A strategy for extrapolation in accelerated testing

Some problems in engineering and applied science require building a model of a newly observed phenomenon and extrapolating that model beyond the regime in which data has been gathered. A simple example is accelerated testing, in which devices are stressed at extreme conditions for a short time — for example, months — in order to build a model of how they behave over decades under moderate stress. In such a situation, choosing the wrong model can cause costly errors in estimating reliability. By integrating physical and statistical thinking, we have developed a strategy for identifying certain kinds of modeling errors. In this paper, we give one example in which we have experimental indications of such an error that would not have been identified without our new approach. We also list some examples that gave no such indications.