论文部分内容阅读
针对传统的综合满意度模型在研究多响应时较少考虑噪声因素的问题,本文通过引入噪声因素对传统的满意度模型实施改进,进行多响应稳健优化分析。首先,构建包含可控因素和噪声因素的多质量特性的响应曲面模型;其次,借助传统综合满意度模型的构建方法,将多质量特性的响应曲面模型整合为改进的满意度模型;最后,将信噪比作为衡量改进的满意度模型稳健优化的指标,得到稳健优化的参数组合。实证研究表明:构建同时包含可控因素和噪声因素的改进的满意度模型是可行的,在此模型的基础上利用信噪比能够有效地找到多响应稳健优化的参数组合。
Aiming at the problem of less comprehensive consideration of noise factors in the traditional comprehensive satisfaction model, this paper improves the traditional satisfaction model by introducing noise factors, and makes a multi-response robust optimization analysis. Firstly, a multi-quality response surface model with controllable factors and noise factors is constructed. Secondly, the multi-quality response surface model is integrated into an improved satisfaction model by using the traditional comprehensive satisfaction model construction method. Finally, Signal-to-noise ratio (SNR), as a measure of the robust optimization of the improved satisfaction model, yields a robustly optimized combination of parameters. The empirical research shows that it is feasible to build an improved satisfaction model that contains both controllable and noise factors. Based on this model, the signal-to-noise ratio can effectively find the optimal combination of multiple response robust optimization parameters.