Bill Meeker
Distinguished Professor
Department of Statistics
Iowa State University
Ames, Iowa

Service Life Prediction of Field-Exposed Units Based on Laboratory Accelerated Degradation Test Data

Photodegradation caused by ultraviolet (UV) radiation is the primary cause of failure for organic paints and coatings, as well as many other products made from organic materials and exposed to sunlight. Accelerated testing provides timely information about product lifetime. This paper describes methods to model accelerated test data for a specific coating, in which spectral UV wavelength and intensity, temperature, and relative humidity were controlled over time. We use a nonlinear mixed-effects model to describe the relationship between the degradation path and the experimental variables with random effects used to account for unit-to-unit variability. The parameters in the model are estimated by using the method of maximum likelihood. We show how to extend the model to allow for time-varying covariates and apply the method to predict outdoor degradation where the explanatory variables are uncontrolled but can be predicted.

Key Words: Coatings, Photodegradation, Nonlinear model, Random effects, Reliability,

UV exposure.

Joint work with Yuanyuan Duan and Yili Hong at Virginia Tech and Deborah L. Stanley, and Xiaohong Gu at NIST