Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

- JMP User Community
- :
- Discussions
- :
- DOE custom design factor combinations

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

Highlighted

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Apr 8, 2019 2:55 AM
(3228 views)

I am designing an experiment (2 level DOE) where my factors come from two sources, an incubator that has 4 factors to be controlled and shake flasks with 3 factors. I have 3 incubators and each one can hold 12 shake flasks. I want to implement a design where the shake flask settings (12 shake flasks each with different settings) are the same for each incubator setting. Is this possible in jmp?

1 ACCEPTED SOLUTION

Accepted Solutions

Highlighted

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

When you define the factors, change the Changes setting from Easy to Hard for the factors around the incubator.

Learn it once, use it forever!

2 REPLIES 2

Highlighted

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

When you define the factors, change the Changes setting from Easy to Hard for the factors around the incubator.

Learn it once, use it forever!

Highlighted
##

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Re: DOE custom design factor combinations

You can replicate the runs over the three incubators, from a DOE point of view it would be less efficient (you use more runs than strictly necessary), compared to considering 8 independent factors, incubator entity being one of those.

I can imagine you may have practical reasons to do the same thing on each incubator. In these situations, I like to construct a DOE table and use the "Evaluate Design" feature. The platform already provides comparison to full factorial design, I always look at "fractional increase of confidence interval". You can also compare to other scenarios, e.g., a D-optimal design with the same number of runs.

I can imagine you may have practical reasons to do the same thing on each incubator. In these situations, I like to construct a DOE table and use the "Evaluate Design" feature. The platform already provides comparison to full factorial design, I always look at "fractional increase of confidence interval". You can also compare to other scenarios, e.g., a D-optimal design with the same number of runs.

Article Labels

There are no labels assigned to this post.