Collision Risk Hotspot Identification Based on AIS Data and DBSCAN Algorithm

The objective of the present study is to identify ship near-miss areas and identify the hotspot waters with higher collision risks. The Automatic Identification System (AIS) data from the waters near Zhoushan Island is used in the study. The ship trajectory is interpolated to extract near-misses between ships based on data at the same time. Then, risk modeling is carried out based on the relative parameters between the encountered ships. Finally, the near misses are clustered using DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm. The degree of collision risk is quantified using Collision Risk Index (CRI) models, which is also involved in the DBSCAN algorithm. The results describe the distribution of collision risky sites in the waters near Zhoushan Island. The research results can be used for safe navigation of ships and the development of near classified maps to prevent collisions between passing ships, for deck personnel to take preventive measures when passing through hazardous areas.