ABSTRACT
The aim of this paper is to formulate and solve the aircraft maintenance scheduling design optimization problem while considering the effects of a pandemic, such as the COVID-19 pandemic. Aircraft maintenance is dependent on how much the aircraft is in service, among other factors. It is no surprise that the pandemic has significantly impacted airline operations, due to travel restrictions and people’s hesitancy to travel. Thus, it is important for airliners to consider the progression of the pandemic when designing their flight and maintenance schedules. The approach proposed in this paper addresses this issue by integrating several models. The first one is a time series forecasting model to predict future COVID-19 cases – in this paper we use a Long Short-Term Memory (LSTM) network. The second model is a simple neural network for predicting flight frequencies based on historical flight data and the results of the first model. The predicted flight frequencies are used to generate a flight schedule, which serves as input to the third model – the maintenance schedule design optimization model, which is formulated as a mixed binary-integer non-linear optimization problem. The final output from the integrated model is the optimized maintenance schedule and associated costs. To demonstrate the proposed approach, we present an illustrative example with 3 aircraft and perform a sensitivity analysis to gain further insight into the results. Copyright © 2022 by ASME.