Impact of voyage segments on maritime accidents: An analysis of navigational factors and accident causes

Maritime transportation, a cornerstone of global trade, faces significant risks from maritime accidents, which can result in severe human casualties, substantial property loss, and extensive environmental damage. This study aims to improve the understanding of how different voyage segments, coastal waters, open seas, and restricted waters, influence maritime accidents by systematically analysing navigational characteristics and Risk Influential Factors (RIFs) across segments. The study employs a Tree-Augmented Naïve Bayes (TAN) model to quantify the probabilistic influence of RIFs on accident occurrence, enabling the explicit modelling of interdependencies that traditional approaches fail to capture. Scenario analysis is further conducted to assess segment-specific accident patterns and to identify how operational, environmental, and human-centred factors vary across navigational contexts. The results reveal both shared and segment-unique root causes, as well as high-risk transition zones where accident likelihood changes markedly between segments. By integrating voyage-segment analysis with a TAN structure, this paper advances maritime accident modelling beyond prior applications and provides actionable insights for risk-informed decision-making. The findings support the optimisation of route planning, the design of segment-specific and transition-focused safety measures, and the development of more effective maritime safety management strategies across diverse operational environments.