Large vessel berthing decision based on case-based reasoning

In order to solve the complexity and uncertainty problems in the berthing process of large ships, this paper develops a decision support model based on Case-Based Reasoning(CBR). This model integrates CBR technology with a cloud model and BP neural networks. It comprehensively considers multi-dimensional attributes such as vessel characteristics, meteorological and hydrological conditions, and port factors to establish a case framework comprising a basic information domain, a characteristic attribute domain, and a decision support domain. By integrating expert scoring with the cloud model, the model processes the randomness and fuzziness in expert evaluations to optimize the case attribute weights. Furthermore, it utilizes BP neural network to achieve case reuse and decision prediction, thereby reducing subjective errors introduced by manual intervention. In this paper, we collect the berthing cases of Chiwan and Shekou container terminals at Shenzhen Port for model validation. The preliminary verification model can provide relevant decision support for pilots, expand new scenarios of artificial intelligence technology in maritime applications, and provide new ideas for intelligent berthing planning of unmanned ships.