Bayesian Reliability Assessment of Permanent Magnet Brake Under Small Sample Size
Ancha Xu, Binbing Wang, Di Zhu, Jihong Pang, Xinze Lian
Abstract
Permanent magnet brakes (PMBs) have diverse applications in automotive, robotics, medical devices, and aerospace industries. However, evaluating the reliability of PMBs is challenging due to their complex construction and limited data availability. This article proposes a bivariate Wiener model to capture the degradation patterns of two key performance characteristics of PMBs, and introduces an objective Bayesian method to analyze degradation data with small sample sizes. The derivation of Jeffery prior and reference priors for different ordering groups, along with the verification of posterior propriety, are presented. A rejection sampling embedded Monte Carlo algorithm is employed to obtain Bayesian estimates of model parameters, reliability, and mean-time-to-failure of PMBs. Simulation study results demonstrate the superiority of the objective Bayesian method over the maximum likelihood approach, particularly when dealing with small sample sizes. Finally, the proposed method is applied to assess the reliability of PMBs.