In order to ensure the reliability and availability of weapons and military equipment,regular or irregular tests are needed to monitor their performance,and each test requires multiple measurements.Normally,the impact of measurement is not considered in the storage life prediction of the weapons and military equipment,which may result in inaccurate prediction of the storage life due to using directly the performance degradation data.This paper proposes a method for storage life prediction taking the impact of measurement into consideration.According to whether it is tested or not during storage,the degradation process is divided into the storage condition and the working condition.Based on which,a new degradation model is established.Then,the degradation model parameters reflecting working condition and storage condition are separated by Bayesian theory and the EM algorithm,and the prediction result of storage life is obtained.A degradation trajectory during storage is predicted through equipment performance testing data,and is compared with that of the conventional life prediction method.The result verifies the practicality and effectiveness of the proposed method.