ABSTRACT
With the continuous development of large-scale, high-speed and professional ships, and the increasing construction of modern intelligent deep-water ports, the safety of inbound merchant ships receives more and more attention. Thus, it is of great significance for the stakeholders to address the safety management and realize the intelligent decision-making of the inbound merchant ships properly. In this paper, a novel algorithm for modeling human decision-making of inbound merchant ships is proposed. This method can be used to realize the automatic acquisition and representation of the crew's decision-making knowledge in inbound merchant ships analysis. To verify the performance of the model, a case study based on this method is conducted in the Waigaoqiao Phase IV Port of Shanghai. The experimental results indicate that the maneuvering decision recognition model combined with the method of classification interval division, which proposed in this article, possesses a high reasoning speed and can accurately and scientifically standardize the boundary of the interval of influencing factor data and identify current maneuvering behavior. The proposed methods and the evaluation results provide useful insights for effective safety management of the inbound merchant ships.
