Analysis of Ship Accidents in the Istanbul Strait Using Neuro-Fuzzy and Genetically Optimised Fuzzy Classifiers


Erol S., Demir M., Cetisli B., Eyüboğlu E.

JOURNAL OF NAVIGATION, vol.71, no.2, pp.419-436, 2018 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 71 Issue: 2
  • Publication Date: 2018
  • Doi Number: 10.1017/s0373463317000601
  • Journal Name: JOURNAL OF NAVIGATION
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.419-436
  • Keywords: Accident Analysis, Istanbul Strait, Adaptive Neuro-Fuzzy Classifier, Maritime Transportation, Genetically Optimized Fuzzy Classifier, HUMAN ERROR, LINGUISTIC HEDGES, VESSEL ACCIDENTS, RISK ANALYSIS, COLLISION, NETWORK, SYSTEM, MODEL, SIMULATION, CASUALTIES
  • Karadeniz Technical University Affiliated: Yes

Abstract

Marine accident analysis is important for ships passing through narrow, shallow and busy waterways. This study analyses the accidents which have occurred in the Istanbul Strait and proposes both quantitative and qualitative assessments of marine accidents. Marine accidents occurring in the Istanbul Strait are analysed by using a method based on neuro-fuzzy and genetically optimised fuzzy classifiers. It can be concluded that accident severity increases when poor weather conditions prevail in the Strait regardless of ship size. Therefore, solutions to reduce unwanted events should be prioritised by accounting for weather conditions and the capacity of the vessels. This analysis indicates that the safety level would be significantly improved if all the vessels follow the passage guidelines.