OCEAN ENGINEERING, cilt.343, 2026 (SCI-Expanded, Scopus)
Cargo inspection in chemical tanker ships is a critical process that impacts operational efficiency, safety, environmental protection, and financial performance. This paper presents a systematic approach integrating Process Hazard and Operability (Process HAZOP) analysis with an improved Z-numbers Bayesian Belief Network (BBN) to assess and quantify risks associated with cargo nomination, acceptance, and potential rejection. The proposed methodology evaluates risk factors across all stages, including load port (prior to loading and during/after loading), and discharge port (pre-discharge and during discharge). The integration of improved Z-numbers enhances uncertainty management by incorporating expert judgment, while BBN facilitates causal relationship analysis and dynamic risk updates. A case study demonstrates the application of this approach, identifying high-risk scenarios and offering insights to optimize cargo handling and mitigate rejection risks. The results of the study reveal that the rejection risk during the cargo loading process is 1.53E-01. On the other hand, the most critical deviations are found in D13 (Manifold sample/first foot sample/after-loading sample off-spec), D14 (Cargo contamination-load port), and D1 (cargo tank is not clean). The improved Z-numbers BBN model, providing reliable, data-driven guidance for crews, inspectors, and Health, Safety, Environment, and Quality (HSEQ) managers to enhance safety and operational performance.