An optimal designed degradation experiment for reliability improvement

This paper’s main purpose is to describe how an optimal designed degradation experiment can be figured out to improve product’s reliability. Insetad of using a known termination time of an experiment, the criterion of minimizing the total cost of experiment is adopted. A key quality characteristic of a lighting product is its luminosity or luminous flux as measured in lumens, say Y (t), which is measured at time t. Let the observed relative degradation D(t) = d(Y (t), Y (t0)), where Y (t0) is the initial luminosity measured at time t0. A useful and simple model for observed degradation is: