Demand change detection in airline revenue management

Demand shocks—unobservable, sudden changes in customer behavior—are a common source of forecast error airline revenue management systems. The COVID-19 pandemic has been one example highly impactful macro-level shock that significantly affected demand patterns and required manual intervention from analysts. Smaller, micro-level shocks also frequently occur due to special events or competition. Despite their importance, detection methods employed by airlines today are often quite rudimentary practice. In this paper, we develop science-based framework based on statistical hypothesis testing which enables fast shocks. Under simplifying assumptions, show how the properties detector can be expressed analytical closed form demonstrate expression is remarkably accurate even more complex environments. Simulations used successfully identify positive negative both volume willingness-to-pay. Finally, discuss could integrated into an system allow for practical use