Product Engineering
Resolution V aliases only three-factor interactions — which in a finishing process are almost never real.
— runs
Fractional design, replicated, plus centre points
The resolution formula is a heuristic that holds for the standard minimum-aberration designs and can overstate what an arbitrary fraction achieves — the alias structure of the specific generator is what matters and must be checked before running anything. Effects counted as estimable assume the resolution supports them, which fails below Resolution V where two-factor interactions confound with each other. Centre points detect curvature but do not let you model it; a significant curvature result means the design must be augmented to a response surface. Randomisation of run order is not optional and is not represented in any of these numbers.
Design of Experiments Factor & Run Planner — free, with the formula and a worked example, at Textile School.