Artificial Intelligence–Based Predictive Maintenance Framework for Marine Propulsion Systems

Abstract

Maritime transport has adapted to recent political and economic shifts by addressing stringent pollution reduction requirements, redrawing transport routes for safety, reducing onboard technical incidents, managing data security risks and transitioning to autonomous vessels. This paper presents a novel approach to predictive maintenance in the maritime industry, leveraging Artificial Intelligence (AI) and Machine Learning (ML) techniques to enhance fault detection and maintenance planning for naval systems.  Marine propulsion systems are critical to the safe and efficient operation of ships, and unexpected failures can lead to significant operational downtime, increased maintenance costs, and safety risks. Conventional maintenance strategies are largely time-based or reactive, limiting their effectiveness under varying operating conditions. This paper proposes an artificial intelligence–based predictive maintenance framework for marine propulsion systems using multi-sensor operational data. The proposed framework integrates data acquisition from key propulsion components, signal preprocessing, and intelligent condition assessment using machine learning techniques. An AI model is developed to automatically learn fault-related patterns and degradation trends, enabling early fault detection and accurate health state prediction. Experimental results demonstrate that the proposed approach improves fault diagnosis accuracy and supports condition-based maintenance decisions when compared with traditional maintenance strategies. The findings highlight the potential of artificial intelligence to enhance reliability, reduce unplanned downtime, and optimize maintenance planning in marine propulsion systems.