Reliability analysis of products based on proportional hazard model with degradation trend and environmental factor

Huiling Zheng, Xuefeng Kong, Houbao Xu, Jun Yang*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

27 Citations (Scopus)

Abstract

Degradation data and lifetime data have been broadly used for assessing product and system reliability. To utilize the two kinds of data simultaneously, the proportional hazard (PH) model with degradation data as a covariate is proposed for reliability analysis. However, most existing works focus on modeling the PH model with degradation state as a covariate, while the degradation trend is ignored, which makes the PH model unable to carry out reliability prediction directly. Confronted with that, a new PH model with the degradation trend and environmental factor as covariates is developed in this paper. The Wiener process is firstly applied to depict the degradation trend, then the degradation trend and temperature are used as covariates to establish the PH model, and a closed-form of the reliability is derived by the Taylor approximation. Based on the degradation data under actual ambient conditions, the real-time updated reliability prediction is provided to guide the health management of products and systems. The simulation study validates that the proposed model outperforms two existing models in terms of Mean Square Error (MSE). Finally, a real-world example of MOSFET is presented to demonstrate the implementation of the proposed method.

Original languageEnglish
Article number107964
JournalReliability Engineering and System Safety
Volume216
DOIs
Publication statusPublished - Dec 2021

Keywords

  • Covariate
  • Degradation trend
  • Proportional hazard model
  • Reliability analysis
  • Wiener process

Fingerprint

Dive into the research topics of 'Reliability analysis of products based on proportional hazard model with degradation trend and environmental factor'. Together they form a unique fingerprint.

Cite this