Analysis of collision accidents onboard ships navigating with pilots
Journal of Transportation Safety and Security, vol.18, no.3, pp.371-400, 2026 (SSCI, Scopus)
- Publication Type: Article / Article
- Volume: 18 Issue: 3
- Publication Date: 2026
- Doi Number: 10.1080/19439962.2025.2562577
- Journal Name: Journal of Transportation Safety and Security
- Journal Indexes: Social Sciences Citation Index (SSCI), Scopus, Academic Search Premier, Civil Engineering Abstracts
- Page Numbers: pp.371-400
- Keywords: Association rule mining, collision, maritime accidents, maritime pilot
- Karadeniz Technical University Affiliated: Yes
Abstract
This study aimed to investigate the contributing factors to collision accidents onboard ships navigating with pilots. The Human Factors Analysis and Classification System was used to classify the causal factors extracted from 44 accident reports produced by various marine accident investigation branches. The reports yielded 286 contributing factors. After classifying the contributing factors and identifying the root causes, Association Rule Mining (ARM) was applied to find connections between the underlying causes. Nine significant rules were identified after determining the threshold values using expert judgment from a scatter graph in an ARM network. First, the study indicated a major correlation between teamwork and pilot–ship crew communication problems. The second most significant finding involved hydrodynamic effects on container ships, impacting maneuver prediction. The third finding indicated a significant relationship between ship-to-ship communication and situational awareness, particularly notable on middle-aged ships. The analysis of the ARM network unveiled new interrelations overlooked in previous research, offering a more comprehensive insight relative to the current literature.