Seaport Hazardous Cargo Loading and Unloading Risk Assessment Using Interval Type-2 Fuzzy Sets and Bayesian Networks

Abstract

Manifold calamitous consequences of seaport hazardous cargo accidents have raised grave trepidation over the safe management of hazardous cargo. Fewer investigations have aimed at the multifaceted risk associated with intricate seaport hazardous cargo loading and unloading operations. To that end, this study analyzes the multifaceted risks associated with seaport hazardous cargo loading and unloading operations. Bayesian network (BN) is employed and amalgamated with interval type-2 fuzzy sets to account for intricacies and uncertainties concomitant to the causation factors. Power average operator on aggregation of domain expert opinions, and experts scores in quantifying conditional probability tables are employed to further augment the viability and diagnosis accuracy of the analysis. The results reveal policy and regulations failures, errors and violations, and communications issues as the prominent accident causation factors. To identify the critical accident causation factors, a sensitivity analysis was performed. Based on these insights, the study proposes practical applications in terms of development and abidance of rules, resource upkeeping, and critical operations to mitigate the risk of seaport hazardous cargo loading and unloading operations.