Fault Diagnosis of Marine Engine Systems Using Optimized XGBoost with Grid Search Hyperparameter Tuning
DOI:
https://doi.org/10.33795/jartel.v16i1.9639Keywords:
Fault Diagnosis, Marine Engine Condition Monitoring, Machine Learning, XGBoost, Grid SearchAbstract
The reliability of the main engine is a critical factor in ensuring vessel safety and operational effectiveness, particularly in the maritime industry, which requires continuous transportation services. Conventional maintenance approaches, such as time-based maintenance and corrective maintenance, remain widely applied; however, they often fail to provide early detection of potential failures. Disturbances such as overheating, lubrication degradation, fuel contamination, and turbocharger malfunction can significantly reduce engine performance and may result in unexpected operational shutdowns. Therefore, data-driven approaches based on machine learning are needed to improve monitoring capability and prediction of engine conditions. This study develops a ship engine condition classification model using the XGBoost algorithm based on multivariate data. The model training was conducted using five hyperparameter configurations with three train–test data split schemes (0.2, 0.3, and 0.5). Experimental results demonstrate stable performance, with accuracy ranging from 0.843 to 0.850, and improved performance as the proportion of training data increased. Models incorporating regularization showed better generalization capability when trained on larger datasets. Furthermore, hyperparameter optimization using Grid Search produced the optimal configuration for the final model. The findings indicate that XGBoost effectively handles complex sensor data and has strong potential to support reliable early fault detection in ship engines
References
W. Busse, N. Siswantoro, and M. N. Bintang, “Dynamic Information System for Failure Analysis with Its Application on Ship Main Engine,” International Journal of Marine Engineering Innovation and Research, vol. 8, no. 3, p. 570, 2023.
A. Ali and A. Abdelhadi, “Condition-Based Monitoring and Maintenance: State of the Art Review,” Jan. 01, 2022, MDPI. doi: 10.3390/app12020688.
Y. Gholipour, M. Zare, M. Vaziri Sereshk, and Y. Gholipour, “A Comprehensive Review of Maintenance Strategies: From Reactive to Proactive Approaches,” Central Asia and the Caucasus, vol. 26, 2025.
R. Gagana Erwin Asmara, T. Sistem Perkapalan, F. Teknik dan Ilmu Kelautan, and U. Hang Tuah Surabaya, “Analisa Kegagalan Sistem Bahan Bakar Kapal dengan Menggunakan Metode Preliminary Hazard Analysis (PHA) dan Fault Tree Analysis (FTA),” HEXAGON, vol. 3, 2022.
P. Prasetyo, A. Seno, and P. Pelayaran Sumatera Barat, “Analisis Penyebab Terjadinya Overheat pada Main Engine di Kapal Self Propelled Oil Barge Tirta Samudra XVIII,” Jurnal Cakrawala Bahari, vol. 5, no. 2, pp. 5–10, 2022. [Online]. Available: http://jurnal.poltekpelsumbar.ac.id/index.php/jcb
M. H. Park and W. J. Lee, “Comprehensive Review of Shipboard Maintenance Management Strategies,” Sep. 01, 2025, Elsevier B.V. doi: 10.1016/j.rineng.2025.106671.
M. Mursidi and A. Sarjito, “Implementation Strategy of Ship Engine Maintenance Management System to Improve Operational Efficiency,” Jurnal Aplikasi Pelayaran dan Kepelabuhanan, vol. 15, no. 2, pp. 201–214, Feb. 2025, doi: 10.30649/japk.v15i2.137.
A. S. Kalafatelis, N. Nomikos, A. Giannopoulos, G. Alexandridis, A. Karditsa, and P. Trakadas, “Towards Predictive Maintenance in the Maritime Industry: A Component-Based Overview,” Mar. 01, 2025, MDPI. doi: 10.3390/jmse13030425.
T. Kirketerp-Møller, M. W. Hyldgaard, J. Cai, A. I. Dodis, and N. G. M. Rytter, “Data-Driven Predictive Maintenance for Two-Stroke Marine Diesel Engines Using Machine Learning and MLOps,” Journal of Ocean Engineering and Science, 2025, doi: 10.1016/j.joes.2025.11.011.
I. T. Jolliffe and J. Cadima, “Principal Component Analysis: A Review and Recent Developments,” Apr. 13, 2016, Royal Society of London. doi: 10.1098/rsta.2015.0202.
L. van der Maaten and G. Hinton, “Visualizing Data Using t-SNE,” 2008.
Z. Qinghe, X. Wen, H. Boyan, W. Jong, and F. Junlong, “Optimised Extreme Gradient Boosting Model for Short-Term Electric Load Demand Forecasting of Regional Grid System,” Scientific Reports, vol. 12, no. 1, Dec. 2022, doi: 10.1038/s41598-022-22024-3.
J. H. Friedman, “Greedy Function Approximation: A Gradient Boosting Machine,” The Annals of Statistics, vol. 29, no. 5, pp. 1189–1232, 2001.
T. Chen and C. Guestrin, “XGBoost: A Scalable Tree Boosting System,” Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Aug. 2016, pp. 785–794, doi: 10.1145/2939672.2939785.
Q. A. Hidayaturrohman and E. Hanada, “A Comparative Analysis of Hyper-Parameter Optimization Methods for Predicting Heart Failure Outcomes,” Applied Sciences (Switzerland), vol. 15, no. 6, Mar. 2025, doi: 10.3390/app15063393.
P. Calle et al., “Integration of Nested Cross-Validation, Automated Hyperparameter Optimization, High-Performance Computing to Reduce and Quantify the Variance of Test Performance Estimation of Deep Learning Models,” Computer Methods and Programs in Biomedicine, vol. 272, Dec. 2025, doi: 10.1016/j.cmpb.2025.109063.
