Management-Action-Embedded Sector-Demand Prediction Models

Abstract

This paper describes a class of traffic flow management-action-embedded sector-demand prediction models. The models consist of the open-loop prediction, which is the prediction without management action, and the management-action model. The use of periodic autoregressive modeling approach enables the model to capture both the midterm (30 min to 2 h) trend based on the historical data and the short-term (less than 30 min) transient response based on recent observations. For severe weather days, both storm precipitation and echo tops were used to form a weather index to approximate the management actions due to reduced capacity. In addition to traditional trajectorybased sector-demand prediction methods, which predict only the open-loop behavior of the National Airspace System adequately for short durations of up to 20 min and are vulnerable to uncertainties, this class of models provides a reliable short-to midterm (both open-and closed-loop) sector-demand prediction that accounts for various traffic flow management actions. A combination of closed-loop and open-loop models provide decisionmakers the full range of traffic behavior.