Sustainable Maintenance Melalui Prediksi Preventive Maintenance di Plant Cold Roll Mills (CRM) PT Krakatau Steel (Persero) Tbk dengan Algoritma Naïve Bayes Classifier dan Decision Tree

  • Hidayatudin Shodiqin STMIK Likmi
Keywords: preventive maintenance, data mining, naïve bayes, decision tree, PT. Krakatau Steel

Abstract

PT. Krakatau Steel (Persero),Tbk is a state-owned company engaged in the largest steel producer in Indonesia. The increasing number of steel product production at PT Krakatau Steel needs to be supported by excellent production facilities. The production process can be stopped if there is damage to the machine. Companies need to predict when the machine must be maintained so that sustainable maintenance can be carried out properly. The purpose of this research is to prevent unexpected damages, especially for equipment that has potential damage. Researchers predict preventive maintenance for maintenance locations, types of maintenance, and cost centers at the CRM (Cold Roll Mills) plant. Data Mining processing using the Naïve Bayes Algorithm to help find predictions for two types of maintenance (WP: Preventive & WE: Emergency). The data is reprocessed using the Decision Tree algorithm to determine which maintenance locations need maintenance activities. The results showed that Preventive Maintenance in Plant CRM (Cold Roll Mills) was running well in only 1.39% of Emergency Maintenance data from a total of 5034 records. Prediction results from the Naïve Bayes algorithm resulted in 8 emergency maintenance records with a class precision of 88.89%. Preventive maintenance data is 2416 records as predicted, and only one record predicted to emergency maintenance with a precision class of 99.96%. This research uses data testing 2626 records with an accuracy rate of 99.92%. This study uses data testing 2626 records with an accuracy rate of 99.92%. The result of the Decision tree is that it can show the location, maintenance activities, types of maintenance, and which cost centers should receive treatment

References

[1] M. M. Baharuddin, T. Hasanuddin, and H. Azis, “Analisis Performa Metode K-Nearest Neighbor untuk Identifikasi Jenis Kaca,” Ilk. J. Ilm., vol. 11, no. 28, pp. 269–274, 2019.
[2] A. Ilham, “Komparasi Algoritma Klasifikasi Dengan Pendekatan Level Data Untuk Menangani Data Kelas Tidak Seimbang,” J. Ilm. Ilmu Komput., vol. 3, no. 1, pp. 9–14, 2017.
[3] Bustami, 2013, Penerapan Algoritma Naive Bayes Untuk Mengklasifikasi Data Nasabah Asuransi”, TECHSI : Jurnal Penelitian Teknik Informatika, vol. 3, no.2, hal. 127-146.
[4] CTI, B. R., Gafar, A. A., Fajriani, N., Ramdani, U., Uyun, F. R., Ransi, N., dkk, 2017. Implemetasi k-means clustering pada rapidminer untuk analisis daerah rawan kecelakaan. Prosiding seminar nasional riset kuantitatif terapan, Vol. 1
Published
2022-06-09