An Operational Risk Analysis Model for Container Shipping Systems considering Uncertainty Quantification. (May 2021)

• Record Type: Journal Article
• Authors: Nguyen, Son; Chen, Peggy Shu-Ling; Du, Yuquan; Thai, Vinh V.
• Is Part Of: Reliability engineering & system safety. Volume 209(2021)
• Publication Date: 2021-05
• Subjects: Risk assessment -- Container shipping -- Bayesian Network -- Evidential Reasoning -- Operational risk; BN Bayesian network -- ER Evidential reasoning -- HE Hazardous events -- RMI Risk Magnitude Index -- UI Uncertainty Index -- DoB Degree of belief -- UDoB Unassigned DoB -- CPT Conditional probability table -- ICT Information and communications technology -- QRA Quantitative risk analysis -- RMP Risk mitigation/prevention -- TEU Twenty-foot equivalent unit -- UQM Uncertainty quantification module -- CSOR Container shipping operational risk -- FMEA Failure mode and effects analysis -- TOPSIS Technique for order of preference by similarity to ideal solution; Reliability (Engineering) -- Periodicals; System safety -- Periodicals; Industrial safety -- Periodicals; Fiabilité -- Périodiques; Sécurité des systèmes -- Périodiques; Sécurité du travail -- Périodiques; 620.00452
• DOI: 10.1016/j.ress.2020.107362
• Languages: English
• ISSNs: 0951-8320

Abstract

Highlights: A quantitative risk analysis model for container shipping operations is proposed and validated. An uncertainty quantification module (UQM) is integrated into the risk analysis model. Evidential Reasoning is employed to aggregate and separate distributed subjective probabilities. The study confirms the criticality of outcome and evidential uncertainties awareness in risk assessment. The UQM reveals experts' limited ability in assessing the knowledge base of risk assessments. Abstract: Different uncertain factors obstruct the analysis of operational risks in container shipping, especially those rooted in the subjectivity of multiple risk assessments and their aggregation. This paper proposes a risk analysis model featuring a quantification of the uncertainty. Bayesian probability theory is employed to quantify the risk magnitude, while a dedicated module to handle uncertainty is enabled by Evidential Reasoning and a set of three uncertainty indicators, including expert ignorance, disagreement among experts, and polarization of their assessments . The situation of risk is diagnosed by risk ranking and visualized by risk mapping, using both Risk Magnitude Index and Uncertainty Index. The functionality of the proposed model in identifying critical and uncertain risks was demonstrated in an organizational-scale case study, followed by an examination of validity criteria and a sensitivity test. The case study reveals the physical flow as the dominant origin of high-ranking risks with potential significant consequences such as piracy, dangerous cargoes, and maritime accidents; while information and financial operational risks are more uncertain, especially cargo misdeclaration and unexpected rises of fuel costs.