Key points are not available for this paper at this time.
This research develops and rigorously evaluates a comprehensive analytical framework for the optimization and comparative assessment of time-based and condition-based maintenance strategies applied to assets experiencing stochastic degradation. Asset deterioration is modeled using a homogeneous Gamma process, accurately representing monotonic degradation processes subject to inherent variability. The study systematically investigates three maintenance policies : Block Replacement (BR), Periodic Inspection and Replacement (PIR), and Quantile-based Inspection and Replacement (QIR). To quantitatively balance economic efficiency and cost predictability, a novel performance-robustness objective function ϕ is introduced, integrating the expected maintenance cost rate and its variability into a unified metric. Extensive Monte Carlo simulations demonstrate that QIR, through the dynamic adjustment of inspection intervals based on conditional reliability thresholds, consistently yields the lowest expected maintenance costs and superior control of cost variability across a wide spectrum of economic scenarios. In contrast, BR is shown to incur significantly higher expected costs and pronounced variability, while PIR offers moderate improvements by incorporating fixed-interval inspections. Sensitivity analyses further reveal that optimal maintenance parameters must adapt to fluctuations in downtime cost rates, underscoring the necessity of responsive, condition-based strategies. The proposed methodological framework and objective function provide maintenance practitioners with a robust decision-making tool for selecting strategies that minimize total cost of ownership while ensuring financial stability, thereby supporting the adoption of adaptive, data-driven maintenance practices in modern industrial asset management.
Cheikh et al. (Fri,) studied this question.