الفهرس 
Chapter I:IntroductionOnly 14 pages are availabe for public view
 |  | from | 190 |  |  |
| | | from | 190 | | |
Abstract
Industry 5.0, with its emphasis on human-machine collaboration and redefined industrial processes, stands as a transformative phase succeeding Industry 4.0. Within this landscape, Prognostics and Health Management (PHM) have emerged as pivotal components in leveraging industrial big data and smart manufacturing. This study serves as a testament to the efficacy of machine learning methodologies in analyzing industrial facility data, marking a crucial stride from Industry 5.0 towards Industry 4.0 paradigms.
Focusing on vibration monitoring as a predictive indicator, this research aims to forecast maintenance requirements for the forced blower within TCI Sanmar Chemicals. Leveraging vibration data collected during the manufacturing process, a spectrum of machine learning algorithms—Logistic Regression (LR), Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Extreme Gradient Boosting (XGBoost), Multilayer Perceptron (MLP), and Random Forest (RF)—is employed to redefine predictive maintenance strategies within the Industry 5.0 framework.
Validation of these methodologies encompasses the use of evaluation metrics such as the Matthews Correlation Coefficient (MCC) and Receiver Operator characteristic Curve (ROC). The fundamental objective remains establishing a tangible relationship between machine failures attributed to vibrations and the predictive capabilities of these machine-learning approaches.
MCC emerged as the pivotal metric for evaluating algorithmic performance post-GridSearchCV. After Hyperparameter tuning, MLP exhibited exceptional progress, achieving the highest MCC of 0.850 alongside an impressive AUC of 0.961. These substantial advancements underscore the efficacy of GridSearchCV in significantly enhancing MLP’s predictive capabilities, showcasing notable improvements in both MCC and AUC metrics post-optimization.
The findings of this study underscore the potential of machine learning applications in prediction machinery status and preemptively identifying potential failures based on vibration analysis, thereby bridging the innovative advancements from Industry 5.0 towards Industry 4.0.