AI-Based Industrial Management for Enhancing Operational Manufacturing Processes of Medical Bed Parts via AI-Driven Quality

Gholampoor, Hadi

Journal of Industrial Engineering and Management
Vol 1, No 2 (2023)

In the realm of medical equipment manufacturing, ensuring the quality of each component is crucial due to the direct impact on patient safety and product reliability. This study introduces a novel application of machine learning within industrial management to enhance the operational manufacturing processes of medical bed parts. Utilizing a Random Forest classifier, we developed a predictive model based on five critical features collected during the manufacturing process: the physical dimensions of Length, Width, Height, Weight of the parts, and the operator involved in manual grinding. The classifier aimed to predict whether each part would be defective or accepted before assembly, potentially revolutionizing the traditional quality control approach by reducing dependency on post-manufacturing inspections and minimizing human error. The model was trained on a dataset of 500 parts, with a class distribution reflecting a significant imbalance between defected and accepted pieces. Despite this, the classifier achieved a high accuracy of 97.0% on the test set, demonstrating robustness and reliability in predicting part quality. Feature importance analysis revealed that while physical attributes like Weight and Height significantly influenced predictions, operator variability also played a crucial role, indicating areas for operational improvement through training and standardization. This research highlights how integrating AI into industrial manufacturing processes can significantly enhance efficiency, reduce waste, and ensure higher standards of quality control, setting a precedent for future applications in similar high-stakes manufacturing environments