A Wiener-based degradation model with logistic distributed measurement errors and remaining useful life estimation

作者：Shen, Yan^*; Shen, Lijuan; Xu, Wangtu

来源：Quality and Reliability Engineering International, 2018, 34(6): 1289-1303.

DOI：10.1002/qre.2325

• Gibbs sampling
• logistic distribution
• maximum likelihood value
• measurement error
• MCEM algorithm

Degradation data analysis, which investigates degradation processes of products to extrapolate the lifetime properties, is an effective method for reliability analysis. But degradation data that reflect a product's inherent randomness of degradation are often contaminated by measurement errors. To deal with the problem, this paper proposes a Wiener-based model with an assumption of logistic distributed measurement errors and adopts the Monte Carlo expectation-maximization method together with the Gibbs sampling for parameter estimation. Based on the model and parameter estimates, an efficient algorithm is proposed for a quick calculation of maximum likelihood value. Also, the estimation of remaining useful lifetime is discussed. Simulation results show that the proposed model is relatively better and more robust in comparison with the Wiener process with Gaussian noises. Finally, the application of the proposed model is illustrated by an example.

• 出版日期2018-10
• 单位厦门大学