Showing results for 
Show  only  | Search instead for 
Did you mean: 
Submit your abstract to the call for content for Discovery Summit Americas by April 23. Selected abstracts will be presented at Discovery Summit, Oct. 21- 24.
Discovery is online this week, April 16 and 18. Join us for these exciting interactive sessions.
Choose Language Hide Translation Bar

JMP Academic Webinar - Teaching Design of Experiments with JMP

Published on ‎12-09-2022 01:10 PM by Staff | Updated on ‎06-21-2023 03:37 PM

Post questions or comments below. Download the JMP Journal file in the Attachments section to the right.


Design of experiments (DOE) is a critical skill to develop in the modern science or engineering student. JMP's world-class suite of DOE tools is used extensively in industry, and JMP's visual, no-code interface makes its DOE tools great for classroom use, too. This webinar reviews the use of JMP for teaching DOE, with a focus on both classical and modern optimal designs, analysis of DOE data, and free DOE teaching resources offered by the JMP Academic Program.

Helpful Links

Register for upcoming webinars:

Free 30-day trial:

JMP Academic Program:

Free teaching materials:


Contact the JMP Academic Team at

Tue, Feb 7, 2023 01:00 PM EST
Tue, Feb 7, 2023 02:00 PM EST

How to put many responses in DOE when i use Taguchi plan?


@cedrick07, Taguchi designs typically target only a single response, and JMP's Taguchi Array allows only one response. You could try creating a composite response, using some function (a weighted average, for example) to meaningfully combine all your responses into one. You also could use a multi-objective optimization approach, where you collect all responses, then analyze each separately in Fit Model, then use the general Profiler under the Graph menu to perform a tradeoff analysis.


Taguchi designs have fallen out of favor with many statisticians. If you are considering Taguchi designs because you need to optimize the within-specification rate in the presence of noise factors, you could consider running another type of design that supports multiple responses (perhaps a Custom Design), manipulating both the noise and control factors as in a Taguchi desing, and then using a simulation experiment to identify optimal settings given inherent variation in the noise factors. See this video for more information on this approach:


Ross Metusalem
JMP Academic Ambassador

Thank you very much for your reply.
I am a student and I would like to know how I can get assistance in setting up my experience plan.

I have three different factors which I want to study in terms of their influence on six different responses. But for reasons of limited finance, I want to have fewer samples with three levels each. I've tried the Taguichi method but I'm not satisfied. Also, with the response surface design, but I don't know how to insert the six responses and also the three levels of each factor in JMP.

If you are on a limited budget, you should consider the Custom Designer, which will provide the optimal design given restrictions you place on the run size (among other design parameters). It easily accommodates multiple responses.


You might work through the DOE Intro Kit to get familiar with the Custom Designer, and then view one or more Mastering JMP webinars on DOE to learn more of the details. If you have further questions to pose to the JMP User Community, I recommend posting them to the general discussion board.

Ross Metusalem
JMP Academic Ambassador

So, i want to know how could i create a composite response if I use Taguchi Plan?

You'd first want to determine whether creating a composite response makes sense conceptually. It may be the case that the individual responses cannot be combined mathematically in a meaningful way. But if you do determine a meaningful way to combine them, you could use JMP's Formula Editor to create the formula to calculate the composite response. You would use that formula in the response column of your Taguchi design table.

Ross Metusalem
JMP Academic Ambassador

hey, i can't reproduce the 13 runs in the Definitive Screening Design demo. I could only get 17 runs even after clicking "No Blocks Required". Could you explain the 13 runs please..  




Hi @colliDer88, you are almost there! Where you configure "No Blocks Required", there are also 4 extra runs added by default (for some good reasons, esp. a nice boost of statistical power). Change 4 to 0 extra runs, and you will get a 13 run DSD.