turn on suggestions

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

Showing results for

- JMP User Community
- :
- Discussions
- :
- Adding conditional constraint while creating the d...

Topic Options

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

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

Aug 25, 2015 9:26 PM
(1109 views)

Hi,

I am creating a design and I want to add one constraint. I have 7 continuous variables and of which 2 variables are related by the below constraints I wish to define.

If(X3=1) then (X5 is not equal to -1)

If(X3=-1) then (X5 is not equal to 1)

Basically I don’t want these two factors to be at opposite extremes at the same time.

If(X7=1)then (x4 is not equal to 1) = Here I don’t want both the factors to have high value simultaneously.

Can you please help me to define these constraints.

1 ACCEPTED SOLUTION

Accepted Solutions

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

Solution

Vinay: The most efficient way to accomplish your request is to use the JMP Custom Design path and then within the design set up phase of the work flow you'll be offered the opportunity to "Define Factor Constraints". See below (from JMP version 12...earlier versions will look a bit different):

If you are using JMP version 12 (as shown above) you'll have three paths by which you can define the constraints you desire. My personal preference is to select the "Use Disallowed Combinations Filter" because this path uses JMP's long standing Data Filter construct to define the disallowed combinations. To me this path is more intuitive with the slider bars or ability to enter values in design units compared to the "Specify Linear Constraints" path...but your mileage may vary.

Here is the URL to the JMP version 12 online documentation that explains in detail the various paths to defining factor constraints:

Then once you have your design created, I suggest constructing some simple JMP Fit Y by X plots of the factors included in the various constraints to double check you've actually excluded certain regions of the design. Sometimes it's hard to verify these exclusions by starring at the design table alone.

Here is a link to a nice JMP On Demand Webinar that a colleague of mine, Don McCormack facilitates that describes additional details...especially if you want to use the "Specify Linear Constraints" path:

Advanced Design of Experiments | JMP

2 REPLIES

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

Vinay: The most efficient way to accomplish your request is to use the JMP Custom Design path and then within the design set up phase of the work flow you'll be offered the opportunity to "Define Factor Constraints". See below (from JMP version 12...earlier versions will look a bit different):

If you are using JMP version 12 (as shown above) you'll have three paths by which you can define the constraints you desire. My personal preference is to select the "Use Disallowed Combinations Filter" because this path uses JMP's long standing Data Filter construct to define the disallowed combinations. To me this path is more intuitive with the slider bars or ability to enter values in design units compared to the "Specify Linear Constraints" path...but your mileage may vary.

Here is the URL to the JMP version 12 online documentation that explains in detail the various paths to defining factor constraints:

Then once you have your design created, I suggest constructing some simple JMP Fit Y by X plots of the factors included in the various constraints to double check you've actually excluded certain regions of the design. Sometimes it's hard to verify these exclusions by starring at the design table alone.

Here is a link to a nice JMP On Demand Webinar that a colleague of mine, Don McCormack facilitates that describes additional details...especially if you want to use the "Specify Linear Constraints" path:

Advanced Design of Experiments | JMP

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

Aug 28, 2015 1:54 AM
(875 views)