Chapter 10 
Space-Filling Designs
 
Space-filling designs are useful in situations where run-to-run variability is of far less concern than the form of the model. Sensitivity studies of computer simulations is one such situation. For this case, and any mechanistic or deterministic modeling problem, any variability is small enough to be ignored. For systems with no variability, randomization and blocking are irrelevant. Replication is undesirable because repeating the same run yields the same result. In space-filling designs, there are two objectives:
• Prevent replicate points by spreading the design points out to the maximum distance possible between any two points.
• Space the points uniformly.
The following methods are implemented for these types of designs:
• The Sphere-Packing method emphasizes spread of points.
• The Latin Hypercube method is a compromise between spread of points and uniform spacing.
• The Uniform method mimics the uniform probability distribution.
• The Minimum Potential method minimizes energy designs in a hypersphere.
• The Maximum Entropy method measures the amount of information contained in the distribution of a set of data.
• The Gaussian Process IMSE Optimal method creates a design that minimizes the integrated mean squared error of the gaussian process over the experimental region.
Figure 10.1 Space-Filling Design
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset