Replicate

a complete repeat of the design

A replicate repeats every run of the design, each one set up again from scratch. Replicates give an estimate of the pure error and therefore real p-values, and protect the result against a single bad run.

Measuring the same part twice is a repeat measurement, not a replicate: it only shows the gauge noise.

Calculate it

DOE: 2-Level Factorial Design: 2^k factorial: run plan, effects, Pareto, interaction plots and best settings.

Related terms

• Factorial design (2^k): every combination of the factor levels
• Pure error: noise estimated from replicated runs
• Randomisation: running the trials in random order