EXPERIMENTAL-DESIGN FOR A CLASS OF ACCELERATED DEGRADATION TESTS

Traditionally, reliability assessment of new devices has been based on accelerated life tests. This approach is not practical for highly reliable devices, such as lasers, which are not likely to fail in experiments of reasonable length. An alternative approach is to monitor the devices for a period of time and assess their reliability from the changes in performance (degradation) observed during the experiment. In this article, we propose a methodology for designing experiments for degradation processes in which the amount of degradation over time levels off toward a plateau (maximum degradation) that is a function of stress. We provide (a) the stress levels for the experiment, (b) the proportion of devices to test at each stress level, (c) the times at which to measure the devices, and (d) the total number of devices to test. We apply the proposed methodology to an actual example.