A novel flexible interval maintenance strategy for multi-unit heterogeneous degrading systems under uncertainty
Coordinated maintenance strategy optimization for multi-unit heterogeneous degrading systems under uncertainty is significantly challenging. Unlike conventional fixed-interval approaches, this study proposes a flexible interval maintenance (FIM) strategy that pursues a unified system-level maintenan...
| Autores: | , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2026 |
| País: | España |
| Institución: | Universitat Politècnica de Catalunya (UPC) |
| Repositorio: | UPCommons. Portal del coneixement obert de la UPC |
| Idioma: | inglés |
| OAI Identifier: | oai:upcommons.upc.edu:2117/459138 |
| Acceso en línea: | https://hdl.handle.net/2117/459138 https://dx.doi.org/10.1016/j.ress.2026.112227 |
| Access Level: | acceso embargado |
| Palabra clave: | Interval-based maintenance Flexible preventive maintenance Uncertainty Two-stage stochastic programming Conditional value-at-risk |
| Sumario: | Coordinated maintenance strategy optimization for multi-unit heterogeneous degrading systems under uncertainty is significantly challenging. Unlike conventional fixed-interval approaches, this study proposes a flexible interval maintenance (FIM) strategy that pursues a unified system-level maintenance policy while allowing flexible execution timing based on remaining useful life information. A two-stage stochastic programming model is developed to support the implementation of the FIM strategy, and the conditional value-at-risk (CVaR) method is incorporated to reflect risk-averse decision preferences. The first stage determines the bounds of the flexible maintenance interval, while the second stage optimizes maintenance timing under various degradation scenarios. A case study on railway freight wagon wheel overhauls demonstrates that the FIM strategy outperforms fixed-interval policies in reducing both cost and the number of threshold overruns. Sensitivity analyses examine the influence of uncertainty and risk preferences on the resulting maintenance strategies. The proposed approach provides a structured and flexible framework for system-level maintenance planning under degradation uncertainty. |
|---|