Using ADS-B Data to Develop New Metrics for Flight Efficiency

Each day, ADS-B signals from aircraft enable real-time tracking of hundreds of thousands of flights on the Flightradar24 network. Thanks to the reliability and volume of that data, it is now playing a valuable role in detailed analysis of air traffic management. Flightradar24 provided extensive flight data, enabling researchers at Centro de Referencia I+D+i ATM (CRIDA), Boeing Research & Technology Europe (BR&TE) and Centre for Applied Data Analytics (CeADAR) to conduct an ADS-B based air traffic performance assessment to measure air navigation service provider and airline flight efficiency in Europe.

Objectives

Air navigation service provider (ANSP) and airline perspectives on efficiency often differ, where airlines are concerned mainly with schedule adherence and fuel consumption, while ANSPs focus on more in-depth components like sector capacity, air traffic controller interventions, emissions, and noise. ANSPs have their own reporting requirements and Key Performance Indicators (KPIs) to evaluate their performance and management of the air traffic. Being able to evaluate how ANSPs are measured and the impact on airlines’ performance through ground delays, reconfiguration of the airspace or controller interventions is essential to improve airlines’ strategies and maximize aircraft capabilities.

This research, carried out as part of the AURORA (Advanced User-centric efficiency metRics for air traffic perfORmance Analytics) project has two main areas. First, AURORA aims to propose new metrics to assess the operational efficiency of the ATM system and to measure how fairly the inefficiencies in the system are distributed among the different airspace users.

Currently, a flight path deemed the most efficient by the ANSP, in terms of sector capacity and controller intervention, could be quite inefficient from an airline’s perspective due to additional length or fuel consumption. These new metrics will be developed to encapsulate the airspace users’ operational objectives, considering fuel consumption, schedule adherence and cost efficiency of the flights.

AURORA will also explore and test techniques borrowed from big data, data science, and information management fields for the collection and analysis of massive amounts of data (such as full ADS-B coverage of European airspace). These techniques will allow AURORA to propose a new framework for ATM decision-making based on real-time performance monitoring.

Methods

This specific research presents a methodology based on the reconstruction of trajectories from ADS-B surveillance data and the generation of user preferred trajectories. Comparing both sets of trajectories it is possible to obtain different efficiency KPIs, including future ones that could better accommodate the airline’s view on efficiency, such as fuel consumption, cost or equity and environmental impact such as emissions or noise, with a higher degree of realism.

Given a set of recorded ADS-B surveillance tracks in an airspace, the trajectory reconstruction algorithm is able to identify the different trajectories flown in that airspace, infer the evolution of the aircraft dynamics, and generates the evolution of the complete set of the aircraft state variables, such as position, mass, speeds or thrust setting. According to the particular KPI that is being evaluated, different reference trajectories based on the initial flight plan or a particular cost function can be generated to compare to the flown one.

Results

Example flights between London (LHR) and Amsterdam (AMS)

City Pair AMS-LHR

ID ROUTE KEP¹ KEA¹ FEP_DW² FEA_DW²
1 AMS-LHR 1.04 13.59 2.14 10.78
2 AMS-LHR 1.04 23.56 6.47 25.27
3 LHR-AMS 11.89 13.10 11.27 6.49
4 LHR-AMS 11.89 20.13 10.99 8.00

The results summarized in the table above show the value of some performance indicators defined in AURORA applied to 4 flights covering the city-pair Amsterdam (AMS)—London (LHR). The ideal values for the KEP and KEA indicators is 0, meaning no difference was performed in terms of distance. For FEP_DW and FEA_DW negative values can be achieved if the flight consumed less fuel than the geodesic routing. KEP and KEA indicators are also computed by EUROCONTROL in the PRU reports with a slightly different approach; KEP and KEA are calculated outside the 40NM around origin and destination to focus on capturing en-route inefficiencies, while in AURORA they are calculated for the whole flight and include weather considerations. This follows the recommendations from the AURORA Airspace User Advisory group formed by some of the major airlines flying in Europe, indicating that their efficiency will be better assessed if the efficiency indicators are calculated covering the whole trajectory.

The values in the performance indicators reveal flights subject to inefficiencies. When analyzing in detail the cause of the inefficiencies, it can be observed that different runway configurations as the one considered in the flight plan had an impact not only in the TMA part but also in the en-route, as the flight will need to adapt to this new situation. It is remarkable the correlation existent between KEP and FEP_DW, and KEA and FEA_DW. At a first glance, it seems that indicators using fuel consumption follow the tendency of those indicators calculated using distance, however the value of the correlation varies, and here is where new fuel consumption indicators might provide a better understanding on the penalty to the airlines than to the inefficiencies of the ATM system. The new indicators proposed in AURORA will show these differences with respect to the already-computed indicators from the PRU.

Future results set will include differentiation between route phases (TMAs and en-route), new indicators such as vertical efficiency indicators or equity indicators and efficiency indicators calculated for large datasets.

The next figures summarize the different profiles of the 4 selected flights. The circles in the Route profile images show the 40NM cut made by the PRU when computing KEP and KEA.

Flight 1

Flight 2

Flight 3

Flight 4

Finding In-flight Efficiencies

By developing new metrics and KPIs, a more holistic vision of airspace efficiency takes shape, striking a better balance between the needs of airspace users and the Air Navigation Service Providers. The results of the AURORA project research should be of great value to the overall SESAR 2020 program, and especially useful for the performance review commission. Finding ways to increase flight path efficiency with airspace users in mind will not only benefit ANSPs and airline business models, but passengers as well.

 

1. KEP* & KEA* – Comparison between the length of a trajectory and the shortest distance between its endpoints. The trajectory used as reference could be the last filed flight plan (KEP – Key performance Environment indicator based on last filed flight Plan – indicator) or the surveillance data (KEA – Key performance Environment indicator based on Actual trajectory – indicator).

2. FEP_DW* & FEA_DW* – Comparison between fuel consumption of a trajectory and fuel consumption of the shortest distance between its endpoints, considering weather impact in both cases. The trajectory used as reference could be the last filed flight plan (FEP_DW indicator) or the surveillance data (FEA_DW indicator).