The Multi-response Integrated Design Of Both Parameter And Tolerance Based On Uncertainty

With the development of science and technology,product quality,cost,and time to market have become the focus of domestic and foreign market competition.Traditional serial design can no longer adapt to the rapidly changing market demand.From the perspective of integration,how to use quality engineering technique to design and produce high-quality and low-cost products in a short period has become the key to enterprises winning the competition.In order to achieve the quality design of the product,the relationship between product/process input and output variables must be explored.However,in practice,the complexity of the process,test error,and other factors will affect the accuracy of the experiment data to a certain extent.These would result in uncertainty in the modeling process.In addition,actual products often need to consider multiple quality characteristics(multiple responses)simultaneously.Therefore,how to reduce the influence of uncertain factors on the economic design of multiple responses products through modeling and optimization techniques has attracted the attention of researchers.In this thesis,the economic design problem of parameters and tolerances is taken as the research object from the perspective of integrated design.We take quality and cost as the goal based on response surface methodology(RSM).And then we do in-depth research on model parameter uncertainty,input parameter uncertainty,and the robust modeling of multi-response concurrent parameter and tolerance design.This study is important for organizations to improve product quality,reduce costs,and enhance market competitiveness.The research contents of this thesis include:Firstly,taking parameter and tolerance integrated design for multi-response based on the location effect as the research object,a new expected quality loss function considering the correlation of responses,the tolerances of the design variables,and the uncertainty of the model parameters is proposed.Furthermore,the optimization objective function including tolerance cost and quality loss function is established.The results of practical example and simulation test analysis show that the proposed method not only improves the robustness and the performance of the model prediction,but also can significantly reduce the total cost while improving product quality.Secondly,for multi-response parameter and tolerance integrated design based on location and dispersion effects problems,this thesis introduces the model parameter uncertainty and the dispersion effect of the quadratic loss function into the modeling of multi-response parameter and tolerance design.Through this,the constructed model can make a reasonable trade-off between location effect,dispersion effect,and tolerance costs,to achieve a lower total cost.The effectiveness and robustness of the proposed method are illustrated with a practical example and a random simulation example.Thirdly,for parameter and tolerance integrated design of multi-response considering predicted response variability problems,this thesis uses interval analysis theory to realize product quality design under uncertain environment.And then an integrated optimization model is constructed based on the idea of robust design,from the perspective of location and dispersion effects.The model includes quality loss,tolerance cost,qualified product,and scrap costs for optimizing multi-response issues.This method also incorporates the loss caused by the variation of the predicted response value into the optimization model,which improves the robustness of product quality at a lower economic cost.Finally,taking parameter and tolerance integrated design for multi-response based on noise variables uncertainty as the research object,the parameter uncertainty in noise variable and model based on the interval estimation theory are analyzed.Then from the perspective of robustness and economy,optimization object function is constructed for solving the multi-response optimization problem in quality design,which includes the location and dispersion effects of quality loss and tolerance cost.This method simultaneously incorporates the two important uncertainty into the objective function.This method not only has good robustness to the disturbance of uncertainty,but also can make a reasonable trade-off between quality loss and tolerance cost to achieve a lower total cost.