Two-Level Fractional Factorial Designs

Abstract

Fractional factorial designs with factors at two levels are the most commonly used in practice. For the same number of factors, they have smaller run size than designs at more than two levels. This run size economy makes them attractive for studying a large number of factors. A fundamental question in this context is the choice of designs. The minimum aberration criterion is commonly used for selecting optimal designs. Theoretical results on minimum aberration designs are given in this chapter. Results on related criteria like maximum resolution and maximum number of clear effects are also considered. A catalogue of two-level fractional factorial designs with 16, 32, 64, and 128 runs is given.