Pengembangan Predictive Maintenance Copilot Berbasis Physics-Informed Machine Learning untuk Prediksi Risiko Kegagalan Mesin dan Estimasi Remaining Useful Life
DOI:
https://doi.org/10.54895/intech.v7i2.3732Keywords:
AI Copilot, Physics-Informed Machine Learning, Predictive Maintenance, Remaining Useful Life, , Sistem Pendukung KeputusanAbstract
Industri minyak dan gas membutuhkan strategi pemeliharaan yang mampu mendeteksi risiko kegagalan dan membantu menentukan prioritas tindakan. Penelitian ini mengembangkan Predictive Maintenance Copilot sebagai prototipe Sistem Pendukung Keputusan yang mengintegrasikan prediksi risiko, Physics-Informed Machine Learning, estimasi Remaining Useful Life, dan AI Copilot. Penelitian menggunakan pendekatan mixed method dengan dominasi kuantitatif-komputasional. Pemodelan dilakukan menggunakan AI4I 2020 Predictive Maintenance Dataset dengan membandingkan Logistic Regression, Decision Tree, Random Forest, dan XGBoost. Pendekatan physics-informed diterapkan melalui fitur turunan dan physics guardrails, sedangkan Remaining Useful Life dihitung secara terpisah berdasarkan tool wear, duty cycle, dan wear increment. XGBoost menghasilkan F2-score tertinggi sebesar 0,7373, accuracy 0,9489, dan recall kelas gagal 0,9412. Evaluasi terhadap 10 responden memperoleh tingkat penerimaan 90,8%, sedangkan seluruh pengujian perangkat lunak berhasil dijalankan. Hasil penelitian menunjukkan bahwa sistem berpotensi mendukung penyajian informasi pemeliharaan secara lebih terstruktur, tetapi masih memerlukan validasi menggunakan data mesin industri aktual.
Downloads
References
[1] Direktorat Jenderal Minyak dan Gas Bumi, “LAPORAN KINERJA Tahun 2025 Direktorat Jenderal Minyak dan Gas Bumi,” Indonesia, 2025. Accessed: Mar. 12, 2026. [Online]. Available: migas.esdm.go.id
[2] E. O. Nwulu, T. Y. Elete, O. V. Erhueh, O. A. Akano, and K. O. Omomo, “Machine learning applications in predictive maintenance: Enhancing efficiency across the oil and gas industry,” International Journal of Engineering Research Updates, vol. 5, no. 1, pp. 013–027, Sep. 2023, doi: 10.53430/ijeru.2023.5.1.0017.
[3] Y. Wu, B. S. Sicard, and S. A. Gadsden, “Physics-informed machine learning: A comprehensive review on applications in anomaly detection and condition monitoring,” Expert Syst. Appl., vol. 255, p. 124678, 2024, doi: 10.1016/j.eswa.2024.124678.
[4] M. H. de Beaulieu, M. S. Jha, H. Garnier, and F. Cerbah, “Remaining Useful Life prediction based on physics-informed data augmentation,” Reliab. Eng. Syst. Saf., vol. 252, p. 110451, 2024, doi: 10.1016/j.ress.2024.110451.
[5] N. Hafidhoh, A. P. Atmaja, G. N. Syaifuddiin, I. B. Sumafta, S. M. Pratama, and H. N. Khasanah, “Machine Learning untuk Prediksi Kegagalan Mesin dalam Predictive Maintenance System,” Jurnal Masyarakat Informatika, vol. 15, no. 1, pp. 56–66, May 2024, doi: 10.14710/jmasif.15.1.63641.
[6] B. Sugito, M. Iqbal, and Z. Sitorus, “Analisis Metode FMEA Dan SPC Pada Proyeksi Losses Produksi Dan Prediksi Perawatan Pompa Minyak Di PT.Pertamina Ep Zona 1 Rantau Field Kuala Simpang,” Jurnal Riset Sistem Informasi Dan Teknik Informatika (JURASIK), vol. 10, pp. 468–479, Feb. 2025, doi: 10.30645/jurasik.v10i1.889.
[7] M. Wang, X. Su, H. Song, Y. Wang, and X. Yang, “Enhancing predictive maintenance strategies for oil and gas equipment through ensemble learning modeling,” J. Pet. Explor. Prod. Technol., vol. 15, no. 3, Mar. 2025, doi: 10.1007/s13202-025-01931-x.
[8] Y. Meinard and A. Tsoukiàs, “What Is Legitimate Decision Support?,” in Intelligent Decision Support Systems: Combining Operations Research and Artificial Intelligence – Essays in Honor of Roman Słowiński, S. Greco, V. Mousseau, J. Stefanowski, and C. Zopounidis, Eds., Cham: Springer, 2022, pp. 207–224. doi: 10.1007/978-3-030-96318-7_11.
[9] U. Dereci and G. Tuzkaya, “An explainable artificial intelligence model for predictive maintenance and spare parts optimization,” Supply Chain Analytics, vol. 8, p. 100078, 2024, doi: 10.1016/j.sca.2024.100078.
[10] E. M. I. Tutu, B. Pinontoan, E. Tenda, C. A. J. Soewoeh, M. I. Takaendengan, and S. C. W. Ngangi, “Pengembangan Chatbot Perpustakaan Berbasis Retrieval-Augmented Generation (RAG) pada Telegram,” INTECH, vol. 7, no. 1, pp. 55–65, May 2026, doi: 10.54895/intech.v7i1.3435.
[11] F. A. Dzaky and D. Kurniawan, “Implementation of Scrum Framework Agile Method to Develop Integrated Asset Management Information System at Universitas Diponegoro Inventory Module,” Jurnal Masyarakat Informatika, vol. 14, no. 1, pp. 53–69, 2023, doi: 10.14710/jmasif.14.1.52605.
[12] S. Matzka, “AI4I 2020 Predictive Maintenance Dataset,” 2020, UCI Machine Learning Repository. doi: 10.24432/C5HS5C.
[13] A. C. Praniffa, A. Syahri, F. Sandes, U. Fariha, and Q. A. Giansyah, “PENGUJIAN BLACK BOX DAN WHITE BOX SISTEM INFORMASI PARKIR BERBASIS WEB,” Jurnal Testing dan Implementasi Sistem Informasi, vol. 1, no. 1, pp. 1–16, 2023, doi: 10.55583/jtisi.v1i1.321.
Additional Files
Published
Issue
Section
Categories
License
Copyright (c) 2026 Rasyid Naufal Saputra, Heni Rahmawati, Purwanto (Author)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
