The objective of the AIA demonstrator in ICARUS has been to identify – through data analysis – new methods for further optimizing the overall airport operational performance, in an evidence-based, data-driven manner. In order to understand the current airfield performance and baseline capacity in AIA, and to properly perform an analysis that shall lead to improved planning of flight schedules per season, historical data for the runway, aprons, gates, aircraft stands, gates, terminals and local airspace (from all airport stakeholders and certified aviation organization sources) need to be leveraged together with additional aviation-related data sources.
In order to effectively use the ICARUS platform, AIA extracts the necessary data from its Airport Operations Database (AODB) and its IT users upload them in the ICARUS platform, performing all necessary configurations. The AIA data are available in the platform, explored and analyzed by a data scientist in order for the business/operational users that eventually access the ICARUS platform to view the results of the analysis that affect their operations.
To this direction, through ICARUS, AIA has run several descriptive and predictive analytics cases to optimize airport operational areas such as: (a) Airport operations modelling, (b) Aircraft turnaround Delay prediction, (c) Passenger traffic prediction and (d) Aircraft parking stands allocation, that are all interrelated to the core Airport operational performance.
In particular, the demonstrator scenario has allowed AIA to discover hidden patterns and implicit relations between operations and scheduling, and learn from the past experience in order to:
- Improve the planning of flight schedules per season with the airlines operating to AIA, by early identifying the airport’s traffic high and low periods of time and through traffic forecasting at an hourly level.
- Achieve optimum utilization of ground services equipment in collaboration with ground handlers, through more accurate and timely predictions of future delays based on the airline, the route and the type of the aircraft, taking also into account the seasonality (day and time, as well) and the weather.
- Optimize the airport operation services by effectively scheduling the position and slot allocation, identifying the allocation of aircrafts to the existing positions and time slots (if necessary) at a given day.
Going beyond traditional data processing
With the help of the ICARUS platform, AIA is able to further embrace data-related technologies and increase its overall data readiness in terms of data sharing and data analytics, going beyond the existing and traditional data processing tools that are usually available in corporate MIS (Management Information Systems) and are currently in place in AIA. During its demonstration activities, AIA had the opportunity to experiment with ML technologies and various data preparation methods which can be utilized with existing airport data and provide additional insights to further optimize the AIA operations and airport slot and resources scheduling. The results achieved during the demonstrator help identify certain conditions that affect the airport operations performance providing AIA’s flight scheduling and operations personnel with substantiated evidence that will be taken into consideration when scheduling the airport’s operations with the aim to achieve higher utilization and efficiency schedules.
Increasing the data reach
In addition, ICARUS offers aviation-related data from various sources and helps AIA to increase its data reach in an effortless manner, by finding and exploring additional data “on-the-go” while performing or trying to improve an analysis, which became evident with the use of OAG data and open weather data relevant to AIA. Through its experimentation in ICARUS, AIA has the credentials to reach bilateral data sharing agreements between the airport and its relevant parties in a more formalized manner, with less effort and in less time, in order to complement its data availability with different data assets that have been identified and could be leveraged by the airport in scenarios beyond airport capacity planning and optimization.
Creating value from data
Many important lessons learnt came out of the AIA experience in the ICARUS Platform regarding the steps and processes around the data analysis lifecycle, including the data acquisition, curation, mapping, cleaning, uploading, preparing and analyzing phases. The engagement of data scientists and engineers needs to go hand-in-hand with the active involvement of business users and domain experts in order to quickly embrace and maximize value creation from the use of the ICARUS platform from the very beginning.
In conclusion, AIA will significantly benefit from ICARUS in transitioning to the Artificial Intelligence era, by augmenting its baseline data analysis that is already performed in its IT infrastructure with advanced machine learning and deep learning analytics to improve its operational efficiency. The ICARUS project has considerably helped in fostering and establishing a data analysis and sharing culture which can be considered as the legacy of the project for AIA.
Blog post prepared by AIA
