Ship fire and explosion accident statistical analysis based on fault tree and Bayesian network

Abstract

To identify the potential causal factors of ship fire and explosion (F&E) accidents and develop effective prevention strategies, this study conducted a statistical analysis of 271 investigation reports on ship F&E accidents and identified 46 basic events. By combining Fault Tree Analysis (FTA) and Bayesian Networks (BN), six key causal factors were recognized: fuel leakage, wooden structures, plastics and rubber, main engine hot surfaces, short circuits, and line faults. This study also revealed two main and two secondary accident paths. The two main accident paths are as follows: one is for fuel or lubricant leakage contact with high-temperature surfaces (e.g., main engine, exhaust system, and turbocharger) to start a fire; the other is for short circuits or line faults to ignite surrounding combustible materials (e.g., wooden structures, plastics, and rubber). Finally, this study proposes measures to prevent ship F&E accidents in the following three aspects: strengthening the management of combustible materials, paying attention to ignition source control, and improving accident path monitoring.