Multi-objective Robust Optimal Design Based on Fuzzy Mathematics

In robust optimal design, the design objective often involves multiple aspects such as “bringing the mean of performance on target” and “minimizing the variations”. Current ways of handling multiple aspects using either the Taguchi’s signal-to-noise ratio or the weighted-sum method are not adequate. And in process and production design, there are a lot of fuzzy factors and random factors. In this paper, we solve multi-objective robust optimal design problems from a practical perspective by fuzzy mathematics. Firstly, we review the discrete tolerance design model which is often used in computer aided tolerance design. Secondly we gain the satisfactorily global optimal solution of the multi-objective problem by use of principle of fuzzy closeness optimization. At last results from applying the above techniques to an optimal example show that the proposed approach is effective.