On the Cutting Edge

Abstract:The U.S. Department of Defense sought to maximize the benefits of Design of Experiments to improve its testing and analysis of armament systems, cut expenses and reveal potential trouble spots. Collaborative work by chemists, statisticians and engineers at two arsenal sites resulted in significant improvements in a chemical coating process to inhibit corrosion. The goal was to extend the time to failure, but defining the corrosion process was complicated by nine possible process variables and resulting interactions. Systematically designed testing was needed. Using regression model predictions and a definitive screening design, the team fine-tuned factor-level settings and achieved statistically meaningful results with 22 runs rather than testing a potential 8,000 combinations. The targeted data enabled them to develop and validate a model through confirmatory runs, sensitivity analyses and large scale simulations. The experimental process highlighted the value of statistical rigor and …

Access this article
Other ways to access this article

Social Bookmarking

Digg, delicious, NewsVine, Furl, Google, StumbleUpon, BlogMarks, Facebook



Thanks - good application article. At a good level to send around to my customer engineers and managers.
--Ken Johnson, 01-11-2017


Featured advertisers