Reliability Assessment of Bearings Based on Performance Degradation Values under Small Samples

It is difficult to obtain the lifetime data of a long lifetime bearing from a test with limited time. Therefore, to apply the method of reliability assessment based on lifetime data to the high reliability and long lifetime bearings would be impractical. The performance degradation data, which contains reliability information, could be used in the reliability assessment. However, the methods based on performance degradation data are often applied in a large sample situation. In this paper, a method suitable for a small-sample situation based on a distribution-based degradation model and a bootstrapping method combined with the Monte Carlo method (DDBMC) is proposed. This method is put forward to enlarge the sample size and estimate the distribution parameters. Then, the function between distribution parameters and time can be obtained by using the least square method. In this paper, the reliability of the ball bearings under a small sample is assessed to verify the proposed method. Finally, the proposed methodology was applied to assessing the reliability of bearings and shown to be efficient in the reliability assessment of bearings under small samples.