Highlights A binary logistic regression model is developed to describe shipping accident injury severity. A zero-truncated binomial regression model is developed to estimate the mortalities resulting from fatal shipping accidents. Both the probability of fatal accidents and mortalities are greater for accidents occurred in adverse weather and darkness conditions. Sinking has the largest effects on the increment of fatal accident probability and mortalities.
Abstract Shipping movements are operated in a complex and high-risk environment. Fatal shipping accidents are the nightmares of seafarers. With ten years’ worldwide ship accident data, this study develops a binary logistic regression model and a zero-truncated binomial regression model to predict the probability of fatal shipping accidents and corresponding mortalities. The model results show that both the probability of fatal accidents and mortalities are greater for collision, fire/explosion, contact, grounding, sinking accidents occurred in adverse weather conditions and darkness conditions. Sinking has the largest effects on the increment of fatal accident probability and mortalities. The results also show that the bigger number of mortalities is associated with shipping accidents occurred far away from the coastal area/harbor/port. In addition, cruise ships are found to have more mortalities than non-cruise ships. The results of this study are beneficial for policy-makers in proposing efficient strategies to prevent fatal shipping accidents.
Investigation of shipping accident injury severity and mortality
Accident Analysis and Prevention ; 76 ; 92-101
2015-01-04
10 pages
Article (Journal)
Electronic Resource
English
Investigation of shipping accident injury severity and mortality
Online Contents | 2015
|Reliability of Statistical Road Accident Injury Severity Models
Transportation Research Record | 1996
|Reliability of Statistical Road Accident Injury Severity Models
Transportation Research Record | 2019
|Reliability of Statistical Road Accident Injury Severity Models
British Library Conference Proceedings | 1996
|Reliability of Statistical Road Accident Injury Severity Models
Online Contents | 1996
|