Modeling Traffic Prediction Uncertainty for Traffic Management Decision Support

TFM personnel are known as Traffic Management Coordinators (TMCs) or Traffic Management Specialists (TMSs), depending on the facility in which they work. The general term for these personnel is traffic managers. One of their primary responsibilities is to ensure that traffic at national airspace system (NAS) resources (e.g., airspace sectors, airports) does not exceed levels that can be safely managed by controllers. Traffic managers also endeavor to ensure fair and equitable treatment for all NAS users, i.e., operators of commercial, general aviation, military, and other aircraft. Air Traffic Flow Management (TFM) is the process of balancing demand for airspace and airport resources with the capacity of those resources, in order to achieve both safe and efficient traffic throughput. Demand is typically estimated by predicting flight trajectories, and comparing the predictions to capacity metrics for airports and airspace. The effectiveness of TFM decision-making depends on the accuracy of these predictions. This effectiveness can be improved not only by improving prediction accuracy, but by quantifying the uncertainty in those predictions. When the uncertainty is known, decision analysis and risk management techniques can be applied to improve decision-making performance. To support this goal, a novel method has been developed for measuring and simulating uncertainty in traffic demand predictions. This method employs empirical observations of traffic characteristics to develop statistical models of the error distributions in demand predictions, which in turn can be used for Monte-Carlo simulation of specific traffic scenarios. Preliminary statistical results are presented here, as well as a discussion of simulation applications for both analysis and real-time decision-support tasks.