Investigation of tugboat accidents severity: An application of association rule mining algorithms

ÇAKIR E., FIŞKIN R., SEVGİLİ C.

RELIABILITY ENGINEERING & SYSTEM SAFETY, cilt.209, 2021 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 209
  • Basım Tarihi: 2021
  • Doi Numarası: 10.1016/j.ress.2021.107470
  • Dergi Adı: RELIABILITY ENGINEERING & SYSTEM SAFETY
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, PASCAL, Aerospace Database, Communication Abstracts, INSPEC, Metadex, zbMATH, Civil Engineering Abstracts
  • Anahtar Kelimeler: Tugboat accidents, Data mining, Association rule, Accident severity, NEGATIVE BINOMIAL REGRESSION, MARINE ACCIDENT, VESSEL ACCIDENTS, DETERMINANTS, CASUALTIES, RISK, EVALUATE, SAFETY, MODEL
  • Dokuz Eylül Üniversitesi Adresli: Evet

Özet

This paper aims to investigate tugboat accidents using various association rule mining algorithms. A total of 477 tugboat accident records obtained from the Information Handling Services (IHS) Sea-Web database for the period of 2008-2017 were analysed. Apriori, Predictive Apriori and FP-Growth algorithms were employed to extract the association rules of the tugboat accidents dataset. The present study revealed that tugboats aged over 20 years are crucial indicators for serious accidents. Hull/machinery damage and collision type accidents, on the other hand, constitute more than half of the total tugboat accidents. Association rule mining also showed that four of the five rules for serious accidents are attributed to hull/machinery damage. The results of this study are thought to be beneficial for tugboat and ship operators, port management and public authorities regarding the awareness of the factors affecting tugboat accidents.