Abstract: Affected by measurement dispersion, instrument error and other factors, there is a measurement uncertainty problem in the ac-celerated degradation test. In this paper, the propagation and evaluation method of the measurement uncertainty of the accelerat-ed degradation test is proposed to improve the accuracy and credibility of the test evaluation results. Firstly, the paper introduces the sources of measurement uncertainty in accelerated degradation tests and analyzes the propagation effect of measurement uncertainty in the parameter estimation process of the Wiener process accelerated degradation model. Secondly, the Guide to the Expression of Uncertainty in Measurement (GUM) method and Monte Carlo method (MCM) are used to evaluate the influence of performance observation data measurement uncertainty on product reliability evaluation results at normal stress levels respec-tively. Finally, utilizing an actual type of polymer sensors for space’ accelerating degradation testing, results of different methods are compared and the fitness of the proposed methods is verified.

Key words: accelerate degradation testing, measurement uncertainty, GUM, Monte Carlo method, polymer sensor