A Robust Design Method Reflecting Design Intent

This study proposes a set-based design method for deriving multi-objective solutions while addressing various requirements in the early design stage. By representing uncertain design variables and performances as range sets, diverse range solutions that satisfy all requirements are derived. Furthermore, by including noise factors in the design variables as a range, a robust design that achieves stable performance under various conditions is realized. The derived range solutions are divided equally based on the design variables, and evaluated based on the evaluation index set by the designer to extract a highly evaluated range solution that reflects the designer's intention. By analyzing these range solutions in detail, it is possible to understand how the designer's preferences are reflected in the design variables and performances. To validate the proposed method, it was applied to the structural design of a reusable rocket. Using a compactness-focused evaluation metric, the method successfully identified multiple high-performing design solutions that reflected the designer’s intention, demonstrating its effectiveness in supporting decision-making during the early stages of design.