Subscribe Bookmark RSS Feed

How to look for optimum mean with low variation from DOE

albiruni81

Community Trekker

Joined:

Jul 15, 2014

Hi All,

 

May i know what is the method that we can do in order to find the optimum mean from our DOE which has the lowest possible variation?

 

Rgrds

 

Irfan

18 REPLIES
ian_jmp

Staff

Joined:

Jun 23, 2011

(I'm assuming that you have already collected your data).

 

You could view the second video (by Robert Anderson, 'Using historical production data to identify manufacturing process improvements') at this link (SAS profile required). It uses observational data, but the mechanics of how to use JMP are the same.

 

albiruni81

Community Trekker

Joined:

Jul 15, 2014

May i know what is the JMP version that Robert Anderson is using in his webcast?
Peter_Bartell

Joined:

Jun 5, 2014

At a high level, there are any number of possible ways to arrive at a solution which I like to call 'on aim, with minimum variability'. Using DOE you could take a purist Taguchi style approach and use his signal to noise ratios (lots of reasons to avoid this method...but I don't want to turn this thread into a Taguchi vs. Classical methods discussion). Another approach is to model the mean and variance of the response as two separate and distinct responses...then using JMP's co-optimization capability to help balance deviation from target with minimum variance. A third approach is through simulation.

albiruni81

Community Trekker

Joined:

Jul 15, 2014

Hi Peter, do you have any example of how we do model the mean and also the variance of the response?Since for the variance we need multiple data from the same condition runs. Are you able to elaborate on the simulation approach what do you mean by this
Peter_Bartell

Joined:

Jun 5, 2014

Myers et. al. give a good overview of simultaneously modeling mean and variance in this reference:

 

http://amstat.tandfonline.com/doi/abs/10.1080/00031305.1992.10475869?src=recsys

 

Essentially it's no different than modeling two responses...of course you'll need replication within your design to estimate the variance for each treatment combination. All the usual co-optimization, simulation tools in the Fit Model platform, for the specific modeling personality you choose, will come into play.

 

On the simulation side of things one path is within the JMP Prediction Profiler (which I assume you'd use, since you can fit a model of your experimental results) you can use the Simulator from the Profiler framework to assign target values for each predictor variable, distributional forms for each variable, and estimates of mean and variance for predictors. Add other sources of noise as you see fit. You can even run a simulated experiment with the assumed mean and variance for each factor setting within the Profiler...so lots of different ways to go at this from a simulation point of view.

markbailey

Staff

Joined:

Jun 23, 2011

You can always used the log-linear variance model through the Fit Model dialog for this purpose. Change the fitting personality to Loglinear Variance. Then you define the linear predictor for the Main Effects (mean) and another for the Variance Effects as you normally would for a multivariate linear model. This way you also have a profiler for each. You can save the fitted models as column formulas and then use them in other platforms outside of the fitting.

You do not need replicates for this model, but they help. You do need a large number of degrees of freedom for the error.

Read more about it in Help > Books > Fitting Linear Models. Chapter 10 is devoted to this platform.

Learn it once, use it forever!
albiruni81

Community Trekker

Joined:

Jul 15, 2014

Dear Mark,

Any webcast showing how to use the loglinear variance
Jeff_Perkinson

Community Manager

Joined:

Jun 23, 2011

I don't know if we have a video demonstrating the Loglinear Variance Model but the documentation has a good example.

-Jeff
albiruni81

Community Trekker

Joined:

Jul 15, 2014

Hi mark, my company is still using JMP 5.1.2, from the loglinear example it shows that an example from the injectionmolding datasets. According to the example it was determined from the screening design example 'Preliminary investigation determined that the mean response only seemed to vary with the first two factors, Mold Temperature, and Screw Speed, and the variance seemed to be affected by Holding Time', then the question I have is how does the screening design able to show this response