A design of experiments approach to set creation is described for implementation of set-based design (SBD). The design points in the experimental design space generated during set creation represent individual design alternatives exhibiting unique characteristics and performance. Exploration and evaluation activities leading to design selection in SBD are based on the feasibility and viability of the design points, so the way they are generated is important. Several Latin hypercube designs are presented and summarized. An application of Latin hypercube sampling is demonstrated as a method of set creation, which produces sets of design alternatives that are depicted as a point cloud with a level of design space coverage that is not possible with traditional curve fit or response surface techniques.

This content is only available via PDF.
You can access this article if you purchase or spend a download.