In the present study a comprehensive investigation of the potential for integrating artificial intelligence methods into the predictive maintenance system of marine machinery is conducted. The application of AI in this context emerges as one of the key factors contributing to the enhancement of reliability and safety in shipping. The aim of the study is to compare the performance of various machine learning and deep learning algorithms based on an analysis of contemporary scientific publications. As a result of the comparative analysis a multifactorial methodology for selecting the most suitable model has been developed taking into account not only prediction accuracy but also the volume and quality of the input data computational costs and the transparency of the obtained conclusions. It is shown that hybrid approaches — in particular when convolutional neural networks are used for feature extraction from vibration and acoustic signals and LSTM networks are used for time series analysis — demonstrate the highest accuracy in predicting the remaining useful life of critically important equipment (main engines generators etc.). The scientific novelty of the work lies in the proposal of an integrated framework for the selection of AI solutions that ensures a balanced consideration of accuracy computational complexity data requirements and interpretability of results. In conclusion it is substantiated that the transition to object-oriented technical maintenance based on AI enables a substantial reduction in operating costs and failure risks compared to reactive and planned preventive strategies. The obtained conclusions are of practical interest to researcher’s shipbuilding engineers and data analysis specialists engaged in the development of intelligent monitoring systems.

predictive maintenance, artificial intelligence, marine engineering, machine learning, deep learning, fault diagnosis, remaining useful life, marine equipment, digital twin, vibration analysis

United Nations Conference on Trade and Development (UNCTAD). Review of Maritime Transport 2023. [Electronic resource]. - Access mode: https://unctad.org/publication/review-of-maritime-transport-2023 (date of access: 24.06.2024).

Je-Gal H. et al. Time-Series Explanatory Fault Prediction Framework for Marine Main Engine Using Explainable Artificial Intelligence //Journal of Marine Science and Engineering. – 2024. – Vol. 12 (8). https://doi.org/10.3390/jmse12081296.

Jimenez V. J., Bouhmala N., Gausdal A. H. Developing a predictive maintenance model for vessel machinery //Journal of Ocean Engineering and Science. – 2020. – Vol. 5 (4). – pp. 358-386. https://doi.org/10.1016/j.joes.2020.03.003.

Fu L., Zhang L., Tao J. An improved deep convolutional neural network with multiscale convolution kernels for fault diagnosis of rolling bearing //IOP Conference Series: Materials Science and Engineering. – IOP Publishing, 2021. – Vol. 1043 (5). – pp. 1-11. https://doi.org/10.1088/1757-899X/1043/5/052021.

Maione F. et al. A Machine Learning Framework for Condition-Based Maintenance of Marine Diesel Engines: A Case Study //Algorithms. – 2024. – Vol. 17 (9). https://doi.org/10.3390/a17090411.

Tang S. et al. Deep transfer learning strategy in intelligent fault diagnosis of rotating machinery //Engineering Applications of Artificial Intelligence. – 2024. – Vol. 134. https://doi.org/10.1016/j.engappai.2024.108678.

Han P. et al. Fault detection with LSTM-based variational autoencoder for maritime components //IEEE Sensors Journal. – 2021. – Vol. 21 (19). – pp. 21903-21912. https://doi.org/10.1109/JSEN.2021.3105226.

Wen Q. et al. Transformers in time series: A survey //arXiv preprint arXiv:2202.07125. – 2022. - pp. 1-9. https://doi.org/10.48550/arXiv.2202.07125.

Hasan A. et al. Leveraging digital twins for fault diagnosis in autonomous ships //Ocean Engineering. – 2024. – Vol. 292. https://doi.org/10.1016/j.oceaneng.2023.116546.

Digitalization in the maritime industry. [Electronic resource]. – Access mode: https://www.dnv.com/maritime/insights/topics/digitalization-in-the-maritime-industry/ (date of access: 18.05.2025).

Cheliotis M., Lazakis I., Theotokatos G. Machine learning and data-driven fault detection for ship systems operations //Ocean Engineering. – 2020. – Vol. 216. https://doi.org/10.1016/j.oceaneng.2020.107968.

Wang K., Hu Q., Liu J. Digital twin-driven approach for process management and traceability towards ship industry //Processes. – 2022. – Vol. 10 (6). https://doi.org/10.3390/pr10061083.

Kaklis D. et al. Enabling digital twins in the maritime sector through the lens of AI and industry 4.0 //International Journal of Information Management Data Insights. – 2023. – Vol. 3 (2). https://doi.org/10.1016/j.jjimei.2023.100178.

What is predictive maintenance? . [Electronic resource]. – Access mode: https://www.ibm.com/topics/predictive-maintenance (date of access: 05/22/2025).