With 220,000 flights and 31.5 million passengers during 2018, Dublin Airport is critically dependent on systems operating at maximum efficiency, for everything from aircraft arrival, departure and turnaround times, to the management of security queue times to optimise the passenger experience. DAA, the company that manages Cork and Dublin airports, is addressing these and other challenges through the use of data science and artificial intelligence (AI).
“DAA is a very diverse business,” says Marc Mullan, vice president of data and analytics with DAA. “It has significant overseas businesses through its ARI and DAA International subsidiaries in addition to Dublin Airport Central and office development on the airport’s campus. In 2014, a strategic investment was made in the establishment of a data analytics function reporting to the chief information officer (CIO). Our role is to create trusted information and actionable insights about our business, passengers and customers to drive business outcomes.”
The 12-strong team works with analysts across numerous departments in DAA to develop analytical solutions. “It’s very hard for one group to cover all the ground,” he notes. “We have adopted a hub and spoke model and created a community of analysts across the business. This approach enables business units to take advantage of the scale of the centralised data platform and data science expertise while having autonomy in terms of creating their own dashboards and reports. We have been building a culture and a community and this allows us to see opportunities where the technology can be used.”
[Read more about Microsoft's research into AI in Ireland ]
One of those uses is in the security screening area. Dublin airport has seen a 45 per cent growth in traffic over the past five years and naturally, this presents challenges at busy times. “We had 14.6 million passengers departing last year. A total of 56,600 of them were on our busiest day in June and 23 per cent of them went through security before 7am,” Mullan says. “There are significant peaks over the course of a day and every day is different. We are now able to forecast in five-minute blocks how many people will arrive at the airport.”
Using analytics to predict the numbers of people and their time of arrival to the airport allows DAA to ensure that there are staff in sufficient numbers at the security areas. The system not only takes into account historical data and current information on flight schedules but also a range of external factors which can impact demand.
“We are able to compare the forecast to the reality and see if it is on profile or not,” he says. “This helps make sure the passenger experience is as good as possible. It’s a very data-driven process, but the key thing is how we use the insights. You still need the expert judgement of human decision making. Data science and AI provides more powerful insights when combined with the business knowledge and judgement of subject matter experts in the field.”
It is exciting to see that, closer to home, AI is being adopted to drive efficiencies at Dublin airport
Out on the airfield, IoT and advanced analytics are being used for the new EU Airport Collaborative Decision Making initiative which aims to reduce delays and improve flight punctuality at European airports. The process involves real-time data sharing by all stakeholders including Dublin Airport, air traffic control, the airlines and ground handling operators.
“All movements on the airfield are tracked using IoT technology,” Mullan explains. “Using this rich geospatial data we are developing models to determine the most important factors impacting on-time performance and to use these insights to aid decision making and to drive action plans for improvements.”
The sheer volume of data involved in these projects and initiatives presents its own issues. “The ability to process this data and provide insights in near real-time is a challenge,” says Mullan. “We are working with Microsoft to take advantage of the power available in the Azure cloud and the flexibility it provides to scale to meet requirements. When we are building models initially, we tend to use a lot of high power processing but when the models are operational the requirement decreases. Flexibility is one of the great benefits of working with Microsoft. We get computing power as and when we need it.”
“Most enterprises today are in the earliest stages of AI implementation. CIOs struggle to identify where and how to realise the business value of AI,” said Ger Perdisatt, enterprise director of Microsoft Ireland. This is why it is exciting to see that, closer to home, AI is being adopted to drive efficiencies at Dublin airport.
“Microsoft’s work with the DAA is using AI and data analytics to help optimise the passenger journey within the airport; be it predicting passenger numbers or improving the queuing process for the ever-increasing numbers of people using Dublin Airport annually.”
[ For more information and to register to attend click Opens in new window ]