How Does ch-aviation Estimate and Forecast Aircraft Utilisation?

ch-aviation's aircraft utilisation estimates and forecasts are built using a layered methodology designed to maximise accuracy at the most granular level possible. Here is how it works.

Seasonal Split

Utilisation is estimated and forecasted separately for two seasonal periods:

• Summer: May – October
• Winter: November – April

This seasonal approach reflects real-world operational patterns and improves forecast precision.

Tiered Data Hierarchy

Forecasts are generated using the most specific data available, falling back to broader averages when needed:

• MSN + Operator level (most precise): Where sufficient historical data exists, forecasts are built around the unique combination of an individual aircraft serial number (MSN) and its operator. Averages are calculated per calendar month, based on the number of calendar days and seasonal patterns.
• Aircraft variant + Operator level: If MSN/operator data is insufficient — for example, for aircraft not yet delivered — the model falls back to averages at the aircraft variant and operator level.
• Aircraft variant level (broadest): Where variant/operator data is also insufficient, variant-level averages are used. This level is not surfaced to customers by default, as it is less accurate at the individual MSN level. It is, however, available on request — primarily for OEM customers conducting industry-wide or regional fleet type forecasting who acknowledge and accept this broader level of estimation.

Proxy Logic for New or Rare Variants

For aircraft variants with limited data, ch-aviation applies proxy logic — using closely related variants as a reference. For example, B737-7 utilisation may be estimated based on historical B737-700 data, ensuring forecasts remain as grounded as possible even for newer or less common types.

Data Sources

Current utilisation estimates and forecasts are based on OEM (Original Equipment Manufacturer) data as reported by the carriers themselves. ch-aviation is actively working to incorporate ADS-B data into the estimation and forecasting models, which will further enhance accuracy going forward.

If this article does not answer your question or resolve your issue, you can always submit a ticket and our Customer Support team will get back to you as soon as possible.