Statistical characterization of airplane delays.

Area of Science:

• Aviation analytics
• Statistical modeling
• Transportation science

Background:

• Aviation is crucial for global connectivity, with customer satisfaction heavily influenced by flight delays.
• Quantifying flight delays across numerous flights, airports, and airlines presents a significant analytical challenge.

Purpose of the Study:

• To develop a statistical procedure for comparing mean delays and extreme delay events.
• To analyze flight arrival delay data from UK airports between 2018 and 2020.
• To investigate the impact of the COVID-19 pandemic on flight delay statistics.

Main Methods:

• Statistical analysis of flight arrival delay data.
• Development of a comparative procedure for mean and extreme delays.
• Identification of delay distribution patterns, including power-law decay.

Main Results:

• A method to compare mean delays and extreme events across airlines and airports was established.
• Large flight delays exhibit a power-law decay distribution.
• Significant shifts in delay statistics were observed during the COVID-19 pandemic.
• Flight delays can be modeled as a superposition of simple distributions, forming a superstatistics.

Conclusions:

• The study provides a robust statistical framework for analyzing and comparing flight delays.
• Understanding delay patterns, especially extreme events, is vital for airline and airport operations.
• The COVID-19 pandemic markedly altered flight delay characteristics, highlighting the dynamic nature of air travel.