We have an RSS feed! Just add
http://feeds.feedburner.com/amstat/IJyi
to your favorite RSS reader. At least, it works in feedly. If you have trouble with it, let me know.
Several weeks ago, we posted a request for collaboration from Willis Jensen at W.L. Gore & Associates. We've had two responses so far. If you are in industry with a research idea that could use some collaboration, please contact me at smuckerb@miamioh.edu.
Another way we might foster collaboration in this space: Perhaps you are in academia, and you have a research idea that you think is relevant to practitioners. We could post a simple write-up of the idea/work, and invite comment. I'm particularly interested in ideas that could be evaluated by industry practitioners as to viability.
Today, we have a report of some collaborative research by Yili Hong at Virginia Tech. It's the first of two posts highlighting some research he did with DuPont.
----------
Collaborative Research Report: Reliability testing of polymer-composite materials
In 2011, I was fortunate to be selected to be one of the recipients of the DuPont Young Professor Award. The award provides an excellent connection for establishing a collaborative relationship with the DuPont Applied Statistics Group. Since then, we have worked on multiple projects. This post will have two parts, detailing two published projects in reliability that resulted from the collaboration with DuPont. The two projects are test planning for polymer composite materials and accelerated destructive degradation test (ADDT) data analysis.
Polymer-composite materials are key components in many areas such as transportation and alternative-energy industries, because they are more lightweight than metals and alloys but still retain comparable levels of strength and endurance. To understand the long-term performance of such materials, a cyclic fatigue testing procedure is often used. The current industrial standards include test plans with balanced designs and equal spacing of stress levels which, in many cases, are not the most statistically efficient designs. In this project, we developed optimal designs for fatigue testing, which are based on a physical model adapted from the fatigue literature. We showed that such designs are more suitable for modeling cyclic fatigue of polymer composites than the model used in the current industrial standards.
King, C., Hong, Y., DeHart, S. P., DeFeo, P. A., and Pan, R. (2016), Planning Fatigue Tests for Polymer Composites, Journal of Quality Technology, Vol. 28, pp. 227-245.
Yili Hong
Department of Statistics
Virginia Tech
yilihong@vt.edu