RISK ANALYSIS OF GROUNDING ACCIDENTS BY MAPPING A FAULT TREE INTO A BAYESIAN NETWORK

Şakar C., Töz A. C., Büber M., Köseoğlu B.

APPLIED OCEAN RESEARCH, vol.113, 2021 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 113
  • Publication Date: 2021
  • Doi Number: 10.1016/j.apor.2021.102764
  • Journal Name: APPLIED OCEAN RESEARCH
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED)
  • Keywords: Ship accident, Grounding, Risk analysis, Fault tree analysis, Bayesian network analysis, MARITIME TRANSPORTATION SYSTEMS, MARINE TRANSPORTATION, RELIABILITY-ANALYSIS, BELIEF NETWORKS, HUMAN ERROR, MODEL, COLLISION, OPERATIONS, FRAMEWORK, SPILL
  • Dokuz Eylül University Affiliated: Yes

Abstract

Groundings, which account for approximately one-third of ship accidents, are classified as very serious with tragic consequences. Therefore, identifying the factors causing groundings is essential to reduce their effects and to prevent recurrence. Thus, the principal purpose of this study is to analyse the risks of grounding accidents using an integrated model. First, a fault tree analysis (FTA) was applied to create a risk hierarchy that defines the level of relationship among factors; and a Bayesian network (BN) analysis was conducted to evaluate the level of their effect on groundings. The study found that of the 34 factors related to the safety of navigation, E10/violation of COLREG Rule-5 (look-out), E9/improper use of ECDIS, and E8/improper use of radar significantly effect groundings. The findings also indicate that maintaining a safe navigation watch is crucial to prevent future accidents. This study makes a significant contribution to the relevant literature as one of the few pieces of research using the combined FTA-BN as a dynamic risk assessment methodology for analysis of grounding accidents. We recommend that further research use other risk assessment methods that consider different ship types for a better comparison.