Scholarlink

<p>Traditional condition-based maintenance (CBM) strategies primarily focus on single-component systems, with limited solutions addressing interdependencies in dual-component configurations. This study proposes a maintenance optimization model under finite time horizons, integrating Wiener-Gamma processes to characterize composite degradation patterns and Copula functions to quantify stochastic dependencies between components. A dependent competing risk model is established by incorporating random shocks, accounting for concurrent failure mechanisms from degradation, dependencies, and external shocks. Maintenance decisions are optimized via Markov decision process (MDP), evaluating preventive, corrective, opportunistic, and continuation actions with cost parameters linked to degradation severity and dependency levels. Numerical validation demonstrates the methodology’s effectiveness, with sensitivity analysis revealing a positive correlation between component dependency strength and total system costs. Results indicate that selection of Copula functions significantly influences maintenance decision-making, providing critical insights for designing dependency-aware policies in dual-component systems.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Optimizing maintenance strategies for dependent dual-component systems under finite time constraints

  • 期刊论文

  • 发表日期：

    2026年2月2日

查看全文
  • Rongxing Duan,
  • Tong Zhou,
  • Yihe Lin,
  • Zikang Zou,
  • Qi Zhong

摘要

Traditional condition-based maintenance (CBM) strategies primarily focus on single-component systems, with limited solutions addressing interdependencies in dual-component configurations. This study proposes a maintenance optimization model under finite time horizons, integrating Wiener-Gamma processes to characterize composite degradation patterns and Copula functions to quantify stochastic dependencies between components. A dependent competing risk model is established by incorporating random shocks, accounting for concurrent failure mechanisms from degradation, dependencies, and external shocks. Maintenance decisions are optimized via Markov decision process (MDP), evaluating preventive, corrective, opportunistic, and continuation actions with cost parameters linked to degradation severity and dependency levels. Numerical validation demonstrates the methodology’s effectiveness, with sensitivity analysis revealing a positive correlation between component dependency strength and total system costs. Results indicate that selection of Copula functions significantly influences maintenance decision-making, providing critical insights for designing dependency-aware policies in dual-component systems.