Beginner Question

[#1]

Hi community, 

this is my first post ever in a forum and also the first time I'm using DoE for experiments. I have watched the DoE Video from JMP and have a question or rather look for confirmation: 

I have an experiment with two continuous factors (e.g. range of pHs) and one categorical factor (e.g. type of acid). I want to use RSM but apparently, it won't work when a categorical factor is included. Hence, I have decided to go for a "Custom Design" and included all the factors. Now the system is saying "Categorical factors cannot appear in polynomial model terms. Only the quadratic terms for continuous factors are being added to the model.".

Would a more senior DoE user say this is a correct approach? And is there anything I need to keep in mind? Alternatively, I was thinking of doing it like all the others and assume the categorical factor (e.g. type of acid) as a continuous factor. But this just doesn't make sense in my point of view. Thank you for any help.

[#2]

@P_Bartell Thank you for pointing that out. I have another question. Why are the surface plots different when generating them over "Graph" or via the "Model" -> "Fit least Square"? Not from a result point of view but from the shape. How can I make it smooth and not so edgy? 

Thank you in advance and kind regards.

[#3 · Accepted solution]

Yes, exactly. Try to achieve the level given by the design but when it deviates a lot, like 12 instead of 14, then just update the level for the analysis.

[#4]

To add just a bit to @Mark_Bailey 's sound guidance...strictly speaking you can create a what I'll call pseudo RSM (think like a full factorial plus central composite as an illustrative example) design with categorical and continuous factors...but, and here's the BIG BUT, the design will be very inflexible and in all likelihood have a much larger number of runs, with a predefined model underlying the design...I won't go into how you could create the design because @Mark_Bailey 's advice is way more efficient and rational. Use the JMP Custom Design platform to create your design.

[#5]

First, welcome to the community.  I have some thoughts on the questions you pose, but to be honest, you have not provided enough information regarding the actual situation for me to provide most appropriate advice.  For example, you don't describe the response variable.

My thoughts:

1. RSM is meant to "map" the design space so you may select optimum conditions for the selected response variable.  It should be done after you have thorough knowledge regarding which, of the few, factors have a significant effect on the response and more importantly, you understand the effect of noise (factors you are not willing to manage or control).  You should already have a first order model...

2. Creating a continuous regression model with a categorical factor is a bit non-sensical.  You could create a model for each category, but including the term in the model makes no sense.  You don't have an inference between the categories nor could there be any non-linear terms. (it's either this acid or that acid type, there is no continuum)

3. You could hypothetically convert acid type into a continuous variable if you can quantify the differences in the types (e.g., amount of a certain chemical compound), but my guess is your types are hydrochloric vs. sulfuric or nitric or something like that.

Are you sure you need the non-linear terms?  Have you predicted what the model should be (based on science, not statistics)?  What are your hypotheses?  Typically we build models starting with first order and augmenting this through iteration, rather than a one-shot experiment.  We experiment to understand, not to pick a winner.

Experimentation is extremely powerful, so have at it. 

[#6]

Hi Everyone, I am trying to find a word "PASS" in column Test. then I want to be able to write another word "FAIL" in a cell just above the PASS in same column

Thanks for your help

Sam

[#7]

dt=current data table();
theRows=dt<<get rows where(:text=="PASS");

:test[theRows[1] -1 ] = "FAIL";

[#8]

Hi Jim, thanks for your help, the code works only on one row, I have few hundred similar rows that need to be updated

Thanks

Sam

[#9]

I tried to put it in a loop, still does only the first "PASS" it finds, it won't continue to find the next "PASS"

Thanks

Sama

dt = Current Data Table();
R = N Rows( dt );
For( i = 1, i <= R, i++,
    theRows = dt << get rows where( :test == "PASS" );
    :test[theRows[1] - 1] = "FAIL";
);

[#10]

Thanks Jim, I was able to figure it out; just needed to replace the [1] by [0] on the 2nd line.

Sam

    theRows = dt << get rows where( :test == "PASS" );

    :test[theRows[0] - 1] = "FAIL";