COVID crisis-aware maritime risk assessment: A Bayesian network analysis

Maritime transportation is a vital component of global trade, yet maritime accidents pose significant risks with far-reaching consequences, including human casualties, economic losses, and environmental damage. The high-risk nature of this sector calls for in-depth, data-driven analysis to enhance risk assessment and accident prevention. While traditional approaches such as probabilistic risk analysis have advanced the understanding of maritime safety, they often overlook the evolving nature of risk under global crises, such as the COVID-19 pandemic (2020), the Ever Given blockage in the Suez Canal (March 2021), ongoing geopolitical conflicts (e.g., Russia-Ukraine since 2022), and the recent Red Sea crisis (2024). To overcome this critical research gap, this study proposes a crisis-aware maritime risk assessment framework based on Bayesian Network (BN), operationalised through a Tree-Augmented Naïve Bayes (TAN) model, using the COVID-19 pandemic as a case study. By analysing maritime accident patterns before and after the pandemic, the model reveals shifts in accident dynamics and emerging risk factors. The BN approach enables objective, interpretable analysis of how underlying causes and safety interventions have evolved in response to the crisis. Additionally, this study indirectly assesses the effectiveness of safety measures implemented during the pandemic and highlights areas for improvement to enhance future resilience. The findings provide actionable insights for policymakers, regulators, and industry stakeholders, supporting the development of more adaptive and robust maritime safety strategies to address future global disruptions.