Screening Design
A DoE plan aimed at filtering the truly effective factors out of many candidates — not at optimising.
Also known as: screening experiment, factor-screening design
A screening design studies many factors together with few runs to separate the "vital few" from the "trivial many". The goal is not the optimum but a reduction in factor count for follow-up optimisation designs.
- Fractional Factorial 2ᴵᴵᴵᵏ⁻ᵖ — classical choice with 2 levels per factor; small run count, main effects aliased with interactions
- Plackett-Burman — very efficient for many factors (up to ~20) in multiples of 4 runs; also only main effects identifiable
- Definitive Screening Design (DSD) — 3 levels per factor, additionally detects curvature and isolated two-factor interactions; attractive for moderate factor counts (~6–12)
See also
Used in
In the Algorithm Lab
Sources
- Montgomery, D. C. — Design and Analysis of Experiments, 10th Edition, Chapter 8
- Jones, B. & Nachtsheim, C. J. — A Class of Three-Level Designs for Definitive Screening in the Presence of Second-Order Effects, Journal of Quality Technology, 2011
- Kleppmann, W. — Versuchsplanung, 10. Auflage, Hanser