Comparative analysis of risk factors affecting maritime accidents across voyage segments
Comparative analysis of risk factors affecting maritime accidents across voyage segments
Maritime transportation is a cornerstone of global trade, facilitating the movement of goods and resources across the world. However, it remains highly vulnerable to accidents, which can lead to devastating human, economic, and environmental consequences. This study presents an innovative application of the Tree Augmented Naive Bayes (TAN) model within the framework of Bayesian Networks to assess the impact of 23 Risk Influencing Factors (RIFs) on maritime accidents across three critical voyage segments: coastal waters, open seas, and restricted waters. By incorporating scenario and sensitivity analyses, the proposed method provides a comprehensive understanding of how specific navigational conditions contribute to distinct accident types, such as collisions, groundings, and machinery failures. The findings offer dynamic risk assessment tools that support optimised route planning and targeted accident prevention strategies, tailored to varying voyage contexts. This paper highlights the influence of environmental, operational, and human factors on maritime safety, emphasising the need for proactive and data-driven decision-making in risk management. Furthermore, the study demonstrates the enhanced predictive performance of the TAN model, which allows stakeholders to prioritise interventions and allocate resources more effectively. By advancing the integration of machine learning and Bayesian inference techniques in maritime safety, this research provides actionable insights for policymakers, operators, and safety regulators. The outcomes not only contribute to reducing accident rates but also support the development of sustainable and resilient maritime operations. This work represents a significant step forward in maritime risk analysis, offering a robust framework for enhancing safety, minimising losses, and safeguarding the marine environment.
Wu, Yiheng
0e074f4c-1ed4-4fdf-84e9-6ea1e8a3b1fd
Li, Huanhuan
5e806b21-10a7-465c-9db3-32e466ae42f1
Jiao, Hang
af70e929-ff8d-4020-92a5-34c55f6aba7c
Yang, Zaili
e8ff5fad-c312-4643-8e8b-55d2697fed3a
16 July 2025
Wu, Yiheng
0e074f4c-1ed4-4fdf-84e9-6ea1e8a3b1fd
Li, Huanhuan
5e806b21-10a7-465c-9db3-32e466ae42f1
Jiao, Hang
af70e929-ff8d-4020-92a5-34c55f6aba7c
Yang, Zaili
e8ff5fad-c312-4643-8e8b-55d2697fed3a
Wu, Yiheng, Li, Huanhuan, Jiao, Hang and Yang, Zaili
(2025)
Comparative analysis of risk factors affecting maritime accidents across voyage segments.
In 2025 8th International Conference on Transportation Information and Safety (ICTIS).
IEEE..
(doi:10.1109/ictis68762.2025.11215050).
Record type:
Conference or Workshop Item
(Paper)
Abstract
Maritime transportation is a cornerstone of global trade, facilitating the movement of goods and resources across the world. However, it remains highly vulnerable to accidents, which can lead to devastating human, economic, and environmental consequences. This study presents an innovative application of the Tree Augmented Naive Bayes (TAN) model within the framework of Bayesian Networks to assess the impact of 23 Risk Influencing Factors (RIFs) on maritime accidents across three critical voyage segments: coastal waters, open seas, and restricted waters. By incorporating scenario and sensitivity analyses, the proposed method provides a comprehensive understanding of how specific navigational conditions contribute to distinct accident types, such as collisions, groundings, and machinery failures. The findings offer dynamic risk assessment tools that support optimised route planning and targeted accident prevention strategies, tailored to varying voyage contexts. This paper highlights the influence of environmental, operational, and human factors on maritime safety, emphasising the need for proactive and data-driven decision-making in risk management. Furthermore, the study demonstrates the enhanced predictive performance of the TAN model, which allows stakeholders to prioritise interventions and allocate resources more effectively. By advancing the integration of machine learning and Bayesian inference techniques in maritime safety, this research provides actionable insights for policymakers, operators, and safety regulators. The outcomes not only contribute to reducing accident rates but also support the development of sustainable and resilient maritime operations. This work represents a significant step forward in maritime risk analysis, offering a robust framework for enhancing safety, minimising losses, and safeguarding the marine environment.
This record has no associated files available for download.
More information
Accepted/In Press date: April 2025
Published date: 16 July 2025
Venue - Dates:
International Conference on Transportation Information and Safety, , Granada, Spain, 2025-07-16 - 2025-07-19
Identifiers
Local EPrints ID: 511420
URI: http://eprints.soton.ac.uk/id/eprint/511420
PURE UUID: 8b71cbfc-c1df-417a-a780-ac4e07789621
Catalogue record
Date deposited: 14 May 2026 16:35
Last modified: 21 May 2026 02:13
Export record
Altmetrics
Contributors
Author:
Yiheng Wu
Author:
Huanhuan Li
Author:
Hang Jiao
Author:
Zaili Yang
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics