Answers are grounded in this paper and cite their sources.
Buckling is a critical failure mode of steel columns, strongly affected by uncertainties in material properties, geometry, and loading. This study develops a FORM-based analytical framework for global reliability sensitivity analysis considering multiple failure modes, namely Euler elastic and elasto-plastic buckling in accordance with Eurocode 3. Semi-analytical expressions of first-order and total-effect Sobol' indices are derived directly from FORM outputs, eliminating additional model evaluations. The proposed approach requires on the order of 10²–10³ model evaluations, compared to ns(n+2)≈9×107for equivalent pick-freeze Monte Carlo sensitivity estimators, a speed gain greater than 105. Multivariate normal integrals are efficiently computed using a minimax tilting technique. Numerical results show excellent agreement with Monte Carlo estimates, with differences typically below 5% for sensitivity indices and about 8–15% for failure probabilities. The analysis identifies yield strength (≈ 0.40) and axial load (≈ 0.30) as the most influential parameters. Geometric imperfections exhibit moderate first-order effects (≈ 0.10) but significant total-effect contributions (up to ≈ 0.40), indicating strong interaction effects. Their influence increases markedly with slenderness, especially near the transition zone (λ̄ ≈ 1.0). The proposed framework enables efficient and physically interpretable sensitivity analysis of structural stability problems involving multiple interacting failure modes.
No takes yet. Share an insight, caveat, or question.
Mouloud Hamhami (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: