Identifying critical risk influencing factors for autonomous ship navigation in winter conditions

Abstract The shipping industry’s transition towards greener and more autonomous solutions is crucial for achieving maritime sustainability and enhancing operational performance. This shift necessitates assessing the risks introduced by these innovations, particularly in the challenging winter conditions of the Baltic Sea. This study proposes a framework to identify the critical risk influencing factors (RIFs) affecting the combined use case of Maritime Autonomous Surface Ships (MASS) and winter navigation. The framework applies the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology to conduct a systematic literature review (SLR) and identify relevant RIFs, which are refined using expert input. The model uses pairwise comparisons to estimate the relative occurrence probability of RIFs. These values are then used to develop a Bayesian Network (BN) model, producing a ranked list of critical risks and their interdependencies. Furthermore, the uncertainty assessment of the BN model indicated that the model’s uncertainty is moderate, primarily due to data limitations. The model’s capability to identify high-risk situations provides valuable insight for decision-makers, supporting the safe integration of autonomous technologies and sustainable maritime development. The results highlight the need for future empirical validation to enhance risk assessments to support the safe and sustainable growth of autonomous shipping.