Significance of Design of Experiments (DOE)
Design of Experiments (DOE) is a crucial statistical methodology for testing plans that assess how various factors impact desired outcomes. It involves iterative processes to optimize solutions and make informed decisions based on data analysis results. Learning from historical figures like Edison and Einstein, DOE emphasizes the importance of systematic testing and continual improvement in various fields. Utilizing DOE platforms and methodologies can help in efficient experimentation and achieving desired results effectively.
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
You are allowed to download the files provided on this website for personal or commercial use, subject to the condition that they are used lawfully. All files are the property of their respective owners.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.
E N D
Presentation Transcript
Questions from the Thinker What is Design of Experiments? And why is DOE important? 1d 2d Think about it . . . . . . . R1 R2 3d XN ... x1x2x3x4x5 n-dimensional => . 0d black-hole <= Infinity Consider Consulting Corporation
Discussion: Edison & Einstein DOE Edison: I didn t fail 10,000 times, I learned 10,000 ways that won t work. Einstein: I could have worked out the physics in a week. Edison: But would you prove it s the optimum solution? Einstein: Yes, I see the value of Design of Experiments! Consider Consulting Corporation
Terminology of the DOE Science DOE: statistical modeling of test plans Factors: measurables affecting response Response(s): the desired measurable(s) JMP/DOE: interactive software platforms Evaluation: statistical goodness measure ! Iteration: repeating the designs and tests (on products, or enterprise, or services!) Consider Consulting Corporation
The full suite of DOE platforms Definitive Screening Design {Ver.11} n-factor Response Optimizer ? Consider Consulting Corporation
Envision, Plan, Do, Check, Act PLAN (design) the experiment; DO the experiment by performing the steps; CHECK the results by testing the information; and ACT on the decisions based on those results. Act Check Continual Improvement {settle in } Do W. Edwards Deming: Dr William Edwards Deming Plan was an American statistician, professor, author, lecturer and consultant. He is perhaps best known for the cycle of Envision Envision >> Plan >> Do >> Check >> Act! {[(repeat)]} Consider Consulting Corporation
The JMP Starter window Consider Consulting Corporation
Full-Factorial The basis of the science is to choose ranges for the n factors and assign a top value and low value for each factor. The factors can be varied and starting optimum results calculated. The icon picture represents an n-dimensional cube and if a middle dot is shown, it means a middle value can also be used. Math example: 5! = 5x4x3x2x1 = 120 cases Five factorial has 120 combinations! (If too costly, we can try to do better.) Consider Consulting Corporation
Custom Design The JMP Custom Design allows you to tailor your plan of experiments to meet specific requirements. Custom Design is often the first choice that one choses from the available design options. This is most advisable! Design professionals begin with Custom Design. Consider Consulting Corporation
{RSM} (response surface method) Response Surface Design At some point, you want to do some auxiliary statistical graphics, visualization, and optimization exploration. You can do this with JMP to look for local and global maximum or minimum points on the factor surface. Visually, see maximum peaks (or minimums) on plotted surfaces. Assess if you are at localized peaks or valleys, or are they are the global solution you seek. Max Min Consider Consulting Corporation
Example of DOE Sonic, America s Drive-In We are hiring, and growing! What determines good new-hire candidates? Consider Consulting Corporation
Sonic, Americas Drive-In We are hiring, and growing! What determines good employees? Let s consider the hiring process. Who should be hired based on initial information, and augmented in the future by performance reviews and other measurable factors? First Iteration: Use JMP/DOE Custom Design with a response factor of service times performance: I.e. for the last year of hiring, RATING = (weeks served) times (performance rating). Begin in Oklahoma City, and then in California. Propose the tests to the Sonic CEO. Future Iterations: Use JMP/DOE Augment Design to add additional factors, to be determined, and additional new-hire data as it becomes available expanding to additional states. Consider Consulting Corporation
We are hiring, and growing! What determines good employees? Add ratings later. Try a Custom reduced test set.
Sonic, Americas Drive-In We are hiring, and growing! What determines good employees? The design of 12 test runs is a lot less than the 54 from before. These 12 test cases constitute the new design. These 12 runs will compete very well with the 54 runs ! They will be sufficient to begin iteration after fine tuning your plans and statistical analysis. Try other JMP DOE models.
What we have covered . . . Steps: Think, Design, Test, Analyze, Act! Improvement involves iterations (repeat) Custom Design starts your path . . . Consider specialized design platforms If possible, use full-factorial design As need, try other JMP/DOE platforms. Consider Consulting Corporation and Software Intelligence Corporation Consider Consulting Corporation and Software Intelligence Corporation
Employ the entire design theory Involve key people Determine the factors Start with Custom Design Analyze every step of the way Continual improvement is the goal Ever learn and expand knowledge Consider Consulting Corporation and Software Intelligence Corporation
Thanks for attention! Let s talk in the hallway Enjoy the conference! A poster paper by Charles Edwin Shipp CharlieShipp@aol.com @ShippAhoy (Twitter) LinkedIn/in/CharlieShipp Begin again Begin again Consider Consulting Corporation and Software Intelligence Corporation
