Want to learn about how to design and analyze economical screening experiments? Take this course on Modern Screening Designs at Discovery Summit 2017.
If you have faced a new system, then you have asked that question. The new system might be a new product, a new piece of equipment, a new process, or anything really that is new and not well understood.
Ultimately, you might need to change this system, but first you need to know which variables will affect the desired changes. That is to say, you need an answer to that question, “Which of my factors are important?”
To get that answer, you need a screening experiment. (Notice that the word "screening" may have different meanings than used here.)
The screening experiment I am referring to is designed to address this question in the form of the first order model. Quite literally, a first-order approximation of the new system will suffice to answer the initial question. Potentially important second order effects in the response, such as interaction or curvature, are usually ignored at this stage for the sake of economy. The success of such economic designs depends on how well a given situation lives up to the key screening principles: sparsity of effects, hierarchy of effects, heredity of effects, and projection of the design.
This course follows JMP Software: Custom Design of Experiments and covers the principles of design first and the various methods of design second. The early screening design methods are presented to illustrate these principles and highlight some of the trade-offs that are made. A quick progression toward modern design methods illustrates how economical design may provide more information.
Are you planning to attend the 2017 JMP Discovery Summit in St. Louis? Pre-conference training is an important and popular part of the conference experience. As a Discovery Summit attendee, you’ll get 25% off of this training course. Register soon for the Oct. 16-17 offering to learn about screening with modern designs.