cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Try the Materials Informatics Toolkit, which is designed to easily handle SMILES data. This and other helpful add-ins are available in the JMP® Marketplace
Choose Language Hide Translation Bar
Richard_Davies
Richard_Davies
Level I

Predicting the effect of a process on a number of variables

Created: Jul 15, 2022 04:37 AM  |  Last Modified: Jun 8, 2023 5:52 PM (824 views)

Hi Everyone,

I have a query that I'm hoping JMP and the community can resolve.

I work with powders and powder rheology for additive manufacturing and use JMP to analyse rheology data. Most of our raw materials require a treatment to flow in the process, and I can see trends between the untreated and treated powders from the rheology data. I have a small dataset of powders before and after treatment, and can predict process performance using partition models or neural networks with some accuracy.

Is there a method in JMP to predict the effect of "treatment" on a raw material based on the data set. 

I'm using the standard JMP v15.

 

Thanks,

Richard

0 Kudos
1 REPLY 1
P_Bartell
P_Bartell
Level VIII

Re: Predicting the effect of a process on a number of variables

Jul 15, 2022 07:33 AM (809 views)  |  Posted in reply to message from Richard_Davies 07-15-2022

Maybe I'm missing something wrt to how you've stated your question, "Is there a method in JMP..." So is your question around trying estimate the effect of 'treated' vs. 'untreated'? If so, simple adding a parameter to any of JMP's modeling techniques called say, Treatment?, with two levels 'treated' and 'untreated' (and any interaction effects as well if using linear modeling methods) and you should be able to estimate those parameters as long as you have sufficient degrees of freedom (you mention a small dataset...so df may become an issue). Or maybe I've just misinterpreted the information you've provided?

0 Kudos