95 / 2025-05-14 22:23:42

An Online Predictive Maintenance Decision-making Framework Considering Imperfect Maintenance via Deep Reinforcement Learning

Imperfect maintenance,Predictive maintenance,Deep reinforcement learning,Remaining useful life

Predictive maintenance based on remaining useful life (RUL) prediction has attracted increasing attention in recent years due to its ability to leverage future equipment states for decision-making. As a common practice in industrial settings, incorporating imperfect maintenance into decision-making can yield greater economic benefits, but it also imposes higher demands on all aspects of predictive maintenance. Existing methods regarding RUL prediction under imperfect maintenance suffer from issues in parameter updating, thereby limiting the real-time capabilities of predictive maintenance decision-making. Additionally, research in predictive maintenance decision-making considering imperfect maintenance is often confined to solving for fixed decision variables, which restricts the flexibility and availability of decisions. To address these limitations, this paper proposes an online predictive maintenance decision-making framework considering imperfect maintenance. Firstly, we propose an AI-based RUL prediction method considering the impact of imperfect maintenance, which can provide accurate prediction results based on online data for subsequent decisions in real time. Secondly, the decision-making problem considering imperfect maintenance is modeled as a deep reinforcement learning problem, and real-time dynamic decision-making output is achieved by the trained agent. The proposed framework is applied to the maintenance of aircraft turbofan engines, demonstrating superior economic benefits compared to other maintenance strategies.