Computational Approaches for Forecasting Operational Risks in the National Airspace System

Models and algorithms for assessing the risks to aircraft operating in the National Airspace System are discussed. These are in support of a ground-based risk assessment system for use by the controllers or risk managers for providing advisories to the aircraft. All the available information about aircraft motion, operational plans and any known adverse environmental factors are used in a traffic simulation of the National Airspace System to forecast emerging risks. The simulation employs aircraft performance models, flight plans, and arrival/departure procedures. It incorporates algorithms for conflict resolution, merging and spacing of aircraft streams and for go-around procedures. Any known variations in the factors affecting the aircraft operations can also be included in the simulation. Potential risks encountered by the aircraft are evaluated with respect to the Aviation Occurrence Category definitions developed by the Commercial Aviation Safety Team of the International Civil Aviation Organization and the National Transportation Safety Board. Risk measures are computed based on time and control margins available to mitigate the predicted hazards. The risk measures computed using the proposed approaches can be combined with available historical data to compute the risk metrics of the National Airspace System. Simulated air traffic in the San Francisco-Oakland-San Jose terminal area is used to illustrate the computational approach. Example results for four different scenarios are given.