Screen variables with an efficient experimental design tool. Measure effects, rank drivers, and inspect behavior. Visual outputs and exports support faster, smarter process screening.
Use the responsive grid below. It shows three columns on large screens, two on medium screens, and one on mobile screens.
Grand mean: \( \bar{y} = \frac{\sum y_i}{N} \)
Main effect for factor j: \( E_j = \frac{\sum y_{+} - \sum y_{-}}{N/2} \)
Screening sum of squares: \( SS_j = \frac{N \times E_j^2}{4} \)
Contribution percentage: \( \%C_j = \frac{SS_j}{\sum SS} \times 100 \)
A Plackett Burman design is a screening design. It estimates main effects efficiently, but interactions are typically confounded with those main effects.
Example screening study with 8 runs and 4 factors. The response could be yield, strength, conversion, or cycle time.
| Run | Factor A | Factor B | Factor C | Factor D | Response |
|---|---|---|---|---|---|
| 1 | +1 | +1 | +1 | +1 | 60 |
| 2 | -1 | +1 | -1 | +1 | 48 |
| 3 | +1 | -1 | -1 | +1 | 52 |
| 4 | -1 | -1 | +1 | +1 | 43 |
| 5 | +1 | +1 | +1 | -1 | 54 |
| 6 | -1 | +1 | -1 | -1 | 41 |
| 7 | +1 | -1 | -1 | -1 | 47 |
| 8 | -1 | -1 | +1 | -1 | 39 |
It estimates main effects from a screening design, ranks factors by influence, shows approximate signal strength, and recommends settings for maximizing or minimizing the chosen response.
It screens many variables with relatively few runs. This helps identify likely important factors before using more detailed optimization or response surface studies.
No. These designs mainly target main effects. Two-factor interactions can be mixed into those estimates, so you should confirm important factors with a follow-up experiment.
The calculator subtracts the average response at the low level from the average response at the high level. A positive effect means the response rises at the high setting.
The recommendation follows the sign of each main effect and your objective. Positive effects support higher levels for maximization and lower levels for minimization.
It is a screening aid, not a final confirmatory test. The calculator estimates effect error from dummy columns or small effects, then reports an approximate significance measure.
Unused columns become helpful dummy columns. They act like noise estimators and improve effect error estimation, which makes the screening summary more informative.
Run confirmation experiments, add center points if needed, and move to a richer design such as factorial or response surface methods for optimization.
Important Note: All the Calculators listed in this site are for educational purpose only and we do not guarentee the accuracy of results. Please do consult with other sources as well.