Demand forecasting methods for spare parts logistics for aviation: a real-world implementation of the Bootstrap method

Abstract

One of the critical issues that an airline faces in its day-to-day operations is a correct prognosis of the necessary quantity of spare parts that are continuously fed into unexpected maintenance operations. Indeed, there is a critical need for accurate forecasting methods to predict the demand of these spare parts in order to minimize the so-called Aircraft-On-Ground situations. This paper describes the real-world implementation of the Bootstrap method and the assessment of its performance with actual data from aviation logistics. The analysis reveals that the Bootstrap method, while not the most accurate in every case, should be preferred over other popular methods in spare parts forecasting for aviation, because is more agile and can address adequately all categories of demand. A simple decision support system is then presented to assist airline materials managers in using the bootstrap method. The system is expandable and can potentially incorporate other forecasting method as well.