Marine Accident Learning with Fuzzy Cognitive Maps (MALFCMs) : a case study on bulk carrier's accident contributors

Navas de Maya, Beatriz and Kurt, Rafet Emek (2020) Marine Accident Learning with Fuzzy Cognitive Maps (MALFCMs) : a case study on bulk carrier's accident contributors. Ocean Engineering, 208. 107197. ISSN 0029-8018 (https://doi.org/10.1016/j.oceaneng.2020.107197)

[thumbnail of de-Maya-Kurt-OE-2020-Marine-Accident-Learning-with-Fuzzy-Cognitive-Maps]
Preview
 Text. Filename: de_Maya_Kurt_OE_2020_Marine_Accident_Learning_with_Fuzzy_Cognitive_Maps.pdf
Accepted Author Manuscript
License: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 logo
Download (1MB)| Preview

Abstract

Statistical analysis of past maritime accidents may demonstrate the trends for certain contributing factors. However, there is a lack of a technique, which is capable of handling complex nature of maritime accidents by modelling interrelations between contributing factors. Due to the aforementioned complex interrelations and insufficient detail stored in accident databases about these contributors, it was not possible to quantify the importance of each factor in maritime accidents. This situation prevented researchers from considering these factors in risk assessments. Thus, in this research study, a technique for Marine Accident Learning with Fuzzy Cognitive Maps (MALFCMs) has been demonstrated. MALFCM employs fuzzy cognitive maps (FCMs) to model the relationships of accident contributors by using information directly from an accident database with the ability to combine expert opinion. Hence, the results can be considered more realistic and objective, which overcomes the main disadvantage of FCMs by eliminating or controlling the subjectivity in results. In this paper, FCMs were developed for bulk carriers with the aim of assessing the importance of contributing factors. For instance, in collision accidents in bulk carriers, situational awareness and inadequate communication were identified as the most critical factors, with a normalised importance weighting of 4.88% and 4.87% respectively.

CORE (COnnecting REpositories)