题目：Accelerated degradation tests with inspection effects

报告人：陈飘

讲座时间：2022年12月11日（星期日），9：30-10：30

讲座地点：综合楼644会议室

报告人简介：

Dr. Piao Chen is a tenured assistant professor in statistics at Delft Institute of Applied Mathematics, Delft University of Technology. He obtained his PhD in Industrial and Systems Engineering Management from National University of Singapore in 2017, and Bachelor in Industrial Engineering from Shanghai Jiao Tong University in 2013. Dr. Chen’s research focuses on industry big data analytics, reliability engineering, and statistical learning. Most of his work has appeared in top journals in statistics and engineering, including Technometrics, Statistica Sinica, Journal of Quality Technology, IEEE Transactions on Information Theory and IEEE Transactions on Reliability. His work was listed as the annual key achievement by A*STAR and was reported by several leading engineering research news media. Dr. Chen is on the editorial board for International Journal of Data Analysis Techniques and Strategies, and a guest editor for Symmetry. As a principal investigator or team leader, Dr. Chen has led a variety of projects funded by TU Delft, National Research Foundation Singapore, and Ministry of Education Singapore. He served as an organizing committee member for CASE2021, the flagship conference of the IEEE Robotics & Automation Society, and SRSE2022, a leading conference of the IEEE Reliability Society.

摘要：

In this talk, we propose a framework to analyze accelerated degradation testing (ADT) data in the presence of inspection effects. Motivated by a real dataset from the electric industry, we study two types of effects induced by inspections. After each inspection, the system degradation level instantaneously reduces by a random value. Meanwhile, the degrading rate is elevated afterwards. Considering the absence of observations due to practical reasons, we employ the expectation –maximization (EM) algorithm to analytically estimate the unknown parameters in a stepwise Wiener degradation process with covariates. Moreover, to maintain the level of generality for the adaption of the method in various scenarios, a confidence density approach is utilized to hierarchically estimate the parameters in the acceleration link function. The proposed methods can provide efficient parameter estimation under complex link functions with multiple stress factors. Further, confidence intervals are derived based on the large-sample approximation. Simulation studies and a case study from Schneider Electric are used to illustrate the proposed methods. The results show that the proposed model yields a remarkably better fit to the Schneider data in comparison to the conventional Wiener ADT model.