Last time, we announced an RSS feed. Now, thanks to Lara Harmon and the folks at the ASA, we have an RSS button on the ISVC landing page.
As always, we are looking for material for this space. If it has to do with academic/industry collaboration, we're interested.
Today, we have the second of two posts regarding some collaborative research between Yili Hong and DuPont.
----------
Collaborative Research Report: Accelerated destructive degradation test data analysis
Accelerated destructive degradation testing (ADDT) is a widely-used technique for long-term material property evaluation. One area of application is in determining the thermal index (TI) of polymeric materials. There are two approaches to estimating a TI based on data collected from ADDT: the traditional procedure and maximum likelihood (ML) procedures. The traditional procedure is a two-step approach which involves polynomial interpolation and least-squares estimation. The traditional procedure is specified in current industrial standards and is widely used in industrial applications. The ML procedure is frequently used in the statistical literature but rarely seen in industrial ADDT applications. In the project, we compared both approaches with two motivating datasets from the literature, as well as simulation studies. We were able to show that the ML procedure has many advantages over the traditional procedure in terms of estimation performance, uncertainty quantification, material comparisons, and predictions.
Overall, I found these collaborations to be rewarding and I really appreciated my industrial partners. These collaborations not only provided me the opportunities to see the needs of industry, which is important for an industrial statistician, but also provided excellent training opportunities for graduate students.
King, C., Hong, Y., Xie, Y., Van Mullekom, J. H., DeHart, S. P., and DeFeo, P. A. (2016), A Comparison of Traditional and Maximum Likelihood Approaches to Estimating Thermal Index for Polymeric Materials, Journal of Quality Technology, in press.
Yili Hong
Department of Statistics
Virginia Tech
yilihong@vt.edu