What inspired this Wish List Request?
When customers create I-optimal design (response surface designs), there are two possible problems.
- When you create an I-optimal design, if the optimal point is not centered, the prediction variance will be inflated (since many center points are designed in i-optimal design).
- Sometimes G-efficiency is low
Sometimes an A-optimal design solves these problems, but users have to change the weights by trial and error.
Please describe the new or improved feature or solution you would like to see.
Customer want to generate response surface designs with high I and G efficiency.
For example, the approach for compromised design is suggested as in the following paper.
https://www.tandfonline.com/doi/full/10.1080/08982112.2023.2194963
Better ability to consider compromise plans (Pareto front type), including G- and I-optimality.
Easier ability to create and compare different weights for A-optimality (using Design Explolre).
Here is another list of their requests:
- They want to use Design Explore when they do split-plot design.
- The fraction of design space plot is useful for evaluating the design, so it's great if Design Explore could generate this plot.
- It's great if Design Explore has the ability to visualize several criteria at once (e.g., G Efficiency vs. I Efficiency plot).
- It would be nice to have an some indicator that shows whether the design points are clustered at the center point or at the end points (He calculates the squares of the X matrix, but any other good criteria are welcome).
What would the value be to you and/or other users if this new feature is provided?
Customer can create design to predict optimal settings well. To my knowledge, there is no software that can do that.
Is this an external customer request?
W.L. Gore & Associates, Co., Ltd.Tatsuo Yamamoto from Japan.
Is there a competitive software package that has this feature or solution that we should be aware of?
To my knowledge, no.
Will this impact a renewal, expansion, or adoption at a customer, and does this put an account at risk if we do not address this?
It helps user expansion.
