If your expected effects are at least 3x the standard deviation of the response then the power is quite high with the 12 runs:
![power.JPG power.JPG](https://community.jmp.com/t5/image/serverpage/image-id/20499iC4AA9F97F6C76418/image-size/large?v=v2&px=999)
This result can be expected even with the correlations (+/- 1/3) seen here:
![Correlations.JPG Correlations.JPG](https://community.jmp.com/t5/image/serverpage/image-id/20500i419E6724CA5A44C1/image-size/large?v=v2&px=999)
The correlations result in NO bias of the estimates. The correlations inflate the variance of the estimates, which causes a lengthening of the confidence intervals:
![estimation efficiencu.JPG estimation efficiencu.JPG](https://community.jmp.com/t5/image/serverpage/image-id/20501i7D56E2AE162E43FC/image-size/large?v=v2&px=999)
The CI are about 6% longer, a very modest inflation. I don't think these correlations compromise the design performance much.