Small Grant for Exploratory Research: Knowledge Based Parameter Design System for Off-Line Quality Control

9416666奥戈巴这项研究的重点是结合田口方法和统计学的某些方面的质量控制建模工具的开发。该工具的目标用户是设计工程师。该技术是基于参数设计的。参数设计涉及并需要深厚的工艺知识，以及进行详细的统计试验设计和分析以确定关键工艺参数的能力。因此，这项研究还包括开发一个串联结构的原型知识库专家系统，该系统由知识库和算法库组成，以优化最终的设计。有了知识库，用户就可以进行参数设计，而不需要进行详细的统计试验设计和随后的分析。开发的用于参数设计的知识库系统可以弥合田口方法与文献中提出的替代方法之间的一些方法差异。成功实施参数设计可以减少质量损失。它可以提高没有统计背景和资源的设计工程师承担全面统计试验设计的严谨工作的生产力。这一优势在并行工程环境中尤为重要，因为在并行工程环境中，非常需要快速响应设计。

9416666 Okogbaa This research is focused on the development of modeling tools for quality control that combines some aspects of Taguchi approach and statistics. The intended users of the tool are design engineers. The technique is based on parameter design. Parameter design involves and requires deep process knowledge and the ability to carry out detailed statistical experimental design and analysis to identify the critical process parameters. Thus, the research also involves the development of a prototype knowledge base expert system with tandem architecture that consists of a knowledge base and an algorithmic base to optimize the resulting design. With the knowledge-base in place, a user can implement parameter design without engaging in detailed statistical experimental design and the ensuing analyses. The developed knowledge-base system for parameter design can bridge the gap between some of the methodological differences between Taguchi approach and some to the alternative approaches proposed in the literature. The successful implementation of parameter design can lead to a reduction in quality losses. It can enhance the productivity of design engineers who do not have the statistical background and resources to undertake the rigors of comprehensive statistical experimental design. This benefit is particularly crucial in a concurrent engineering environment where fast response time to design is highly desirable.