Trajectory segments with detailed predictability performance metrics and clusterisation for two days of traffic over South West Functional Airspace Block

Published: 23 April 2018| Version 2 | DOI: 10.17632/s3m26t7f6v.2
Rocío Barragán Montes


This dataset includes two days of flight trajectory information over South West (SW) Functional Airspace Block (FAB). The flight trajectories have been decomposed into segments with individual information regarding three-dimensional deviations and a global predictability non-conformance metric. Using MakeDensityBasedClusterer algorithm and SimpleKMeans as clusterer in Weka 3.9.1, the 8 clusters assignment is included in this dataset. The attributes acting as explanatory variables are: * Segment type * 2D deviation * Delta flight level threshold normalised at the destination point * Delay range at destination point * Point of origin (A) * Point of destination (B) * Diversion point (C) * Is origin point an airport * Is destination point an airport * P_norm (Predictability_normalised) The attributes studied as response variables are also detailed in this dataset.



Universidad Politecnica de Madrid Escuela Universitaria de Ingenieria Tecnica Aeronautica


Machine Learning Algorithm, Unsupervised Learning, Air Traffic Control, Flight Tracking, Trajectory Planning