Sample Size for DOE

[#1]

Hello,

I write you to have some suggestion regarding on topic linked to the DOE.

I have a industrial process. The outout (response) of the process is the percentage of scrap.

The average of this scrap is around 1.5%.

I want to create a DOE to identify main factors, interactions, model,...

My question is: how I can determine the number of the replicates for each corner (for a specific power) to be sure to have correct statisical results (if in one trial I will obtain "0% of scrap" it is real output change or only a case)?

Thanks in advance for your feedback.

Best Regards,

Simone

[#2]

My suggestion is to be more creative in developing an appropriate response variable (and use multiple responses).  If color is the issue, there are multiple ways to measure this:

• ∆L: the difference in lightness/darkness
• ∆A: the difference in red/green values
• ∆B: the difference in yellow/blue values
• ∆C: the difference in chromaticity
• ∆E: the spectral “distance” between the two colors
• ∆H: the difference in hues

If human perception is of interest, then an ordinal scale might be very useful.  I have attached a paper I wrote to provide guidance on using such a scale.  Binary responses are not very descriptive, require larger sample sizes and lack discrimination.

[#3]

I understand your suggestion regarding the usage of some "∆" and tranform "OK/KO" in some data measurable.

Not easy... but not impossible.

Thanks also for the paper.

Have you a good day.

Best Regards,

Simone