Citation

Paolo F, Gianfranco F, Luca F, Marco M, Andrea M, Francesco M, Vittorio P, Mattia P, Patrizia S. Transp. Res. Proc. 2021; 52: 252-259.

Copyright

(Copyright © 2021, Elsevier Publications)

DOI

10.1016/j.trpro.2021.01.029

PMID

unavailable

Abstract

Navigation safety is a priority both at European and global level. Despite the important progress made over the years, sea accidents remain a major concern and much work is still needed to enhance maritime safety. Knowing the causes and precursors of past accidents is essential to identify the elements on which to intervene to improve safety and reduce the possibility of an accident to occur again. In this study, 1.079 sea accidents from the International Maritime Organization (IMO) database are analyzed using Semi-supervised Recursively Partitioned Mixture Models in an attempt to identify and categorize causal themes from accident data. Special attention is devoted to the human element, which is widely recognized as a primary or precursory cause in most accidents.

23rd EURO Working Group on Transportation Meeting, EWGT 2020, 16-18 September 2020, Paphos, Cyprus

Language: en

Keywords

Clustering; Hierarchical Methods; Human Factors; Sea Accidents; Sea Safety