Six Sigma and Design of Experiments (DoE)
The concept of the design of experiments (DoE), alternatively known as “robust design” or “variability reduction,” has been used to reduce some of the sources of manufacturing variation or manipulate a design toward its intended performance. These attributes make DoE one of the most effective tools for reaching six sigma in design and manufacturing.
DoE influences both ends of the six sigma ratio: manufacturing operations produce parts with defects, either because of tight design specifications or manufacturing process variability. Using DoE, the need to have narrow specification limits can be eliminated, and the product can operate satisfactorily within wide production process variability. Most of the applications of DoE have been made in the production or process development phases of new products, because the use of DoE is most beneficial in multidisciplinary applications, where traditional engineering analysis, simulation, and verification are difficult to achieve.
In design applications and new product development, DoE is very effective in systems design when there is considerable interaction among the system components in achieving system performance. Six sigma tools can be effectively used in the selection of the quality characteristic in DoE experiments because they can point to where the most benefit can be extracted. This chapter will address the issues of using DoE methods in design and manufacturing of new and current products striving to achieve six sigma. The topics to be discussed in this chapter are:
1. DoE definitions and expectations. In Section 7.1, the definition DoE is given, as well as the expectations of proactive improvement of the product and process design. The reasons for using DoEs are discussed, including the effects of noise and other external and internal conditions that contribute to the variability of products and processes.
2. Design of experiments (DoE) techniques. These techniques are introduced in Section 7.2 with an algorithm for conducting a D〇E project, and selection of the quality characteristic.
3. The DoE analysis tool set. These tools are presented in Section 7.3 with case studies for each. They include graphical and statistical analysis of the average and the variance of the quality characteristic.
4. Using DoE methods in six sigma design and manufacturing, DoE design techniques have been used mostly to reduce manufacturing variability. Section 7.4 addresses the use of DoE methods for design engineering applications as well as optimizing manufacturing