Marine accident learning with fuzzy cognitive maps (MALFCMs)

Navas de Maya, Beatriz and Kurt, Rafet Emek (2020) Marine accident learning with fuzzy cognitive maps (MALFCMs). MethodsX, 7. 100940. ISSN 2215-0161 (https://doi.org/10.1016/j.mex.2020.100940)

[thumbnail of de-Maya-Kurt-MethodsX-2020-Marine-Accident-Learning-with-Fuzzy-Cognitive-Maps]
Preview
Text. Filename: de_Maya_Kurt_MethodsX_2020_Marine_Accident_Learning_with_Fuzzy_Cognitive_Maps.pdf
Final Published Version
License: Creative Commons Attribution 4.0 logo

Download (798kB)| Preview

Abstract

Statistical analysis of past accidents in maritime may demonstrate the trends for certain contributing factors in accidents, however, there is a lack of a suitable technique to model the complex interrelations between these factors. Due to aforementioned complex interrelations and insufficient information stored in accident databases, it was not possible to understand the importance of each factor in accidents, which prevented researchers from considering these factors in risk assessments. Therefore, there is a need for a capable technique to estimate the importance of each factor. The results of such a technique can be used to inform risk assessments and predict the effectiveness of risk control options. Thus, this study introduces a new technique for Marine Accident Learning with Fuzzy Cognitive Maps (MALFCMs). The novelty of MALFCM is the application of fuzzy cognitive maps (FCMs) to model the relationships of maritime accident contributors by directly learning from an accident database as well as having the ability to combine expert opinion. As each fuzzy cognitive map is derived from real occurrences supported by expert opinion, the results can be considered more objective. Thus, MALFCM may overcome the main disadvantage of fuzzy cognitive maps by eliminating or controlling the subjectivity in results. • A novel MALFCM method to weight human-contributing factors into maritime accidents has been developed. • With MALFCM method the main disadvantage of traditional FCMs is overcome. • The MALFCM method can produce logical results even by solely using information from historical data in the absence of expert judgement.