Safety Performance Functions to predict Separation Minima Infringements in en-route airspace

Abstract

The objective of this methodology is to characterise the safety of air routes under a systematic framework, in terms of Separation Minimum Infringements (SMI) between en-route aircraft based on models known as Safety Performance Functions (SPF). Bayesian Networks have been selected as techniques with high predictive capability and low probability event estimation. Moreover, they allow the integration of knowledge modelling with data inference. It is complex to develop a Bayesian Network model for SMI prediction. It is necessary to integrate the available knowledge about the precursors of SMIs together with the accumulated knowledge from the analysis of the collected data into a conceptual framework. This framework underpinning the Bayesian Network model focuses on the Closest Point of Approach (CPA) analysis and considers the general scenario of aircraft evolution in an air traffic sector. In order to translate this conceptual framework into a set of causal sub-networks, the ATM barrier and event tree models are used.