Answers are grounded in this paper and cite their sources.
This study presents a framework for the reliability and sensitivity analysis of hull girders, accounting for both local structural element failures and global collapse. Unlike conventional two-stage reliability approaches in ship structures that rely on fitted probability distributions of structural response, the proposed framework employs a surrogate model to directly approximate the implicit limit state function in terms of the basic random variables, thereby eliminating uncertainties associated with distribution selection and parameter estimation. The framework first integrates nonlinear finite element analysis with the progressive collapse approach recommended by the International Association of Classification Societies to identify critical structural components and to evaluate implicit limit state functions at both local and global levels. Subsequently, a PC-Kriging surrogate model is employed within an active learning strategy to efficiently estimate the failure probabilities of individual structural elements and the overall hull girder. Furthermore, since the limit state function is constructed directly based on the local component's material properties, geometric characteristics, and external loading conditions, the proposed framework enables to examine the sensitivity of the global hull girder failure to these parameters. The applicability and efficiency of the proposed approach is investigated through the reliability and sensitivity analysis of a bulk carrier.
No takes yet. Share an insight, caveat, or question.
Kakaie et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: