Robust Design Model Based On Uniform Design And Economic Design Model

Originated by Dr.Taguchi, parameter design is always an economical and effective approach to reduce variation in products and improve product quality during the period of designing. Through finding the appropriate parameters combination, the parameter design making quality attribute that is not sensitive to the importation, thus reduce the amount of products variation. Uniform design, which is created by Dr. K.T Fang, emphasize the model developing data from experimental data by less run. In the process of uniform design, the focus of the experimentation is choosing factor level suitability so that we can get a precision model.In this dissertation, a thorough analysis of all models of robust design and uniform design in existence has been done on the basis of the systemic review of the development history and current research status of robust design and uniform design. Afterwards, a groupe new strategy of uniform cross table design process of three stages based on uniform design and Taguchi methods is advanced, this process makes full of the intense complementary action of characteristics between uniform design and Taguchi methods. After that, we established the uniform robust model and uniform economic model, and then made confirmation using one example. Particularly, the problem of fitting response model with those control factors have variation has been discussed in this thesis, and the model was consistent with actual situations ideally.The characteristics and innovation of this paper is:①The uniform robust design has been proposed based on uniform principle and a new strategy of robust optimization design process of three stages based on uniform design and Taguchi methods is advanced.②The uniform robust and economic optimization model has been established based on uniform design.③For the continuous situation, we studied the uniform robust and economic optimization model which output variation was caused by input variation and noise factors input.Finally, this paper has great application value for two aspects. On the one hand, we can avoid inefficiency by runs increase with using uniform robust design. On the other hand, we can find the accuracy optimal solution efficiently.