Pattern investigation of total loss maritime accidents based on association rule mining

Total loss of a ship is the most serious consequence of maritime accidents, which not only brings huge property losses, but causes serious human casualties and environmental pollution. Therefore, to reduce the occurrence of such accidents, this study investigates the significant patterns involved in total loss accidents using association rule technique. Based on a total of 1554 total loss accident data from 2010 to 2020 in the Information Handling Services (IHS) sea-web database, the ' Eclat algorithm is employed to extract association rules from total loss accident dataset associated with accident types and accident severity. The present study indicates that ships aged over 20 years is the key indicator of casualties in total loss accidents. Hull/machinery damage and foundered are the major types of accidents that frequently contribute to ship total loss. The results would help stakeholders involved understand the characteristics of total loss maritime accidents and put forward preventive measures to reduce the occurrence of similar accidents.

Automated summary

Total loss of a ship is the most serious consequence of maritime accidents, causing massive property losses, human casualties, and environmental pollution. This study investigates significant patterns in total loss accidents using association rule technique and finds that ship age and accident type are key indicators.