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 pagelinkcall_made for the publication and this link for the associated report on our internal archive.