Predictive Maintenance on Dry 8 Production Machine Line Using Support Vector Machine (SVM)

Rasyid, Mohammad Andi
Sukmono, Tedjo
Jakaria, Ribangun Bamban

Jurnal Teknik Industri : Jurnal Hasil Penelitian dan Karya Ilmiah dalam Bidang Teknik Industri
Vol 10, No 2 (2024): December 2024

Machines are the main element in manufacturing companies, and the role of machine performance is vital in the production process. Downtime problems caused by machine damage can significantly affect company productivity. This research implements the support vector machine (SVM) method for predicting Dry 8 production machine line maintenance, which aims to reduce downtime and increase productivity. The SVM method is known for its high accuracy and low error rate. The evaluation process used four kernel functions: linear, radial basis function (RBF), polynomial and sigmoid. The linear kernel function performed best with 99.8% accuracy, 83% precision, recall, and f1-score. These results show that the SVM method can be a viable solution to improve the efficiency of machine maintenance. Keywords: Confusion Matrix, Machine Learning, Predictive Maintenance, Support Vector Machine