Aircraft Maintenance Check Scheduling Using Reinforcement Learning

This paper presents a Reinforcement Learning (RL) approach to optimize the long-term scheduling of maintenance for an aircraft fleet. The problem considers fleet status, capacity, and other constraints schedule hangar checks specified time horizon. are scheduled within interval, goal is to, them as close possible their due date. In doing so, number reduced, availability increases. A Deep Q-learning algorithm used policy. model validated in real scenario using data from 45 aircraft. plan that generated with our compared previous study, which presented Dynamic Programming (DP) based airline estimations same period. results show reduction scheduled, indicates potential RL solving this problem. adaptability also tested by introducing small disturbances initial conditions. After training these simulated scenarios, robustness its ability generate efficient plans only few seconds.