Marine accidents are complex processes in which many factors are involved and contribute to accident development. For this reason, effectively analyse what combination of factors lead an accident event is a complex problem, especially when human factors are involved. State-of-the-art methods such as Human Factor Analysis and Classification System, Human reliability Assessments, and simple Statistical Analysis are not effective in many situations since they require the intervention of human experts with their limitations, biases, and high costs. The authors propose to use a data-driven approach able to utilise the information present in historical databases of marine accident for the purposes of establishing the most influential human factors. For this purpose a two-stage approach is presented: first, a data-driven predictive model is built able to predict the type of accident based on the contributing factors, and then the different contributing factors are ranked based on their ability to influence the prediction. Results on a real historical database of accidents provided by the Marine Accident Investigation Branch will support the proposed novel approach.
Addressing safety is considered a priority starting from the design stage of any vessel until end-of-life. However, despite all safety measures developed, accidents are still occurring. This is a consequence of the complex nature of shipping accidents where too many factors are involved including human factors. Therefore, there is a need for a practical method, which can identify the importance weightings for each contributing factor involved in accidents. As a result, by identifying the importance weightings for each factor, risk assessments can be informed, and risk control options can be developed and implemented more effectively. To this end, Marine Accident Learning with Fuzzy Cognitive Maps (MALFCM) approach incorporated with Bayesian networks (BNs) is suggested and applied in this study. The MALFCM approach is based on the concept and principles of fuzzy cognitive maps (FCMs) to represent the interrelations amongst accident contributor factors. Thus, MALFCM allows identifying the importance weightings for each factor involved in an accident, which can serve as prior failure probabilities within BNs. Hence, in this study, a specific accident will be investigated with the proposed MALFCM approach. Keywords Maritime accidents. Maritime safety. Maritime accident learning with fuzzy cognitive maps (MALFCMs). Human factors. Bayesian networks (BNs)
Maritime transport has strived to reduce accidents and their consequences since its origins, by addressing safety as the priority from the design stage to decommissioning of any vessel. Complex nature of accidents, where numerous factors combine in a complicated structure, in turn, makes accidental learning ineffective. Statistical analysis of past experiences in maritime is good for demonstrating the trends for certain contributing factors in accidents. However, there is a lack of a detailed technique, which is capable of modelling the complex interrelations between these factors. Due to aforementioned complex interrelations between these contributing factors and insufficient information stored in accident databases about these contributors, it was not possible to understand the importance of each factor in maritime accidents, which prevented researchers from considering these factors in risk assessments. Therefore, there is a need for a practical technique, which is capable of estimating the importance of each contributing factor. The results of such a technique can be used to inform risk assessments and predict the effectiveness of risk control options. Thus, in this research study, a new technique for Marine Accident Learning with Fuzzy Cognitive Maps (MALFCMs) has been introduced and explained. The novelty of MALFCM is the application of fuzzy cognitive maps (FCMs) to model the relationships of accident contributors by utilizing information directly from an accident database with the ability to combine expert opinion. Hence, as each fuzzy cognitive map will be derived from real occurrences, the results can be considered entirely objective, and MALFCM may overcome the main disadvantage of fuzzy cognitive maps by eliminating or controlling the subjectivity in results. In this paper, FCMs were developed for various accident scenarios and contributing human factors were assessed. For instance, in collision accidents in bulk carriers, situational awareness or inadequate communication were identified as the most critical factors, with a normalised importance weighting of 4.88% and
The application of data mining techniques is an extended practice in numerous domains; however, within the context of maritime inspections, the aforementioned methods are rarely applied. Thus, the application of datamining techniques for the prediction and ranking of non-conformities identified during vessel inspections could be of significant managerial contribution to the safety of shipping companies, as non-conformities could potentially lead to real accidents if not addressed adequately. Hence, specific data mining methods are investigated and applied in this paper to predict and rank non-conformities on oil tankers using a database recorded by tanker shipping companies in Turkey from 2006 to 2019. The results of this study reveal that specific non-conformities, for instance, inadequate ice operations or inadequate general appearance and condition of hull, superstructure and external weather decks, are not company-based problems, rather they are industry wide issues for all tanker shipping companies.
Human factors (HF) in aviation and maritime safety occurrences are not always systematically analysed and reported in a way that makes the extraction of trends and comparisons possible in support of effective safety management and feedback for design. As a way forward, a taxonomy and data repository were designed for the systematic collection and assessment of human factors in aviation and maritime incidents and accidents, called SHIELD (Safety Human Incident and Error Learning Database). The HF taxonomy uses four layers: The top layer addresses the sharp end where acts of human operators contribute to a safety occurrence; the next layer concerns preconditions that affect human performance; the third layer describes decisions or policies of operations leaders that affect the practices or conditions of operations; and the bottom layer concerns influences from decisions, policies or methods adopted at an organisational level. The paper presents the full details, guidance and examples for the effective use of the HF taxonomy. The taxonomy has been effectively used by maritime and aviation stakeholders, as follows from questionnaire evaluation scores and feedback. It was found to offer an intuitive and well-documented framework to classify HF in safety occurrences.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.