12-02, 10:55–11:25 (Europe/Amsterdam), Ernst-Curie
Inefficiencies in the flight preparation processes (turnaround) are accountable for around 30% of the total delays at Royal Schiphol Group (the Amsterdam airport). This process has been a black box and for this reason, it was quite hard to improve. To open the turnaround black box, Schiphol has developed technology based on computer vision using deep learning that detects many different turnaround-related tasks from images that are streamed from cameras located in the aircraft ramps in real time. In this session, we will explain how this project started, the technologies that we have applied, and the business impact that is generated at enabling the airport to reduce delays.
The Amsterdam airport Schiphol is one of the busiest airports in Europe. Schiphol hosts more than 120 airlines and it is connected to 316 destinations around the globe. In this setup, for Schiphol it is key to keeping a high on-time performance (proportion of flights with no delays), not only to guarantee good service as an airport but also to contribute to the good service of the whole network of airports that Schiphol is connected with.
Schiphol has identified that 30% of the total delays are caused by inefficiencies related to the flight preparation process (turnaround) such as fuelling, passenger boarding, baggage handling, etc. To improve the turnaround process, many different tasks should be measured and monitored to find inefficiencies that can be improved.
In the past, to have reliable information about all relevant turnaround-related tasks, many sensors in vehicles and airplanes needed to be installed and maintained. However, given the number of vehicles involved and many airlines, gathering information and maintaining thousands of sensors was a monumental task making this approach financially unviable.
Today, to open the turnaround black box, Royal Schiphol Group has developed a technology that uses deep learning and computer vision to make real-time detections of more than 33 different turnaround-related tasks from images streamed from cameras located at the aircraft ramps (parking spots). From these detections, Schiphol has developed actionable insights to avoid delays by using real-time alert systems and has gained a deep understanding of the inefficiencies of the turnaround process by analyzing historical data.
In this talk, we will introduce the technologies that we have applied (such as data centric AI), learnings, challenges, and the business impact.
No previous knowledge expected
Santiago has an educational background in industrial engineering, operations research, and data science. In the last years, he has been working extensively with computer vision to make real-time turnaround tasks related to detections with the final aim of reducing flight delays. Santiago's passion is to develop data-driven solutions to help to improve people’s life.
Bringing ingenuity to life using Applied AI