Multi-objective robust optimization using adaptive surrogate models for problems with mixed continuous-categorical variables
New paper published in Structural Multidisciplinary Optimization M. Moustapha and B. Sudret publish a paper on multi-objective robust optimization together with A. Galimshina and G. Habert (D-BAUG/IBI).
The paper entitled Multi-objective robust optimization using adaptive surrogate models for problems with mixed continuous-categorical variables proposes a novel approach for the solution of multi-objective and robust design optimization problems. The latter is formulated using quantiles as measure of robustness and solved by the general purpose non-dominated sorting genetic algorithm II (NSGA-II) upon the introduction of the concept of common random numbers. To reduce the computational cost, a Kriging-based enrichment scheme which is coupled to NSGA-II is proposed. Finally, the framework is adapted to handle problems with mixed continuous-categorical parameters. Two numerical examples and an engineering application on building renovation validate the proposed methodology.
For more information, please follow this external page link for the publication and this link for the associated report on our internal archive.