It seems like there may be some confusion regarding your reviewer's comment. Since CCD itself doesn’t inherently include redundant points, it's important to clarify what exactly the reviewer meant by "repetitive runs."

I suggest re-examining their feedback and asking yourself:

To improve your question here, it might help to specify:

• Whether the reviewer identified specific points in your design that appear repetitive.
• Whether the duplicated results are due to Fluent treating close values as identical.

Providing this additional detail will make it easier for others to offer meaningful advice.

I think I understand the point.

Designs like CCD or BBD are typically developed for physical experiments where one encounters variance, noise, and similar issues. Replications are used to counteract these effects.

If I understand correctly, your simulation doesn't account for such variances, meaning that all results for a given set of input points are identical. Consequently, no replications are needed. In this sense, your data is deterministic.

For deterministic data, space-filling designs are recommended, such as Latin Hypercube Sampling. In these types of designs, you don't have replicates; the focus is more on exploring the design space. This means you have no redundant runs, which in turn saves computing power. Space-filling designs also allow for fitting higher-order polynomial models.