Constrained optimal maintenance strategies for k-out-of-n systems with dependent components and mission duration

Many existing maintenance strategies for k-out-of-n systems lack practical applicability, often neglecting essential operational constraints in real-world scenarios. Performing maintenance during active operations is generally impractical, emphasizing the need to account for mission duration in main...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Reliability engineering & system safety Ročník 264; s. 111265
Hlavní autoři: Kheyri, Azam, Taghipour, Sharareh
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier Ltd 01.12.2025
Témata:
ISSN:0951-8320
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Many existing maintenance strategies for k-out-of-n systems lack practical applicability, often neglecting essential operational constraints in real-world scenarios. Performing maintenance during active operations is generally impractical, emphasizing the need to account for mission duration in maintenance planning. Additionally, the assumption of independent component failures is unrealistic, particularly in challenging environments where failures are often correlated. In critical missions, strategies must go beyond cost-efficiency to maintain system availability at acceptable levels throughout the mission. To address these issues we propose three maintenance policies simple replacement, replacement first, and replacement last to various dependency structures using copula models, including Clayton, Gumbel, and FGM. Our framework integrates mission duration into decision making, optimizing both the number of components and replacement timing to achieve a balance between cost and system availability. By explicitly accounting for component dependencies and the availability constraint, this study provides a comprehensive and realistic strategy for optimizing maintenance in k-out-of-n systems under operational conditions.
ISSN:0951-8320
DOI:10.1016/j.ress.2025.111265