NAS Element Closure Planner

We propose a NAS Element Closure Planner, wherein the tool may be used to plan airspace closures and/or combined airport surface and airspace closures in advance, in addition to the exclusive airport surface closures. The proposed technology applies concepts from statistical modeling and machine learning to reliably predict likely future evolution of airport traffic as well as the evolution of other influencing factors such as runway capacity over time. A machine learning tool will drive multiple what-if analysis simulations, each with a slightly modified “initial condition” which may be defined by flight simulation start times (i.e., gate pushback times) as well as allocated taxi routes. Multiple simulations, each driven by one set of initial conditions will be run for each closure time-window option being investigated. Thereby, for each closure time-window option, we will obtain not just one but a distribution of performance metrics, which is a more realistic estimate of likely performance as opposed to a single point value. This ability to reliably predict future performance and the uncertainty associated with it, is a significant step up from the predictive analytics that are available today to airport airside operations staff. The technology would also be applied to determine the multiple futures of closing airspace for any variety of common reasons that would include a commercial space launch use case.