doe - 10 suppliers of the active material

[#1]

Hi ,

I'd like to design an experiment to investigate the relationship between active material concentration and the final particle size. I want to test 10 different active material suppliers and see how the active material concentration affects the resulting particle size.

Since the materials are very expensive, I'm looking for an efficient experimental design that can provide meaningful insights while minimizing material usage.

Can you please suggest how I should go about designing this experiment?

thank you

[#2 · Accepted solution]

Hello @YanivD ,

Please find below my thoughts:

    Generically, in an attempt to better understand the relationship amongst the factors, usually I would go with a sequential design of experiment approach, using at the beginning 2-level fractional factorial designs with "bold" (most apart as possible) level settings and tight control of the experimental noise (using different noise strategies, such as randomization, mapping noise factors, holding them constant or even better using them as factors, etc) in order to surface the vital few effects and then using them to progress with the experimentation.

    If you are not close to you objective, to save some runs I would start with foldover designs, Res III, to understand the effects size for each factor and their interactions than once you are satisfied with the variation generated in the response you can foldover the design to increase the resolution and start dealing with the confoundings, screening out the non active factors and interactions, then move on with your inference space towards the optimal point. At the very beginning I would not run replicates, instead I would screen out the factors based on the sparsity of effects, using Daniel plots, to understand about the statistical significance and once you are close to your objective (with much fewer factors to deal with) you can use RSM methods to better model the surface near the optimum point, replicates to improve predictions, etc

    Now, about your case, could you please provide more info, are those 10 suppliers supplying the same product and you need to choose one of them? Do you know whether the process factors and its noise would impact the performance of each supplier? Do you know if there is any quantity that could help you predict if a supplier would be better than the other? (this would help to reduce the amount of suppliers to test)

    Considering each supplier as a block, I would run a fractional factorial design, including the concentration as a design factor, process factors and already mapped noise as factors following the same strategy mentioned above instead of a one factor experiment for each of the suppliers (blocks) and including replicates, however this is up to you given the cost of each run and the information you already have (at the end this strategy may also needs a lot of runs due to the replication)

    Once you run the first fractional factorial design for the first block, you might be able to tell the important effects and they would make sense for the other blocks, you can use this knowledge to refine the experiment for the other blocks.

    The block with the factorial designs including the process factors might not sound like the strategy with the minimum number of runs, however I believe it is going to help you build a solid knowledge foundation about your process factors and their interaction with the supplier material.

    Please let me know if you would like to further discuss the topic,

    Sincerely,

[#3]

Hi Emmanuel,

Thank you for the great points and questions - we will try

All the best,

Yaniv

[#4]

Hi @YanivD,

Adding to the great response from @gonzaef, it would be best to start simple and iterate for your DoE.

I'm not sure testing 10 different suppliers in the first experimental round is the most economical way to learn from the differences in raw materials. Furthermore, you may also have to consider batch-to-batch variability, as it may challenge the outcomes and conclusions you gathered from your previous experimental phase, so this would increase even more the number of samples tested/required (as you're not sure that one sample from one supplier may be representative and how much variability each supplier has). 

Instead, if your goal is to better understand how changes in raw materials may affect your process/responses, I would try to find informations/data that could help characterize the raw material, in documents like technical datasheet or by analyzing them prior to their introduction and testing.

• Once you have several factors/parameters that could help describe the variability of your raw material, you can use these factors as covariates in your DoE (thanks to the platform "Custom Design"), to make sure that only the batches that cover the most area of your experimental space (the more different) are selected. Since you won't be able to change the levels of these factors (only the suppliers can control in a certain range these specifications/factors), the use of these factors as covariates (with predefined possible levels) seems appropriate. You would then be able to link your responses directly to raw material characteristics, which helps having a broader view on the topic and help understanding and generalization. 
• You can also select them manually by running a Principal Component Analysis on the several characterization factors of the raw material, and then select only the most dissimilar (far apart) batches, so that you can have a better overview on how these characteristics may change your response(s).

For the DoE choice, there may be too little information to help you set up your design. As we don't know how many factors, which type(s), the constraints you have, your experimental budget, and how difficult to change some factors may be, it's hard to help you. The Custom design platform should be very helpful to deal with a large variety of constraints and design types.

If you would like to deep dive in the DOE choice and construction, feel free to provide more context and informations. 

Some references for further reading/learning about covariates and a Discovery Summit paper that may help you thinking about your topic (and switching from categorical level "supplier" to continuous covariate factors) :

What is a covariate in design of experiments? (jmp.com)

Developer Tutorial - Handling Covariates Effectively when Designing Experiments - JMP User Community

Coding with Continuous and Mixture Variables to Explore More of the Input Space ... - JMP User Commu...

I hope this complementary response will help you,

[#5]

Dear Victor,

Thank you so much for your ideas and solutions - as always including the links are valuable and can help.

all the best,

Yaniv