The figures are reprinted from computer aided design

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ir of compatible parameter sets. The elements of the vector that appear in the sets are assigned their set values and the remaining elements are assigned their nominal values. For example parameters (w, x, y, z ) with nominal values of zero and compatible sets {x = 1, y = 2} and {y = 2, z = 3} yield the vector (0, 1, 2, 3). We also construct a vector for each parameter set that is not compatible with any other set. A set a that is compatible with some b does not need its own vector, since its maximal variation is achieved in the a ∪ b vector. We reduce the number of candidates by culling vectors that are within ǫ of other vectors. We employ Manhattan distance and set ǫ to 1% of the initial tolerance interval. In our examples, culling eliminates 75% of the candidates without missing any failures. We do not construct parameter vectors for larger compatible sets because they are expensive to compute and provide little benefit. The expense is clear: n contacts yield O (n3 ) triples, O (n4 ) quadruples, and s...
View Full Document

This note was uploaded on 08/03/2010 for the course DD 1234 taught by Professor Zczxc during the Spring '10 term at Magnolia Bible.

Ask a homework question - tutors are online