A Multi-Step Analysis Procedure for Robust Design

稳健设计的多步分析程序

• 批准号: 9312698
• 负责人: Kwokleung Tsui
• 金额: $ 8.24万
• 依托单位: Georgia Tech Research Corporation
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1993
• 资助国家: 美国
• 起止时间: 1993-09-01 至 1998-02-28
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9312698&HistoricalAwards=false
• 关键词: Multi; Step; Analysis; Procedure; Robust

9312698 Tsui Robust Design is an important method for improving product or manufacturing process design by making the output response insensitive (robust) to hard-to-control variations (noise). In analyzing robust design experiments, Taguchi (1986) proposed a two- step procedure to identify the "optimal" factor settings that minimize the average quadratic loss. His two-step procedure becomes very effective in practice since it reduces the dimension of the design optimization problem and it does not require re- optimization for future changes of the target value. This research will develop improved statistical methods for analyzing robust design experiments. The researchers develop a multi-step analysis procedure for a variety of robust design problems. Their new procedure substantially reduces the dimension of the design optimization problem and does not require reoptimizaton for future changes of design requirements. This produces savings in the development time of product and manufacturing process design and allows for flexible manufacturing. The procedure also helps engineers better understand the physical mechanism of the product and process and allows the designer to make tradeoffs among the off-target bias and the individual process variances contributed by different noise factors. The research also identifies the appropriate robust design objectives and establishes their relationships with the quality loss functions for various robust design problems. They will strategies for implementing the multi-step procedure for smaller- the-better and larger-the-better problems as well as for the problems with several responses.

小行星9312698 鲁棒设计是一种通过使输出响应对难以控制的变化（噪声）不敏感（鲁棒）来改进产品或制造过程设计的重要方法。 在分析稳健设计实验时，田口（1986）提出了一个两步程序来确定使平均二次损失最小化的“最优”因子设置。 他的两步过程在实践中变得非常有效，因为它降低了设计优化问题的维度，并且不需要针对目标值的未来变化进行重新优化。 本研究将发展稳健设计实验之统计分析方法。 研究人员开发了一个多步骤的分析程序，用于各种鲁棒设计问题。 他们的新方法大大降低了设计优化问题的维数，并且不需要为将来设计要求的变化进行重新优化。 这节省了产品和制造工艺设计的开发时间，并允许灵活制造。 该过程还有助于工程师更好地理解产品和工艺的物理机制，并允许设计人员在偏离目标偏差和不同噪声因素引起的单个工艺差异之间进行权衡。 研究还确定了适当的稳健设计目标，并建立了它们与质量的关系 各种鲁棒设计问题的损失函数。 他们将为小而好和大而好的问题以及具有多个响应的问题提供实施多步骤过程的策略。