Abstract

This abstract is an edited version of the author's original.

Robust design methods are the set of concepts and practices which directly affect the ability of the product or process to withstand the influence of variation and continue to provide high quality performance. Practitioners of Design of Experiments (DOE) methodologies use DOE to improve the overall robustness of product and process designs; however, many engineers do not understand robust design methods and the supporting principles and practices for their application. This paper presents ten specific robust design principles with specific methods and examples for each.The ten robust design principles are:

1. Design in margin,
2. Achieve target,
3. Minimize variability,
4. Characterize and minimize noise effects,
5. Design to the flats,
6. Use parameter combinations,
7. Optimize designs and processes,
8. Use interactions for tuning,
9. Standardize parts, processes and designs, and
10. Eliminate process steps, processes, and parts.

Every robust design concept can be used during product and process development. Some will be used with greater frequency, some will be applied less; however, if the engineering community is not aware of the tool set they will fail to understand how these methods can be applied to improve the overall quality of their products.