WP03 will carry out an initial experimental testing of the new indicators to show if they will allow providing a better representation of the airspace users’ operational efficiency than current indicators such as the Horizontal Flight Efficiency.
Tools and procedures will be designed to perform these laboratory-based experimental measurements. The innovative method to assess the proposed metrics will be based on defining a generic advanced trajectory-based airline cost model that approximately captures, to the extent required for air traffic efficiency assessment, the impact of different aspects of the trajectory (e.g. fuel burn or departure and arrival times) on the airlines’ operational costs. The model will be characterized by its simplicity and flexibility, by not requiring sensitive information from the airspace users and by the fact that it will be applicable to both recorded and streaming data. The challenge of defining a relevant reference to compare the actual performance is overcome through trajectory reconstruction and trajectory generation technologies. Recent advances in trajectory reconstruction technologies make it now possible to accurately infer the evolution of the dynamics of a flying aircraft from basic surveillance information. AURORA’s trajectory reconstruction technologies will be based on model-based analysis techniques that merge basic surveillance information with aircraft performance data and weather and atmospheric models.
In this first phase of the project, the results will be obtained by applying the described tools and procedures to off-line scenarios where the complete set of air traffic data will be already available from the start of the analysis. The challenge in this stage will be to implement the means to assimilate and process the massive amount of data involved in the analysis.
A second stage of the project will comprise the extension of the experimental testing to on-line scenarios where the air traffic data is streamed in real-time as the flights evolve. Trajectories will be reconstructed and generated in near-real-time, offering on-line measurements of the evolution of the different KPIs, and providing initial conditions to generate alternative trajectories on-the-fly, and to forecast the future values of KPIs based on the current state. The challenge in this stage will be to implement the means to do the on-line processing of all the different data sources involved in the analysis using stream-based processing technologies. As stream-based processing is contrasted with batch processing, calculations should be performed as soon as new updated within a data stream become available, rather than waiting to perform analysis until a complete dataset is gathered.