DOE (Design of Experiments) Dependent Factors

J

javgut15

Hello:

I have a complex simulation system. With a DOE I want to see the influence of different variables on my system.

My problem is the following:
I have six factors. Two of them are the upper limit and lower limit for the acceleration. So two factors make reference to the same variable of the system. My problem is that upper limit must be always higher than the lower limit, but their possible value ranges have some values in common (for example lower limit between 1 and 10 and upper limit between 5 and 15). What I want is to introduce this restriction in Minitab when I create the DOE in order not to have impossible designs. What is the best way to do it in Minitab?

Thank you very much in advance.
 

Statistical Steven

Statistician
Leader
Super Moderator
Hello:

I have a complex simulation system. With a DOE I want to see the influence of different variables on my system.

My problem is the following:
I have six factors. Two of them are the upper limit and lower limit for the acceleration. So two factors make reference to the same variable of the system. My problem is that upper limit must be always higher than the lower limit, but their possible value ranges have some values in common (for example lower limit between 1 and 10 and upper limit between 5 and 15). What I want is to introduce this restriction in Minitab when I create the DOE in order not to have impossible designs. What is the best way to do it in Minitab?

Thank you very much in advance.

One solution to this problem is to have a single factor called acceleration with four levels (i.e. 1/5, 1/15, 10/11, 10/15). In other words, find four combinations of the two factors that define your design space. Just a thought.
 
Top Bottom