I-Optimal Design Calculator

Here you can create a custom I-optimal design online. Define your numeric or categorical factors, enter the available levels, choose the number of runs, select the model you want to estimate, and exclude forbidden factor combinations if needed.

I-Optimal Design

An I-optimal design is a design of experiments (DoE) method that selects experimental runs from a candidate set to minimize the average prediction variance over that candidate space. It is especially useful when the main goal is predicting the response well across the design region.

When to Use an I-Optimal Design

• You want a design that supports prediction and response optimization.
• The available factor combinations are constrained or irregular.
• You need to limit the number of experimental runs.
• Your experiment includes numeric factors, categorical factors, or both.

How to Create an I-Optimal Design

1. Select Custom Optimal Design in the DoE calculator.
2. Add your factors, select their types, and enter the available levels.
3. Choose the number of runs and the model that should be estimated.
4. Select I-optimal as the criterion.
5. Add forbidden combinations if some candidate runs must be excluded.
6. Generate the design and export the resulting test plan to Excel.

If your main goal is precise estimation of model coefficients, use a D-optimal design. For constrained experiments where you want to compare both criteria, use the custom optimal design calculator.