Aircraft Line Maintenance Planning Based on PHM Data and Resources Availability Using Large Neighborhood Search

Abstract

Maintenance planning has become a topic of great interest among researchers and industry practitioners in recent years, since it directly impacts the availability and the lifecycle cost of systems. In the aviation industry, maintenance planning becomes even more relevant due to the high availability expectations from aircraft operators and the high costs incurred when an aircraft becomes out of service. For this reason, some minor maintenance activities are carried out near the gate, between two consecutive flight legs. These activities are referred to as aircraft line maintenance. Planning line maintenance activities is critical because a problem in the execution of line maintenance may lead to flight delays and even flight cancellations. This paper presents a methodology for aircraft line maintenance planning including both the troubleshooting tasks and the repair activities. The proposed methodology uses a Large Neighborhood Search (LNS) algorithm in order to find the most appropriated time and location to perform line maintenance activities. The algorithm considers the precedence relation between a troubleshooting task and its respective repair activity, as well as the dispachability constraints included in the MEL (Minimum Equipment List). Resources availability such as spare parts, equipments and personnel are taken into account, as well as the risk of occurrence of an AOG (Aircraft on Ground) event, estimated from PHM (Prognostics and Health Monitoring) data. An AOG event is an event that leads to a flight cancelation. The optimization goal is to minimize the Expected Cost of Repair (ECR) considering both delay and AOG expenses. A numerical example is presented to illustrate the application of the proposed methodology.