Taguchi's method don't involve all variable combination among input factors. Therefore RSM is preferred. For three factors, the Box-Behnken design offers some advantage in requiring a fewer number of runs. For 4 or more factors, this advantage disappears. Therefore, CCD in RSM will be better for your design.

It depends on your experiment variables.

CCD can increase an existing factorial design, while the BBD cannot.

BBD can avoid combined factor extremes, but it has some missing corners. That could be led to worse quality of the regression model.

With 5 factors, you are required to do more experiments (50 runs) if you used CCD, but 46 runs if you used BBD.

I recommend reviewing the information related to your question on the following websites:

Article Response Surface Methodology

https://support.minitab.com/en-us/minitab/18/help-and-how-to/modeling-statistics/doe/supporting-topics/response-surface-designs/response-surface-central-composite-and-box-behnken-designs/

https://develve.net/Response_surface_methodology.html

You also can read this book:

Response Surface Methodology: Process and Product Optimization Using Designed Experiments.

https://www.amazon.ca/Response-Surface-Methodology-Optimization-Experiments/dp/1118916018

Amr M. Bakry Riyadh Abdulmajid Saleh Thabit Dr-Nabil Al-Hajj Jacob Ojobi Omedi Junaid Haider Junaid Haider

Maybe my answer is a little late, but using Statistica, Desing Expert and mini-tab are more recommended.

For me, Develve is not useful.

Regards!