Identifying influential ships in multi-ship encounter situation complex network based on improved WVoteRank approach

Abstract

The ship encounter situation perception is of great significance to improving the safety and efficiency of the Vessel Traffic Service (VTS). The traffic complexity and complex network theories are the critical techniques for evaluating the influential ships in the multi-ship encounter. This paper proposes an influential ships identification method based on improved WVoteRank approach. It considers ships' spatial compactness and risk connectivity and divides the ships in the area into multi-ship encounter clusters. At the same time, the complex network of ship encounter situation is constructed, and the improved WVoteRank approach is introduced to identify influential ships. An evaluation is carried out through the AIS data of Ningbo-Zhoushan port in China. The experimental results reveal that the approach: 1) can effectively identify the ship encounter clusters and influential ships, which simultaneously helps VTS further grasp the spatio-temporal characteristics of complex traffic areas distribution. 2) the improved WVoteRank approach is practical and advanced for multi-ship encounter situation complex network. Therefore, it provides new insights into enhancing marine traffic monitoring capabilities, helping VTS to improve the efficiency and intelligence of marine traffic services.