Abstract:
This work presents a comprehensive collection of applications demonstrating how reliability-based design and probabilistic analysis, heavily reliant on computational tools like the Finite Element method, are implemented across the aerospace, automotive, and ship industries to address uncertainty and enhance safety. In aerospace applications, specialized probabilistic software such as NESSUS and DARWIN are integrated with Finite Element models created in codes like ANSYS and ABAQUS to perform risk assessments on critical components like gas turbine engine rotors, analyze corrosion-fatigue in aircraft joints, model space shuttle debris impact, and conduct probabilistic progressive buckling analysis of space trusses. The automotive sector utilizes large-scale, nonlinear Finite Element simulations in programs like LS-DYNA to conduct stochastic crashworthiness analyses, where uncertainties in material properties and manufacturing tolerances are propagated through the models to predict the probability of meeting safety standards. Furthermore, these computationally intensive FEA simulations are used to build fast-running metamodels, or response surfaces, which enable efficient reliability-based design optimization of complex vehicle systems. For the ship industry, the methodology involves assessing the ultimate limit state of hull girders by using Finite Element Analysis to determine structural capacity, which is then used in reliability calculations that account for time-dependent degradation from corrosion and fatigue over the vessel's service life. Across all industries, the integration of FEA with probabilistic methods provides a rigorous framework for moving beyond deterministic safety factors to a quantitative, risk-informed design process that systematically manages the effects of variability on structural performance.
