Custom design question

[#1]

I'm creating a DOE and haven't in a long time and I can't remember, For continuous levels you input low and high (-1 and 1) and when you make your design it generates different runs. So 3 question, if my variable is temp 50 to 100, how does my model explore temps between (if it's include center points how do I judge how many with such a sliding scale?). Also if my goal is to hit a target (say a distance of 2m) should I input limits? and Is there anyway to tie that response to the variables or does the model do that for me?

[#2]

Hi @hallef18 ,

1. You need to specify quadratic (or higher) effects in the Model outline to get values in between the low and high extremes. Click on RSM or Powers > 2nd. The default is the main effects model and the optimal settings to estimate this model will always be only at the high and low extremes.

2. Yes, you can add the limits for your responses. This will not affect the design that is created but it is information that is "remembered" later when you have run the experiment and built a model.

3. Yes, the model ties the response to the factors. It describes the relationship between them. It enables you to understand how changes in the factors affect the responses. This is the key end result of your experiment.

I hope all this helps. You might want to get some more background on design of experiments. I recommend the Statistical Thinking for Industrial Problem Solving course, which includes a module on DOE.

Phil