- Contents
- Learn about JMP
- Introduction to DOE
- Starting Out with DOE
- Custom Designs
- Construct Designs That Meet Your Needs
- Overview of Custom Design
- Example of a Custom Design
- Custom Design Window
- Custom Design Options
- Technical Details
- Examples of Custom Designs
- Perform Experiments That Meet Your Needs
- Screening Experiments
- Response Surface Experiments
- Mixture Experiments
- Experiments with Covariates
- Experiments with Randomization Restrictions
- Augment Designs
- Definitive Screening Designs
- The Fit Definitive Screening Platform
- Screening Designs
- Overview of Screening Designs
- Examples of Screening Designs
- Screening Design Window
- Screening Design Options
- Additional Examples of Screening Designs
- The Fit Two Level Screening Platform
- Response Surface Designs
- Full Factorial Designs
- Mixture Designs
- Taguchi Designs
- Evaluate Designs
- Explore Properties of Your Design
- Overview of Evaluate Design
- Example of Evaluate Design
- Evaluate Design Launch Window
- Evaluate Design Window
- Evaluate Design Options
- Compare Designs
- Compare and Evaluate Designs Simultaneously
- Overview of Comparing Designs
- Examples of Comparing Designs
- Compare Designs Launch Window
- Compare Designs Window: Specify Model and Alias Terms
- Compare Designs Window: Design Evaluation
- Compare Designs Options
- Prospective Sample Size and Power
- Discrete Choice Designs
- MaxDiff Design
- Covering Arrays
- Space-Filling Designs
- Overview of Space-Filling Designs
- Space Filling Design Window
- Space Filling Design Options
- Sphere-Packing Designs
- Latin Hypercube Designs
- Uniform Designs
- Comparing Sphere-Packing, Latin Hypercube, and Uniform Methods
- Minimum Potential Designs
- Maximum Entropy Designs
- Gaussian Process IMSE Optimal Designs
- Fast Flexible Filling Designs
- Borehole Model: A Sphere-Packing Example
- Accelerated Life Test Designs
- Nonlinear Designs
- Column Properties
- Technical Details
- References
- Index
Chapter 7
Definitive Screening Designs
Definitive screening designs work for factor screening if you have any combination of continuous or two-level categorical factors. These designs are particularly useful if you suspect active two-factor interactions, or you suspect that a plot of a continuous factor’s effect on the response might exhibit strong curvature.
Definitive screening designs are small designs. For six or more factors, there are only about twice as many runs as factors. Yet, they often conclusively identify which of several factors affect the response. In particular, they can detect and identify any factors causing strong nonlinear effects on the response.
Here are areas where definitive screening designs are superior to standard screening designs:
• They help identify the causes of nonlinear effects by fielding each continuous factor at three levels. In standard screening designs, continuous factors have only two levels. You can add center points to screening designs, but these points establish only if curvature exists. They do not allow you to identify the factors responsible for quadratic effects.
• They avoid confounding between any effects up through the second order. For continuous factors, definitive screening designs have main effects that are orthogonal to each other and orthogonal to two-factor interactions and quadratic effects. Two-factor interactions are not completely confounded with each other. Confounding occurs in many standard screening designs with a similar number of runs.
• They avoid the need for costly additional experimentation to resolve ambiguity from the initial results of standard screening designs.
Figure 7.1 Plot of Response against Factor Values Showing Curvature
-
No Comment
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.