Reliability Inference in A First Hitting Time Model with Augmented Data

摘要：

Reliability models traditionally use lifetime information to evaluate the reliability of a device or system. To analyze small failure-time samples within dynamic environments where failure mechanisms are unknown, there is a need for models that make use of auxiliary reliability information. In this paper we present a model suitable for electronics reliability data, where degradation variables are latent and can be tracked by related observable variables we call markers. We develop parametric and predictive inference equations for a data structure that includes terminal observations of the degradation variable. We compare maximum likelihood estimation and prediction to results obtained by Whitmore et al. 1998. Lastly, we motivate future modeling of variable failure thresholds.

关键词： First hitting time model degradation marker process likelihood

作者: VASILIS SOTIRIS ERIC SLUD MICHAEL PECHT

作者单位: Dept. of Mathematics, Univ. of Maryland, College Park, MD, USA

会议类型: 国际会议

会议名称: The 7th Inyernational Conference on Mathematical Methods in Reliability:Theory,Methods,Applications(第七届国际可靠性数学、理论、方法与应用会议 MMR2011)

会议地点: 北京

会议语种:英文

页码: 845-851

在线出版日期: 2011-06-20（万方平台首次上网日期，不代表论文的发表时间）