Definition and analysis of airline spare parts demand classification

Abstract

Effective demand classification of aviation spare parts is essential for enhancing the scientific basis of inventory planning and improving operational efficiency. Traditional classification methods, such as those adopted by Boeing, typically rely on short-term observation cycles. However, this study finds that such approaches tend to produce unstable classification results, posing challenges to inventory consistency. To address this issue, this paper introduces a new metric called Number of Months with Issues (NMWI), which is derived from historical replacement data and used to quantify demand frequency across different observation cycles. Classification consistency is evaluated under one-year, two-year, and three-year periods. The findings suggest that extending the observation cycle improves classification stability, with the two-year cycle offering a more balanced trade-off between responsiveness and consistency. Furthermore, the original classification thresholds are optimized to reduce fluctuations caused by borderline cases. The optimized method significantly enhances classification stability while maintaining interpretability for practical applications. This study proposes a structured, data-driven approach to demand classification and offers a feasible solution for managing spare parts with intermittent demand patterns.