framework provides a foundation for future implementation using industrial image datasets, edge-computing hardware, active-learning pipelines, digital production records, and real-world validation.
Keywords	autonomous quality inspection; flexible manufacturing; machine vision; industrial anomaly detection; deep learning; active learning; smart manufacturing; defect detection; adaptive inspection; human-in-the-loop AI
Field	Engineering
Published In	Volume 2, Issue 1, January-February 2021
Published On	2021-01-11

View / Download PDF File

About AJAMIR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+1-872-302-3851	editor@ajamir.com

American Journal of Advanced Multidisciplinary Innovation and Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

FdQAutonomous Quality Inspection in Flexible Production Environments

Share this