Regional collision risk identification and prediction is important for traffic surveillance in maritime transportation. This study proposes a framework of real-time prediction for regional collision risk by combining Density-Based Spatial Clustering of Applications with Noise (DBSCAN) technique, Shapley value method and Recurrent Neural Network (RNN). Firstly, the DBSCAN technique is applied to cluster vessels in specific sea area. Then the regional collision risk is quantified by calculating the contribution of each vessel and each cluster with Shapley value method. Afterwards, the optimized RNN method is employed to predict the regional collision risk of specific seas in short time. As a result, the framework is able to determine and forecast the regional collision risk precisely. At last, a case study is carried out with actual Automatic Identification System (AIS) data, the results show that the proposed framework is an effective tool for regional collision risk identification and prediction.

Ship density is widely accepted as a basic and major indicator to reflect the marine traffic situation, but it has some limitations in representing the compactness and complexity of ship traffic. To overcoming these limitations, the paper proposes a novel ship density model based on the radial distribution function in molecular dynamics. The proposed model can identify the density and compactness of traffic around each ship and then map the ship density from a microscopic perspective. In addition, the proposed model can identify the global density and the complexity of ship traffic to some extent in the macroscopic perspective. Utilising case studies, the effectiveness of the proposed model is validated through the analysis of ship density in several regions in the Bohai Strait area. The proposed model is developed to help marine surveillance operators gain a better understanding of the traffic situation and to assist them in their work, eventually contributing to navigational safety.

During the process of establishing the analytical quaternion ship domain model, the impact of ship traffic conditions and navigator state was not taken into consideration. However, the significance of these factors in the ship domain cannot be ignored. To create a more realistic representation of changes in the ship domain in real navigation environments, this study further considers the influence of ship encounter course, waterway traffic intensity, relative ship velocity, and the navigator state based on the quaternion ship domain model. As a result, a new dynamic quaternion ship domain model is proposed. To assess the changes in the size and shape of the ship domain under various navigation environments, ship domain scaling and shape transformation functions are introduced. Specifically, this study focuses on analyzing the ship traffic near the Lao Tie Shan Waterway, simulating the size and shape changes of the ship domain during the navigation process in this area. The findings indicate that the size of the ship domain dynamically adjusts to the traffic conditions. Additionally, when the navigator state is excellent, the ship domain takes on an irregular diamond shape with the smallest area, whereas when the navigator state is poor, the shape approximates a rectangle with the largest area. Furthermore, the dynamic quaternion ship domain model proposed in this study is compared to the ship domain models put forth by Goodwin, Davis, and co-authors. The results demonstrate that the dynamic quaternion ship domain model is more compatible and suitable for open waters compared to the static quaternion ship domain model.

With the rapid development of the shipping industry in recent years, the increasing volume of ship traffic makes marine traffic much busier and more crowded, especially in the waterway off the coast. This leads to the increment of the complexity level of marine traffic and poses more threats to marine traffic safety. In order to study marine traffic safety under the conditions of increasing complexity, this article proposed a marine traffic complexity model based on the method in molecular dynamics. The model converted ship traffic to a particle system and identified the traffic complexity by analyzing the radial distribution of dynamic and spatial parameters of ships in a Euclid plane. The effectiveness of the proposed model had been validated by the case studies in the waters of Bohai Strait with real AIS (Automatic Identification System) data and simulated data. The results show that the proposed model can evaluate the marine traffic complexity more sufficiently and accurately. The proposed model is helpful for marine surveillance operators to monitor and organize marine traffic under complex situations so as to improve marine traffic safety.