抄録
Robust design enables the products to ensure robust performance in diverse conditions due to increased individuation of user needs and the globalization of markets. In the previous study, a robust design method which evaluates robustness of adjustable mechanism was proposed by using a control factor with adjustable range. However, the method is not applicable to design problems in which more than one control factor with adjustable range are included. In the present study, the previous method was improved to be applicable to more than one control factor with adjustable range. First, a need of improve computational efficiency is identified as a problem to be solved. Second, to improve computational efficiency, a method to decrease computational effort in both searching solution and evaluating robustness are proposed. Specifically, in the proposed method, Genetic Algorithm is applied to search solution, and several adjustable control factors' values used for evaluation of robustness are extracted from their adjustable range. Finally, the proposed method was applied to two numerical examples and effectiveness was confirmed.
