nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

MODDE from Umetrics

Bill_Worley

Does anyone have experience with MODDE from Umetrics? We have a customer asking for a comparison of JMP vs. MODDE. I checked out the website and it appears to be a nice tool for DOE and analysis. The one thing I did not see was a modeling platform that would allow exploration of different types of models or any sort of model comparison. The graphics look a little dated, but they appear to have a nice optimizer with visualization that is nice for determining robust regions.

Any opinions, experience, and/or documentation would be greatly appreciated.

Best,

Bill

Accepted answers

bernd_heinen

Hi Bill,

I used Modde a bit and it's very helpful for the unmexperien´ced user as it guides him through the whole process of setting up a DOE and then analyzing the result. However it is far from beeing flexible. It does well on traditional designs, but the optimal designs do not work properly. Furthermore you can only analyze experiments for which a trial design exists. I you get additional data you have to fake a design in order to evaluate it.

There are no strategies nor functionality that supports exploratory data analysis or to find optimal models among a variety of options as JMP does. Graphs are only available in the context of a designed Experiment you don't have the variety that JMP has. Many users find Modde Graphs very appealing for their simplicity.

Usually it is good enough to point out that JMP can analyze arbitrary data sets even without a corresponding experimental design to win against Modde.

Hope that helps

Bernd

All comments

bernd_heinen

Hi Bill,

Usually it is good enough to point out that JMP can analyze arbitrary data sets even without a corresponding experimental design to win against Modde.

Hope that helps

Bernd

martindemel

And in the user Folder on \\jmpsmd\users2\Heinen\ModdeInfo Bernd has added some more Information he presented few months ago in an SE Meeting

Bill_Worley

Bernd and Martin,

Thank you both. Very helpful!

Best,

Bill

volker_kraft

Just adding the voice of a customer: A JMP user mentioned that for (classic) DOE he chooses JMP or MODDE interchangeably, but if he knows that the analysis will need a more complex / more challenging model (esp. with Mixed Models in mind) he always starts with JMP Pro right away.