An operational risk awareness tool for small fishing vessels operating in harsh environment

Probabilistic safety assessment using the Bayesian network (BN) has emerged as a popular method for developing risk analysis tools. The method allows for risk-influencing factors from several areas to be captured probabilistically, enabling an easy-to-use safety assessment tool to be developed. Meanwhile, because the resulting tool is a BN model, the use of subject-matter experts in eliciting probabilities for the conditional probability tables (CPT) makes the method subjective. The subjectivity makes the tool's output result less reliable since different experts rarely produce the same probabilities for CPTs. Therefore, the present study proposed a probability-scoring scale that uses pre-determined scores to assign probabilities to CPTs. Using the scale ensures that different experts working on a common CPT produce identical probabilities. That way, the variability amongst experts’ results is minimised while the reliability of a BN's output result increases. The scale was applied to a BN-based risk-awareness (RAw) tool developed for monitoring safety aboard small fishing vessels (SFV). Advanced safety monitoring equipment is lacking aboard many SFVs, especially those in developing countries. Hence, the RAw tool developed in the present study demonstrates how the probabilistic safety assessment method could be leveraged to equip SFVs with safety monitoring tools. The study will benefit SFV owners and operators, the commercial fishing industry, and maritime administrations in charge of ensuring safety aboard SFVs.