Human error probability in ship ballast water management systems: A HEART-based human reliability analysis across the BWMS lifecycle

Ballast Water Management Systems (BWMS) rely extensively on human-in-the-loop activities across planning, operation, monitoring, and compliance. The Human Error Assessment and Reduction Technique (HEART) was applied to quantify task-level Human Error Probabilities (HEPs) across the full BWMS operational lifecycle. A hierarchical task analysis identified fine-grained tasks, which were mapped to appropriate Generic Task Types (GTTs) and combined with context-specific Error-Producing Conditions (EPCs). Subject-matter experts provided Assessed Proportions of Affect (APoA; 0–1), enabling calculation of nominal HEPs. Residual HEPs were obtained by modelling Error-Reduction Strategies (ERSs) as proportional reductions in APoA based on task-specific mitigation mechanisms. Results show that risk concentrates in a compact set of high-leverage tasks—valve line-ups, alarm interpretation and response, bypass decision-making, neutralisation dosing, and D-2 indicative sampling—where co-occurring EPCs such as display or procedural inconsistencies, insufficient verification, interruptions or workload peaks, and ambiguous feedback combine multiplicatively to elevate error likelihood. Design-based controls (e.g., interlocks, LOTO measures, improved HMI feedback and mimic alignment) together with procedural and organisational controls (standardised techniques, job aids, two-person verification, and short no-interrupt windows) substantially reduce residual HEP. The approach provides a transparent and reproducible basis for prioritising interventions and strengthening human-reliability contributions to BWMS safety management and compliance assurance.