The challenge: Planning for the airports of the future despite a shortage of skilled workers
It’s the year 2050. In the parking area reserved for e-shuttles at Frankfurt Airport, a family gets out of their driverless vehicle. Two gripper arms that appeared automatically from the floor have already lifted their suitcases out of the shuttle’s trunk. While the parents pull their suitcases into the terminal building, the children marvel at all the flying taxis landing and taking off again at designated ports. The autonomous shuttle that brought them here is long gone. The prospect of a summer vacation on Crete has put a smile on everyone’s faces. Just 15 minutes later, the family is on its way to the departure gate. A fully automatic conveyor belt has already carried the prechecked suitcases down to the tarmac, where they are now being taken to the aircraft on a driverless vehicle. Biometric scanners, modern screening devices, and AI have made security checks almost fully automatic—no lines, no waiting. When the family reaches the gate, they still have 20 minutes until boarding. Their holiday fever reaches its peak.
Fraport AG is looking to realize a very similar vision by 2050. “The airport of the future should be sustainable and should take all the complications out of traveling—from baggage handling to connecting with the mobility services of the future to digital, automated processes on-site,” explains Verena Dollberg, Program Director Strategy and Digitalization at Fraport. “But to turn this vision into a reality, we have to tackle certain challenges—first and foremost, a shortage of skilled workers, both on our runways and in our offices.” Over the next few years, Fraport is expecting to grow by around 30 percent. At the same time, the airport operator—like many other companies—is expecting to lose around 30 percent of its workforce through aging. AI will be a key component in responding to this impending shortage of skilled workers by boosting the efficiency of processes and supporting employees in their work. “Everyone has been talking about AI for some time now. The trend is really picking up pace, and lots of new solutions are emerging that create efficiencies,” Dollberg says. “But we won’t get to the airport of the future just by introducing AI solutions left and right. It’s a change process that will only work if we get our employees on board and take active steps to open up the discussion on AI solutions. That’s how we can break down prejudices and dispel fears.” Fraport has decided to use Azure OpenAI Service to ensure company-wide availability for FraportGPT—a digital assistant that helps all employees with their day-to-day work.
The solution: Starting the cultural shift to AI solutions with Azure OpenAI Service
Fraport already relies on Microsoft solutions, such as Microsoft 365 for its workstations and Azure for its cloud infrastructure, so it made perfect sense to go with Azure for FraportGPT, too. This way, the solution can be scaled in line with Fraport’s planned growth; besides, the company has already built up knowledge of the infrastructure and how to manage it. The FraportGPT bot is already providing support in a wide variety of areas. Programmers use it to create program statements wherever possible. Fraport’s lawyers ask the bot to summarize long texts. And employees in administrative positions, for instance, in HR, use FraportGPT to find inspiration for initial drafts of emails, letters, or other texts. “As a provider of critical infrastructure, we naturally have strict compliance and security requirements. With Azure OpenAI Service, we can meet these requirements and, at the same time, provide our employees with all the functionality and advantages of ChatGPT as part of our infrastructure,” explains Christian Wrobel, Chief Data Architect at Fraport.
As Fraport had hoped, employees have quickly started experimenting with FraportGPT and trying things out in different areas of the company. “An airport is like a city, with its own administration, infirmary, train station, restaurants, and shopping streets,” Dollberg says. “So, we also rent out many properties. One of our colleagues who is responsible for rental management used FraportGPT to help him come up with a suitable formula for calculating offers.”
In the past, people would have spent a long time juggling numbers and trying things out on their own, but today, they can turn to FraportGPT as a digital assistant that can help them think things through and let them try out new solutions. Fraport didn’t specify any use cases when it rolled the bot out; it simply provided some guidance on how to give the bot instructions, an activity known as prompt engineering. Now, when it comes to tasks made up of lots of smaller, time-consuming steps, the AI stands ready to accelerate them in whichever way suits the employee: by offering outside inspiration, preparing formulas in calculations, or quickly summarizing long texts. “We’re planning to connect FraportGPT to in-house data sources so that our employees can access that knowledge more quickly and easily, and in this way, further increase efficiency,” Wrobel says.
FraportGPT has made the topic of AI more accessible to employees—and that’s a definite success for Fraport. “Everyone has an opinion on AI. Some people oppose it, others are open to it, but what we lacked was a common understanding. Introducing FraportGPT has opened up the discussion on AI. That’s a major step in bringing about a cultural change that will really drive innovation,” Dollberg says. No matter how small the first step into the future is, it’s the most important one, and it gets major projects off the ground. Today, thanks to a simple chatbot that’s easy to start using, Fraport employees can try out artificial intelligence, discuss it, and find out for themselves how AI can help them. And 30 years from now, it will benefit travelers passing through the airport of the future, too.
“We’re planning to connect FraportGPT to in-house data sources so that our employees can access that knowledge more quickly and easily, and in this way, further increase efficiency.”
Christian Wrobel, Chief Data Architect, Fraport AG
Follow Microsoft