Efficient Assembly Line Operations through Machine Vision-Enabled Assembly Part Classification

Assembly line complexity presents significant challenges in integrating various systems, subsystems, and mechanical components. Manual inspection of assembled parts often leads to quality assurance issues, potentially resulting in production downtime, cost overruns, and product failures. This research proposes a novel automated machine vision system to address these challenges by identifying and classifying mechanical parts' spatial dimensions in assembly lines. The study compared the results of deep learning techniques. Specifically, Convolutional Neural Networks and ResNet50 are trained on a high-performance computing infrastructure for robust component identification. Our approach achieves a remarkable 99.68% training accuracy, significantly outperforming existing methods. The developed model is deployed for real-time prediction of components in assembly lines, offering a versatile solution applicable across various manufacturing sectors. A key innovation of this study is integrating the deep learning model with a user-friendly Graphical User Interface (GUI), enhancing operator efficiency and reducing human error. This research contributes to the field of smart manufacturing by demonstrating the feasibility and effectiveness of AI-driven quality control in complex assembly processes, potentially transforming industrial inspection methodologies.