Example: ink mixing
The team plans to vary mixing temperature, impeller speed and mixing time for independently prepared ink batches. The design has 12 edge-midpoint combinations and three center runs. Viscosity will be measured in mPa·s.
Download the factor catalog (CSV)This is the same example used in Help. The CSV lists factor definitions to enter manually on Inputs; importing it does not configure the design. Use Open example in a practice copy in Help to prefill the example settings.
1. Create the workspace
In the sidebar, open Design of Experiments and create an analysis. Name it, select New blank worksheet, then choose Create analysis. The factor catalog is a reference; no measurement data needs to be imported for this task.
2. Choose the design
On Inputs, choose Create design for Task and Response surface designs for Design family. Choose Box–Behnken for Study type.
3. Enter the factors
Under Factors, use Add factor or remove a row until there are 3 factors. Keep them in the order below. Set Type to Numeric for every factor.
Enter the Factor name, Unit, Low and High values below.
| Factor | Factor name | Unit | Low | High |
|---|---|---|---|---|
| A | mix_temperature_c | °C | 20 | 40 |
| B | impeller_speed_rpm | rpm | 400 | 800 |
| C | mixing_time_min | min | 8 | 16 |
4. Set the run plan
Enter 1 in Complete replicates and 3 in Center points per design. Choose Unblocked for Blocks. Check Randomize within blocks and enter 42 in Random seed.
The preview should show Expected 15 runs in 1 block(s).
5. Generate design
Choose Generate design. A new worksheet opens with the randomized run plan and a blank Response 1 column. Open Results to inspect the design summary and complete run sheet. Keep this generated study separate from the recorded analysis example.
6. Interpret results
Check the run plan
Design summary shows 15 runs: 12 edge-midpoint combinations and three center runs, in one block. At each edge point, two factors are at their low or high settings and the third is at its midpoint. No viscosity has been measured yet.
Read the first run
Execution order 1 is standard-order run 13, a center point. Its settings are 30 °C, 600 rpm and 12 minutes. A center can appear anywhere in the randomized order; it is not required to run last.
Keep the design region in mind
This design does not test combinations with all three factors at their extremes. A model prediction at an untested corner has less direct experimental support than a result at a sampled setting. Check the complete run sheet before choosing a region to study.
Conduct and record the study
Prepare independent ink batches in the generated execution order. Measure each batch’s viscosity and enter it in Response 1 against its actual run ID. Keep point types and factor settings together; another experiment’s randomized row order is not a valid response mapping.
More details: read each result section
Complete result tables use six significant digits, as in Help.
Design summary
Family identifies the design. Runs is the total, 15; Base runs is the generator’s base count, 15. Replicates and Blocks record the chosen structure. Actual seed is 42; keep Generator version with it when reproducing the plan. Alpha applies to axial distance in a central composite design. Resolution describes aliasing in a fractional factorial; a dash means the statistic is not applicable.
| Statistic | Value |
|---|---|
| Family | Box–Behnken |
| Runs | 15 |
| Base runs | 15 |
| Replicates | 1 |
| Blocks | 1 |
| Alpha | — |
| Resolution | — |
| Actual seed | 42 |
| Generator version | doe-v2.1 |
- Unavailable settings are shown as —; the saved design rows remain authoritative.
Factor settings
Each row gives the Factor name, modeling Type, Unit and entered Low and High bounds. All factors here are numeric. A dash under Levels means no additional explicit level list was entered. Midpoints are factor settings for center runs, not measured responses.
| Factor | Type | Unit | Low | High | Levels |
|---|---|---|---|---|---|
| mix_temperature_c | numeric | °C | 20 | 40 | — |
| impeller_speed_rpm | numeric | rpm | 400 | 800 | — |
| mixing_time_min | numeric | min | 8 | 16 | — |
Run sheet
All generated runs are shown in execution order. Row ID identifies a run; Standard order identifies its design position; Execution order gives its planned sequence. Block groups runs and Replicate identifies repeated designs. Point type distinguishes factorial, star (axial), edge or center points, as applicable. The remaining columns are actual factor settings in their stated units. The generated worksheet’s Response 1 starts blank; no response values appear in this plan.
| Row ID | Standard order | Execution order | Block | Point type | Replicate | mix_temperature_c | impeller_speed_rpm | mixing_time_min |
|---|---|---|---|---|---|---|---|---|
| doe-13 | 13 | 1 | 1 | center | 1 | 30 | 600 | 12 |
| doe-12 | 12 | 2 | 1 | edge | 1 | 30 | 800 | 16 |
| doe-15 | 15 | 3 | 1 | center | 1 | 30 | 600 | 12 |
| doe-14 | 14 | 4 | 1 | center | 1 | 30 | 600 | 12 |
| doe-9 | 9 | 5 | 1 | edge | 1 | 30 | 400 | 8 |
| doe-6 | 6 | 6 | 1 | edge | 1 | 20 | 600 | 16 |
| doe-7 | 7 | 7 | 1 | edge | 1 | 40 | 600 | 8 |
| doe-10 | 10 | 8 | 1 | edge | 1 | 30 | 400 | 16 |
| doe-11 | 11 | 9 | 1 | edge | 1 | 30 | 800 | 8 |
| doe-1 | 1 | 10 | 1 | edge | 1 | 20 | 400 | 12 |
| doe-5 | 5 | 11 | 1 | edge | 1 | 20 | 600 | 8 |
| doe-3 | 3 | 12 | 1 | edge | 1 | 40 | 400 | 12 |
| doe-8 | 8 | 13 | 1 | edge | 1 | 40 | 600 | 16 |
| doe-2 | 2 | 14 | 1 | edge | 1 | 20 | 800 | 12 |
| doe-4 | 4 | 15 | 1 | edge | 1 | 40 | 800 | 12 |
7. Create a report
Keep Design summary, Factor settings and the complete Run sheet. Record the actual seed, generator version, units and planned execution order. This report documents a design; add measured responses and their analysis separately after conducting the study.
Choose Add to report from the completed run. Open Choose individual items and advanced options to select tables and charts, then choose Add to draft. Hidden items stay in the saved analysis.
Name the report and add the purpose of the study and your interpretation. Choose Save, review Preview, then use Export → Download HTML. The example below contains the full selected content, including results beyond one screen.