Machine Learning Algorithms for Quality Control in Automated Manufacturing: Defect Prediction Accuracy in Technical Education Department, Niger Delta University, Bayelsa State

Abstract

This study investigates the efficacy of various machine learning algorithms in predicting manufacturing defects within the automated manufacturing training environment of the Technical Education Department at Niger Delta University, Nigeria. Using data collected from the department's manufacturing simulation systems over a two-year period (2022-2024), we evaluated the performance of five machine learning algorithms: Random Forest, Support Vector Machines (SVM), Artificial Neural Networks (ANN), K-Nearest Neighbors (KNN), and Gradient Boosting Machines (GBM). The research employed 10-fold cross-validation and performance metrics including accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC-ROC). Results demonstrate that the Random Forest algorithm achieved the highest overall accuracy (94.2%) and F1-score (0.93), making it particularly suitable for defect prediction in the technical education manufacturing context. The findings provide valuable insights for integrating advanced predictive modeling into technical education curricula and improving quality control processes in automated manufacturing environments. This research addresses the critical gap between theoretical machine learning applications and practical implementation in technical education settings in developing economies.