Key points are not available for this paper at this time.
Maritime transport, a cornerstone of global trade, faces escalating risks that necessitate systematic and data-driven assessment methods. This study proposes an integrated and intelligent framework for Maritime risk evaluation, aiming to enhance the accuracy and effectiveness of identifying high-risk scenarios. By leveraging Bayesian network models to capture probabilistic dependencies and employing genetic algorithms combined with the Best-Worst Method (BWM) to optimize criteria weighting, this research introduces a robust multi-criteria decision-making (MCDM) system. Furthermore, the rough comprehensive compromise solution (CoCoSo) method is applied to evaluate the severity of risk factors, incorporating rough set theory and the normal distribution-based Rough Ordered Weighted Averaging(ROWA) weighting approach to better address uncertainty and imprecision. Empirical results Accident quantity(A) and Large ship(L) as the most and least favorable criteria, respectively, with piracy and fire identified as the most critical risk factors. From the perspectives of policy and management, this study aims to provide operational insights for port authorities and Maritime companies to enhance safety management and operational resilience, thereby promoting the sustainable development of maritime transportation.
Bei et al. (Tue,) studied this question.