Reliability acceptance sampling plan for degraded products subject to Wiener process with unit heterogeneity

Huiling Zheng, Jun Yang*, Houbao Xu, Yu Zhao

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

15 Citations (Scopus)

Abstract

The acceptance sampling plan (ASP), designed with the degradation data following the Wiener process, is widely used to verify the reliability requirements of products. Previous studies mainly designed ASPs under the given sampling plan, which lacked the justification of the sampling plan; furthermore, they ignored the unit heterogeneity of products in the degradation modeling, which affects the description accuracy of risks. To solve the above problems, this paper proposes an optimal ASP design method considering the unit heterogeneity. Firstly, the determinant of the Fisher information matrix is used to characterize the parameters estimation accuracy, under the cost requirement, the optimal test time and sample size are determined by maximizing the determinant, and an optimal sampling plan is obtained. Then, with the likelihood ratio order and the monotonicity of the average failure time on the average degradation rate, the average degradation rate is taken as the acceptance index to effectively simplify the original acceptance test problem. On this basis, the decision-making criterion considering the existence of ASP is obtained by solving the risk constraint equations of both parties. Finally, the simulation and a real example are presented to demonstrate the implementation and feasibility of the proposed method.

Original languageEnglish
Article number108877
JournalReliability Engineering and System Safety
Volume229
DOIs
Publication statusPublished - Jan 2023

Keywords

  • Acceptance sampling plan
  • Decision-making criterion
  • Degradation data
  • Sampling plan
  • Unit heterogeneity
  • Wiener process

Fingerprint

Dive into the research topics of 'Reliability acceptance sampling plan for degraded products subject to Wiener process with unit heterogeneity'. Together they form a unique fingerprint.

Cite this