Automatic Identification System-Based Approach for Assessing the Near-Miss Collision Risk Dynamics of Ships in Ports

• Published in: IEEE transactions on intelligent transportation systems Vol. 20; no. 2; pp. 534 - 543
• Main Authors: Fang, Zhixiang, Yu, Hongchu, Ke, Ranxuan, Shaw, Shih-Lung, Peng, Guojun
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.02.2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accidents; Artificial intelligence; Collision dynamics; Estimation; Geographical distribution; Geospatial analysis; Identification systems; marine safety; marine transportation; Marine vehicles; Mathematical model; Oceans; Ports; Risk analysis; Safety management; Ships; time series analysis; Traffic accidents & safety; Traffic flow; Traffic planning
• ISSN: 1524-9050; 1558-0016
• DOI: 10.1109/TITS.2018.2816122

Vessel risk analysis is critical for safe ship navigation and maritime safety management. Near-miss collisions by ships comprise an significant risk, which may be complicated by factors, such as the ship conditions, waterway environment, and driving behavior of any ships encountered. Previous studies have rarely considered how to automatically and adaptively estimate the risk of near-miss collisions for different situations, particularly in port areas. In this paper, we propose an automatic identification system-based approach for adaptively calibrating near-miss collision risk model and assessing a ship's near-miss collision risk by using the vessel's speed and course patterns to obtain a robust estimate of the collision risk. Six measures are employed to determine the hierarchical geographical distribution of the near-miss collision risk for ships in port areas and to identify the high-risk areas. Some predicted high-risk areas were validated as official precautionary areas in the Xiamen Port area. All predicted areas may help the port administration to plan monitoring areas to ensure safe traffic flow in the port.