Marine accident learning with fuzzy cognitive maps (MALFCMs) : a case study on fishing vessels

Tools

Navas de Maya, Beatriz and Kurt, Rafet Emek and Turan, Osman (2019) Marine accident learning with fuzzy cognitive maps (MALFCMs) : a case study on fishing vessels. In: European Safety and Reliability Conference, 2019-09-22 - 2019-09-26, Hannover.

[thumbnail of Navas-de-Maya-etal-ESRC-2019-Marine-accident-learning-with-fuzzy-cognitive-maps]
Preview
 Text. Filename: Navas_de_Maya_etal_ESRC_2019_Marine_accident_learning_with_fuzzy_cognitive_maps.pdf
Accepted Author Manuscript
Download (957kB)| Preview

Abstract

Despite advanced safety systems installed on ships, marine accidents still occurs at a more-or-less constant rate. This situation can be attributed to the fact that accidents occurred in a complex way and the role of humans into past accidents is not properly understood in this process. Furthermore, a number of factors are combined to result in a failure/accident but interrelations of these factors are not well understood. Therefore, shipping industry can benefit from a practical method, which is capable of considering the interrelations and identifying the importance weightings for each factor involved in an accident. Thus, in this paper, a new technique for Marine Accident Learning with Fuzzy Cognitive Maps (MALFCMs) is developed and demonstrated. The method utilises Fuzzy Cognitive Maps (FCMs) to model the relationships by also integrating information from an accident database. By applying accident data instead of expert judgement, MALFCMs may overcome the main disadvantage of FCMs by controlling the subjectivity in results attributed to expert opinion. Within this study, MALFCMs is applied to fishing vessels accident data, in order to compare the results with the findings of an existing report provided by the European Maritime Safety Agency (EMSA). In order to make this comparison, Collision and Fire/explosion accidents were selected and comparatively analysed in this paper. Our study shows that MALFCM can produce results, which are in line with the findings from aforementioned EMSA report.

ORCID iDs

Navas de Maya, Beatriz ORCID logoORCID: https://orcid.org/0000-0002-3595-9401, Kurt, Rafet Emek ORCID logoORCID: https://orcid.org/0000-0002-5923-0703 and Turan, Osman;
CORE (COnnecting REpositories)