Evaluation of air traffic flow strategies based on extended range forecasts using simulation and modeling tools

Our group has built a discrete event model, using Extend™, a commercial software application. The model is designed to allow analysis of various Trans-continental (Transcon) flight strategies. From discussions with System Specialists and Managers at the Air Traffic Control System Command Center, we found that the early morning eastbound Transcon flights are extremely dependent on convective weather forecasts. Several strategies for setting routes for Transcons, when convection was forecast, were described to us.

The model uses the forecast as an indicator of expected Center capacity and the model logic selects the shortest route if capacity allows; otherwise alternative routes are selected or the flight is delayed until capacity becomes available. While enroute, the model checks for actual weather in the Center (which may be different than what was forecast). If the weather is worse than forecast, the flight may air-hold, reroute or be diverted. With this model, an analyst can compare various strategies (e.g., which flights have precedence in entering a center – Transcons or local flights? What are the criteria for air holds vs. diverts?, etc.), all as a function of forecast accuracies.

The model tracks actual routing, delays, holding time, diverts and cancellations. Historical data can be entered for the number of flights and their schedule; Center capacity can also be estimated using historical data. The model can be run using fixed values for numeric inputs or a degree of randomness can be used (again based on variation of historical data). The model is assembled from components representing airports, and centers or sectors. These components can be put together in a myriad of ways to examine other NAS issues.

This presentation will describe the application of the model to the evaluation of several Traffic Flow Strategies based on the Collaborative Convective Forecast Product (CCFP).