Improved robust design optimisation of structures Available to Purchase

Robust design optimisation is conventionally achieved by minimising the expected value and the variance of the performance function by attaching equal importance to each individual gradient of the performance function. But, it is well established in the literature that not all the gradients should be of equal importance when accounting for uncertainty. The present work proposes an improved robust design optimisation algorithm. The basic idea is to improve performance robustness by defining a new index based on factors that are proportional to the importance of the gradients of the performance function. The same concept is used to achieve the robustness of the constraints. Two examples are considered: minimisation of the cost of a steel column; and weight minimisation of a simply supported beam. The results confirm that the proposed formulation is an improvement on conventional robust design optimisation.