Real-time collision risk based safety management for vessel traffic in busy ports and waterways

Journal article (2023)

Authors

M. Li Safety and Security Science - TPM , Hubei Key Laboratory of Inland Shipping Technology, Wuhan University of Technology
Junmin Mou Hubei Key Laboratory of Inland Shipping Technology, Wuhan University of Technology
Pengfei Chen Wuhan University of Technology, Hubei Key Laboratory of Inland Shipping Technology
Linying Chen Wuhan University of Technology, Hubei Key Laboratory of Inland Shipping Technology
P.H.A.J.M. van Gelder Safety and Security Science - TPM
Research Group
Safety and Security Science (TPM) (TU Delft)
Copyright: © 2023 M. Li, Junmin Mou, Pengfei Chen, Linying Chen, P.H.A.J.M. van Gelder
DOI
help_outline
https://doi.org/10.1016/j.ocecoaman.2022.106471
AIS Risk assessment and management Maritime safety management NLVO Non-accident critical events Ship collision accidents
More Info
expand_more

Published Date

2023

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Technology, Policy and Management

Department

Values Technology and Innovation

Research Group

Safety and Security Science

Abstract

Regional risk analysis and management of maritime accidents is one of the fundamental tasks for maritime safety management. With the heavy and complicated maritime traffic in the ports and waterways, accidents, especially ship collision accidents, have been continuously posing threats to the maritime transportation system. To achieve effective and prompt identification of collision risk and to facilitate the stakeholders such as Maritime Safety Administration, this paper proposes an integrated approach for regional collision risk analysis and maritime safety management in busy ports and waterways. Firstly, regional gridding is used to link accident data and traffic data based on geographical location; Secondly, the risk model based on accident data is established. The reliability of the accident risk model is verified by data feature analysis. Finally, non-accident critical events are mined from historical accident data and traffic data as surrogate indicators of collision accidents. A regional real-time risk model is developed for integrating the accident risk model and non-accident critical events risk model by using random forest. A case study in Shenzhen port indicates that the proposed collision risk model can identify high-risk areas and facilitates maritime safety management to improve the safety level of vessel traffic in these areas. In this paper, the regional grid is used to overcome the shortcomings of different scales between data, and a real-time risk model is established by combining accidents and traffic. The 15-year maritime collision accidents are used for collision risk modeling, which improves the performance of the model.