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, maintenance capacity, and other maintenance constraints to schedule hangar checks for a specified time horizon. The checks are sch...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-04-01
|
Series: | Aerospace |
Subjects: | |
Online Access: | https://www.mdpi.com/2226-4310/8/4/113 |
id |
doaj-9a2fd8521df24e9e97b94adb5cd4e5e8 |
---|---|
record_format |
Article |
spelling |
doaj-9a2fd8521df24e9e97b94adb5cd4e5e82021-04-17T23:00:06ZengMDPI AGAerospace2226-43102021-04-01811311310.3390/aerospace8040113Aircraft Maintenance Check Scheduling Using Reinforcement LearningPedro Andrade0Catarina Silva1Bernardete Ribeiro2Bruno F. Santos3Department of Informatics Engineering, University of Coimbra, CISUC, 3030-290 Coimbra, PortugalDepartment of Informatics Engineering, University of Coimbra, CISUC, 3030-290 Coimbra, PortugalDepartment of Informatics Engineering, University of Coimbra, CISUC, 3030-290 Coimbra, PortugalAir Transport and Operations, Faculty of Aerospace Engineering, Delft University of Technology, 2629 HS Delft, The NetherlandsThis paper presents a Reinforcement Learning (RL) approach to optimize the long-term scheduling of maintenance for an aircraft fleet. The problem considers fleet status, maintenance capacity, and other maintenance constraints to schedule hangar checks for a specified time horizon. The checks are scheduled within an interval, and the goal is to, schedule them as close as possible to their due date. In doing so, the number of checks is reduced, and the fleet availability increases. A Deep Q-learning algorithm is used to optimize the scheduling policy. The model is validated in a real scenario using maintenance data from 45 aircraft. The maintenance plan that is generated with our approach is compared with a previous study, which presented a Dynamic Programming (DP) based approach and airline estimations for the same period. The results show a reduction in the number of checks scheduled, which indicates the potential of RL in solving this problem. The adaptability of RL is also tested by introducing small disturbances in the initial conditions. After training the model with these simulated scenarios, the results show the robustness of the RL approach and its ability to generate efficient maintenance plans in only a few seconds.https://www.mdpi.com/2226-4310/8/4/113aircraft maintenancemaintenance check schedulingreinforcement learningq-learning |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Pedro Andrade Catarina Silva Bernardete Ribeiro Bruno F. Santos |
spellingShingle |
Pedro Andrade Catarina Silva Bernardete Ribeiro Bruno F. Santos Aircraft Maintenance Check Scheduling Using Reinforcement Learning Aerospace aircraft maintenance maintenance check scheduling reinforcement learning q-learning |
author_facet |
Pedro Andrade Catarina Silva Bernardete Ribeiro Bruno F. Santos |
author_sort |
Pedro Andrade |
title |
Aircraft Maintenance Check Scheduling Using Reinforcement Learning |
title_short |
Aircraft Maintenance Check Scheduling Using Reinforcement Learning |
title_full |
Aircraft Maintenance Check Scheduling Using Reinforcement Learning |
title_fullStr |
Aircraft Maintenance Check Scheduling Using Reinforcement Learning |
title_full_unstemmed |
Aircraft Maintenance Check Scheduling Using Reinforcement Learning |
title_sort |
aircraft maintenance check scheduling using reinforcement learning |
publisher |
MDPI AG |
series |
Aerospace |
issn |
2226-4310 |
publishDate |
2021-04-01 |
description |
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, maintenance capacity, and other maintenance constraints to schedule hangar checks for a specified time horizon. The checks are scheduled within an interval, and the goal is to, schedule them as close as possible to their due date. In doing so, the number of checks is reduced, and the fleet availability increases. A Deep Q-learning algorithm is used to optimize the scheduling policy. The model is validated in a real scenario using maintenance data from 45 aircraft. The maintenance plan that is generated with our approach is compared with a previous study, which presented a Dynamic Programming (DP) based approach and airline estimations for the same period. The results show a reduction in the number of checks scheduled, which indicates the potential of RL in solving this problem. The adaptability of RL is also tested by introducing small disturbances in the initial conditions. After training the model with these simulated scenarios, the results show the robustness of the RL approach and its ability to generate efficient maintenance plans in only a few seconds. |
topic |
aircraft maintenance maintenance check scheduling reinforcement learning q-learning |
url |
https://www.mdpi.com/2226-4310/8/4/113 |
work_keys_str_mv |
AT pedroandrade aircraftmaintenancecheckschedulingusingreinforcementlearning AT catarinasilva aircraftmaintenancecheckschedulingusingreinforcementlearning AT bernardeteribeiro aircraftmaintenancecheckschedulingusingreinforcementlearning AT brunofsantos aircraftmaintenancecheckschedulingusingreinforcementlearning |
_version_ |
1721523767944937472 |