5 - Screening designs

Abstract

Screening designs are used for screening a large number of process or design parameters to identify the most important parameters that will have significant impact on the process performance. Once the key parameters are identified, subsequent experimentation can be performed using these parameters to understand and analyse the nature of interactions among them using full/fractional factorial designs and response surface methods, if necessary. Plackett–Burman (P–B) designs allow the experimenters to evaluate a large number of process/design parameters in a minimum number of trials (i.e. with minimum budget and resources). One of the stringent assumptions experimenters make is the unimportance of interactions in the early stages of experimentation.