@Victor_G : And also separate repetitions from replications to better understand if the problem comes from measurement process (repetitions), or from the system (with possible uncertainty in fixing factors levels, seen through replications)
Video 3: Uses of covariate factors
2024_Q1_Uses of covariate factors.mp4
Video Player is loading.
Current Time 0:00
/
Duration 6:08
Loaded: 0%
0:00
Stream Type LIVE
Remaining Time -6:08
1x
Chapters
descriptions off, selected
captions settings, opens captions settings dialog
captions off, selected
en (Main), selected
This is a modal window.
Beginning of dialog window. Escape will cancel and close the window.
End of dialog window.
This is a modal window. This modal can be closed by pressing the Escape key or activating the close button.
Isn't Define Factor Constraints something I should just leave alone or pretend doesn't exist? A wise man does nothing by constraint. -Marcus Tullius Cicero If you've felt intimidated by the Define...
@Paul_J : Did that all the time when I worked in food and chem industries.
@Bill_Worley : I'm a big proponent of using covariate candidate sets.
@Victor_G : Covariate sets are very useful in chemistry, when dealing with chemical properties as inputs for example
Video 4: best practices on setting ranges
2024_Q1_Setting ranges best practices.mp4
Video Player is loading.
Current Time 0:00
/
Duration 11:30
Loaded: 0%
00:00
Stream Type LIVE
Remaining Time -11:30
1x
Chapters
descriptions off, selected
captions settings, opens captions settings dialog
captions off, selected
en (Main), selected
This is a modal window.
Beginning of dialog window. Escape will cancel and close the window.
End of dialog window.
This is a modal window. This modal can be closed by pressing the Escape key or activating the close button.
@shs: What works for me, if you start from scratch and you are unsure if your borders work, start with the experiment which you think might be the most critical. Here you waste only one experiment.
You only try tests with -1 levels for all factors, centre point(s) and +1 levels for all factors.
And you can very easily augment your scoping design into a factorial design
@Nicola_Nuti: Agreed, increasing effect size gives more power as well.
@Jonas_Rinne : The idea of scoping designs is super helpfull to check if your ranges make sense. Especially when you have many factors.
@SDF1: Even trials that don't work still provide very useful information for the DOE and the whole exploratory process.
@shs : With too narrow settings, you can never be sure about curvature., which is often the case with temperature.
@Paul_J : setting the ranges is good example of why to keep the science in the stats.
@Bill_Worley : If you have some historical data you can build a model and use the Design Space Profiler to help determine optimal ranges.
@Victor_G : Yes, DoE is a method change, but with more predictability about the outcomes, learnings and with planning made a lot easier (compared to trying One Factor At A Time and relying on best guess).
@SDF1 : Depending on how "new" the research is, sometimes a "pre-DOE" is necessary to determine what are the workable ranges for the factors that you're testing so that when you actually do sit down to create a more detailed/complex DOE, you're more likely to have all runs be successful.
Video 5: Modern DOE´s
2024_Q1_Modern DOE´s.mp4
Video Player is loading.
Current Time 0:00
/
Duration 7:51
Loaded: 0%
0:00
Stream Type LIVE
Remaining Time -7:51
1x
Chapters
descriptions off, selected
captions settings, opens captions settings dialog
captions off, selected
en (Main), selected
This is a modal window.
Beginning of dialog window. Escape will cancel and close the window.
End of dialog window.
This is a modal window. This modal can be closed by pressing the Escape key or activating the close button.
@Victor_G : Really like the metaphor with the roof, thanks for sharing!
@shs : I rarely use classical design anymore, as they are large in experimental runs. Using concepts like SVEM with small designs, or definite screening designs, etc.. can give you designs which can cover large amount of factors with few runs, and that convinces people when they first see how much the experimental efforts can drop using modern designs.
@ryan_cooper : I often like to refer to some of Phil Kay's points in this Community discussion. (The title of the question is misleading, however.) Some discussion on why most things can be done with DSD and Custom design.
Basically, you analyze your different curves with the FDE, extract the functional principal components from each curve, and then use these coefficients as covariate in a Custom DoE.
Looking forward to attending the next DoE Club (and JMP Discovery Summit) !
Victor GUILLER
"It is not unusual for a well-designed experiment to analyze itself" (Box, Hunter and Hunter)
