Abstract

In this paper we propose an objective Bayesian method to analyze the accelerated degradation model based on the new Wiener process stated in Ye, Chen, and Shen (Reliab. Eng. Syst. Safe., 2015, 139: 58–67). The Jeffreys prior and reference priors are derived after reparameterization, and the posterior properties under each prior are validated. A Metropolis random walk algorithm is then employed to generate samples from the posterior distributions, and a simulation study is carried out to reveal the frequentist properties of these priors. Finally, we apply the approach to a real data and obtain the mean-time-to-failure (MTTF) under the usage condition.

摘要

在本文中，我们提出了一种客观的贝叶斯方法来分析基于 Ye、Chen 和 Shen 提出的新维纳过程的加速退化模型（Reliab. Eng. Syst. Safe., 2015, 139: 58–67）。Jeffreys 先验和参考先验是在重新参数化后导出的，并且验证每个先验下的后验属性。然后采用 Metropolis 随机游走算法从后验分布生成样本，并进行模拟研究以揭示这些先验的频率特性。最后，我们将该方法应用于实际数据并获得使用条件下的平均无故障时间 (MTTF)。