Abstract
Maritime incidents, though rare, have a significant impact on both the global economy and the environment. Improving maritime navigation practices always requires new ways to enhance safety, inevitably involving learning from past experiences and mistakes. In this context, probabilistic analysis of incidents and their associated consequences can play a crucial role in creating a safer and more efficient maritime transportation system. Bayesian networks constitute a class of probabilistic models based on statistics, decision theory, and graph theory. This paper describes the analysis of maritime incident statistics by selecting probabilistic parameters influencing the risk of their occurrence. Important parameters from this database are grouped, and a Bayesian network is constructed to illustrate the relationships between them. This, in turn, provides insight into the dependencies existing among the variables in the database and the fundamental reasons for these accidents. The data for this study are based on the Lloyds Register and IMO incident databases from 1990 to 2022. Key factors from this database are grouped, and a Bayesian network is built to show the relationships between the corresponding variables, providing an understanding of the probabilistic dependencies among the variables in the database and the primary causes of these incidents.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have