A Large-Scale Annotated Multivariate Time Series Aviation Maintenance Dataset from the NGAFID

TL;DR

This paper introduces the NGAFID Aviation Maintenance Dataset, the largest real-world multivariate flight and maintenance log dataset, enabling improved predictive maintenance research for aircraft using real operational data.

Contribution

The paper provides a large-scale, real-world aviation dataset with detailed annotations, and offers tools for easy access and benchmarking, advancing predictive maintenance research.

Findings

Contains 31,177 hours of flight data across 28,935 flights.

Includes 2,111 maintenance events with 36 types of issues.

Provides benchmarks with three models using the dataset.

Abstract

This paper presents the largest publicly available, non-simulated, fleet-wide aircraft flight recording and maintenance log data for use in predicting part failure and maintenance need. We present 31,177 hours of flight data across 28,935 flights, which occur relative to 2,111 unplanned maintenance events clustered into 36 types of maintenance issues. Flights are annotated as before or after maintenance, with some flights occurring on the day of maintenance. Collecting data to evaluate predictive maintenance systems is challenging because it is difficult, dangerous, and unethical to generate data from compromised aircraft. To overcome this, we use the National General Aviation Flight Information Database (NGAFID), which contains flights recorded during regular operation of aircraft, and maintenance logs to construct a part failure dataset. We use a novel framing of Remaining Useful Life (RUL) prediction and consider the probability that the RUL of a part is greater than 2 days. Unlike previous datasets generated with simulations or in laboratory settings, the NGAFID Aviation Maintenance Dataset contains real flight records and maintenance logs from different seasons, weather conditions, pilots, and flight patterns. Additionally, we provide Python code to easily download the dataset and a Colab environment to reproduce our benchmarks on three different models. Our dataset presents a difficult challenge for machine learning researchers and a valuable opportunity to test and develop prognostic health management methods