Using DBN and evidence-based reasoning to develop a risk performance model to interfere ship navigation process safety in Arctic waters

Abstract

Risk performance reasoning strategy for LNG ships navigating in Arctic waters is proposed in this paper. Many uncertainties exist in the reasoning of ship navigation risk in Arctic waters, which are influenced by multi-source risk causing events. As a result of the aforementioned concerns, the dynamic Bayesian network (DBN) structure is offered as a solution for an uncertain risk assessment model. The DBN network can benefit from a strategy for solving ambiguous data information based on Dempster–Shafer (D–S) evidence theory and a cloud model. Additionally, a risk performance reasoning technique for LNG tanker collision accidents in Arctic waters is developed. Besides, the marine meteorological reanalysis data, data from ship-borne sensor monitoring, and expert knowledge in the suggested risk performance reasoning method are incorporated. A case study confirmed that the risk performance reasoning of accidents was needed revealed that the main risk in Arctic summer waters is posed by obstacles in the channel that are difficult to detect, such as icebergs and reefs.