CMStatistics 2018: Start Registration
View Submission - CMStatistics
Title: Design and analysis of covering arrays using prior information Authors:  Ryan Lekivetz - SAS Institute Inc. (United States) [presenting]
Abstract: Validating complex engineered systems is proving to be an increasingly difficult task for test engineers given the tight budgetary constraints that they usually face. The behavior of such systems is typically due to many inputs, each of which may have several possible settings. Consequently, the input space for these systems is often so large that only a small fraction of the input space can be used for testing. The challenge for the test engineer is to find a way to sample from the input space in such a way that testing effectiveness is maximized. As it turns out, covering arrays can be used to address this challenge. We introduce the concept of covering arrays and illustrate how they may be used to validate engineered systems, and how prior knowledge of the system can be used in both the design and analysis of a covering array.