Adding a complex constrain to a Custom Design with 4 factors

[#1]

Hello,

I am creating a Custom Design with 4 factors (Factors A, B, C & D).

I have a question about a constraint that has to do with factors B and D.
- Factor B is actually the number of reactions done in sequence in the same experiment. This can be 1 reaction, 2, 3 or more reactions in sequence. I set the range as 1-3 (not higher because I am not interested at the moment in >3 reactions in sequence).
- Factor D is the time I wait in between the reactions in sequence. Thus, I only define Factor D when Factor B is >1. The range is 0.1 to 10 seconds. Factor D would be 0 if Factor B is 1. In other words, when the number of reactions done in sequence (factor B) is just 1, Factor D doesn't exist because it was just 1 reaction and thus, I don't need to wait any seconds since there is no other reaction in sequence performed.

How can I define this constraint in my Custom Design model?

Thank youuuu!!
PS: I use a Mac. 

[#2]

Hi @Victor_G,

Following up on your response from Sept 27 about comparing multiple designs, which you said will be easier to do in JMP17. I was wondering, do you know when we will be able to use this new version of JMP17? It seems a great feature!

1- Also, to make sure I understand this. When you say that you like to compare designs first. You mean on JMP, not in the lab. So you prefer to always compare different designs with JMP DOE before physically testing the design, correct? 

2- I don't recall seeing anything about DOE design comparisons in the "Custom Design of Experiments" course. Would it be possible @Victor_G to provide more information on how you go from "DOE > Design Diagnostics > compare designs" to 1) the power, 2) variance prediction profile and 3) correlation matrix/color map? (I tried it but I am confused on what exactly I should select...)


3- On the same note, what is the meaning of this "Design Diagnostics" information (see screenshot) that one can see right before making the table?

4- Last question Why when I am making the design again, now JMP chose two runs with 'X1' data points as 1.93 and 1.89 (two separate runs) that are not centered at all. X1 is set up as a continuous factor with a range of 1-3. I am okay with the decimals for this factor, however, I wonder why JMP chose those two values so close to each other instead of values more spread out apart in the experimental space?

Thank you very much @Victor_G for your fast and detailed responses!! Super helpful, thank you!!! I am learning  A LOT!

[#3 · Accepted solution]

Hi @ADouyon,

Concerning this new feature in JMP 17, it should be available soon (late October - beginning November), according to the JMP website : New in JMP 17 | Statistical Discovery Software from SAS
There is already a fascinating and brilliant white paper of this new functionality (Design Explorer) available here, which promises very interesting use cases in the selection of an optimal design: Choosing the Right Design - with an Assist from JMP's Design Explorer | JMP

1- Exactly ! Sorry for not being clear, this is exactly what I meant : I prefer to create several designs on my computer with JMP, and choose the most relevant one according to the experimental budget, goal and constraints, instead of going into the lab with the first design created and figuring out later that I may have forgotten some constraints or that my experiments are not all feasible/possible.

2- Sure ! When you have created several designs and the corresponding datatables for each design, you can go to DoE -> Design Diagnostics -> Compare Designs. There, you can select all the designs tables (max 5 designs in total, so 4 selected + the design from where you have clicked on "Compare Designs") and match the factors if they have different names in the tables (if they have the same names like in your screenshot (x1 and x1, ...), JMP will figure out that they are the same, so you don't need to match each factor individually). You should then have the same view as I had. More infos here : Evaluate Design Window (jmp.com)

3- The "Design Diagnostic" informations are values that need to be compared with other designs in order to see the strengths and weaknesses of each design. Each efficiency can go from 0 to 100. Different efficiencies are mentioned:

• D-efficiency is linked to parameter estimation precision, and is very important in screening/factorial designs, to precisely estimate the significance and effect size of effects in the model,
• G-efficiency is linked to the minimization of the prediction variance over the entire experimental space, and is very important in optimization design (focussed more on predictive performance rather than in causal explanations / statistical significance),
• A-efficiency is linked to the minimization and optimization of aliasing between effects ; the highest the A-efficiency, the lower and more precise (and unbiased) will be the estimation of regression coefficients.

More infos here : Design Diagnostics (jmp.com)

4- Very good question, and I don't have a clear answer. This is presumably because of coordinate exchange algorithm (used for custom design) and random starting/generating points for the design. As the design is generated from random points in the design space, the optimal repartition of points in the experimental space may change from one design generation to another. Hence you can see some slight changes in the values when generating again the design. You can manually change these values to the closest value (here 2) without changing the optimality of your design too much, or try generate again the design, eventually by augmenting the number of random starts and/or the design search time (in the red triangle close to "Custom Design" you will find "Number of Starts" and "Design Search Time").

[#4]

Thank you, @Victor_G , for the prompt and clear answer, as well as all the offered materials! It is really useful!!!

[#5]

Hello @Victor_G again! I have a quick question; do you know why I cannot save my model constraints like we can do with the factors and responses?
thank you!
Best,

 

[#6]

Also, @Victor_G do you know how I can specify in the model that my factor values cannot have decimal numbers (ie. 0.5)? Two of my factors are variables that I set up in an instrument. But, the instrument doesn't let you use decimal units because it has to be a whole number. The instrument lets the user change the value of these factors in increments of 1 unit (ie. 21, 22, 23...).

Can I add this in my constraints somehow as well?

Thank you in advance @Victor_G !
Best,

[#7 · Accepted solution]

Hello @ADouyon,

For the factors that can't take any decimal value, you have two options :

1. The first and easy one may be to change the numbers without decimal values in the datatable, by changing the format to "Fixed Dec" and Dec = 0 (see screenshot). But the design may lose a little bit of its optimality, and the optimal values for these factors from the model for the prediction profiler may have decimals.
2. The other way is to treat those factors as "discrete numeric", with specified levels (see my previous comments for more info about Discrete Numeric factor type). This way looks better to me, as this "constraint" is directly taken into account in the model. In your case it might not be interesting to use all the possible levels of your equipment (with an increment of 1) in the levels of the discrete numeric factor, but choosing at least 3 levels (min, medium and max) may be interesting in order to look at possible quadratic effects. In the prediction profiler, it will be treated as categorical factor, but keeping in mind the ordinality of the levels (like 2 < 5 < 7 ... see screenshot for Factor 4), so you can have a look at the trend between each levels.

There might be other options to do it, so don't hesitate to create a new post and check for other users' inputs and responses.

At the end, choose the option you're the most confortable with.
Hope it helps you, 

[#8]

Thank you again @Victor_G for always responding so fast and with such clear answers!! Much appreciated!

(I may try to make an additional post for that one; just out of curiosity :)) 

[#9 · Accepted solution]

Hi @ADouyon,

It seems that you can save and load constraints only if it is done through the option "Specify Linear Constraints".

What you can do in your case, is

1. to generate your DoE entirely with the constraint you added, and then you'll see that a new script in your datatable called "Disallowed Combinations" (see screenshot Disallowed_Combinations-datatable-script.jpg).
2. If you right click on it, then click "Edit", you can copy the JSL code of the disallowed combinations (see Disallowed_Combinations-datatable-script-Edit.jpg) and...
3. Use it (paste it) in the Custom Design creation, in the "Use Disallowed Combinations Script" option (see screenshot "Use-disallowed-combinations-script.jpg).

There might be a smarter and more automatic way to do this (via JSL scripting maybe ?), but this approach work (and you don't need to manually recreate your disallowed combinations).

Hope it will help you !  

[#10]

Thank you very much @Victor_G! I just tried it and was able to do it

[#11]

Hi again @Victor_G,

I have a follow up a couple follow-up questions:

1- If I want to maximize my response to the highest possible value, I still have to define the 'lower limit' for the response, right? But, does this 'lower limit' matter much? Does it matter if I set this to 10 or 50 or 100? (I don't know how to go about this 'lower limit' value ... as I really just care about getting the max response possible).

2- This DOE I am working on is for a biological reaction. Usually, in science/biology, we perform replicate experiments for each condition (run); that means we perform each experimental run in duplicates (or triplicates). I understand, that this is not practical: If my DOE consists of 18 runs total, it does not make sense for me to duplicate all 18 runs, because that would be a total of 36 runs, which would defeat the purpose of me using DOE...
How do I go about deciding the number of replicate runs, then? How to decide if I need just 2 or 4 or 10 replicates?

3- On that same note, in science/biology, researchers usually add some negative and positive control reactions to their experiments as well. That is a reaction we expect to fail because we know in advance it will fail and a reaction we expect to succeed because we have prior knowledge that one works well. How do DOE users go about this in general? 

Thank you very much @Victor_G for your help!!!
Best,