A fuzzy Bayesian network risk assessment model for analyzing the causes of slow-down processes in two-stroke ship main engines

Abstract

ABSTRACT This paper presents a risk assessment approach for analyzing the causes of malfunction-related main engine slowdowns. A fuzzy Bayesian Network-based methodology is used to assess the factors contributing to the engine’s slow-down processes. The model addresses the complexity and uncertainty inherent in maritime operations with fuzzy sets where numerous interrelated factors can affect engine performance, and the Bayesian network to capture probabilistic dependencies. It considers various potential causes of the slow-down of ship engines that the manufacturer provides. Results demonstrate the model's ability to identify the influential factors leading to engine slow-down events and quantify the overall risk. Integrating fuzzy logic and Bayesian Networks comprehensively assesses relevant risk factors. It enables maritime stakeholders to manage engine performance and improves operational safety proactively. Findings can inform decision-makers, enabling the implementation of targeted maintenance strategies, fuel quality control measures, and crew training programs in the maritime industry.