| Abstrakt: |
The emergence of Industry 4.0, also known as the fourth industrial revolution, has brought forth the concept of prognostics and health management (PHM) as an inevitable trend in the realm of industrial big data and smart manufacturing. This study aims to present a proof-of-concept that illustrates how machine learning can be employed to analyze industrial facility data and anticipate the condition of industrial machines. Specifically, a comprehensive case study focusing on vibration monitoring is conducted. The proposed models aim to predict maintenance requirements for the forced blower of a chemical plant by utilizing vibration data obtained during the manufacturing process. To validate the methodology, five different machine learning algorithms, namely logistic regression (LR), support vector machine (SVM), K-nearest neighbor (KNN), extreme gradient boosting (XGBoost), and random forest (RF), are employed. The evaluation metrics used include Matthews correlation coefficient (MCC) and receiver operator characteristic curve (ROC). This study aims to establish a relationship between machine failures caused by vibration and the prediction of both healthy and faulty bearings using the machine learning approaches. The findings indicate that the XGBoost algorithm outperforms other approaches with an MCC of 0.800 and a higher area under the ROC curve. [ABSTRACT FROM AUTHOR] |
Full text from SpringerLink