Robust Designs Through Partially Clear Two-Factor Interactions
主 题: Robust Designs Through Partially Clear Two-Factor Interactions
报告人: Prof. Boxin Tang (Simon Fraser University,Canada )
时 间: 2010-07-09 上午9:00 - 10:00
地 点: 理科一号楼 1418
Orthogonal arrays with clear two-factor interactions are very attractive for industrial experiments as they provide a class of designs that are robust to nonnegligible two-factor interactions. If prior knowledge suggests that some two-factor interactions are negligible before running the experiment, then designs can be found that allow additional factors to be studied while remaining robust to the nonnegligible two-factor interactions. This is done through partially clear two-factor interactions. We study the existence and construction of such robust designs in this paper. Several construction results are presented and examples are provided for illustration. We also prove a result that establishes an upper bound on the maximum number of clear two-factor interactions in an orthogonal array. This is a joint work with my student Ryan Lekivetz.
