Prediction of ship incidents based on BP neural network

This paper investigates the use of a Back-Propagation Neural Network (BPNN) to identify the correlation between relevant parameters and ship collisions. Subsequently, a model is constructed to forecast the number of collisions over the upcoming three years. This model is expected to provide insight into the prevalence of maritime incidents and aid in the formulation of preventative measures. In this paper, BP neural network is used to study the water incidents provided by TSB (The Transportation Safety Board of Canada). The purpose of this paper is to predict the number of collisions in the future from the macro level. To achieve this goal, this paper establishes a data model based on Python. The results of this study can help researchers and channel management agencies better understand collision risk.