Optimal design of degradation testing and life testing based on Wiener process models

Degradation testing and life testing are two types of significant reliability tests for evaluating the lifetime of a product. When it is difficult to obtain failure time data via life testing, degradation testing can also be conducted to collect performance degradation data for parameters estimation and reliability assessment. In this paper, an optimal design method is proposed to determine the number of units allocated in degradation testing and life testing under constrains. Firstly, the degradation characteristic is described by using Wiener process model, and the reliability model and parameters estimations are derived both using degradation data and lifetime data. Secondly, based on V-optimality criterion, under the constraint that test cost does not exceed a predetermined budget, the optimal design of test number in each test is obtained by minimizing the asymptotic variance of the maximum likelihood estimator, which is the p-quantile of the lifetime distribution. Finally, a numerical example is presented to illustrate the use of the proposed method in practice, and show its meaningful in improving the estimation accuracy.