Artificial Intelligence in Aviation - Where are we?

AI – Artificial Intelligence. A huge buzzword over the last few years, presented by aviation media coverage as varying between promise of salvation and the menace of a dark future. However, Artificial Intelligence is not a distant, entertaining Hollywood fantasy like in the Matrix movies. It’s here and it’s changing our daily life. Not with a big bang, but slowly progressing from one day to another, AI is effecting our lives. But, speaking of slowly, think of what has changed over the last 10 years. Did you book your flight on your smartphone? No, you didn’t even own a smartphone, right? Because there were none at that time (okay, they did start to slowly take off in 2009). Did you put your e- ticket in your digital wallet back then, check-in online, pre-order meals or send a parcel with just a few clicks or swipes? Same answer: no. The digital infrastructure we are used to, with its convenient possibilities – it simply didn’t exist. But where are we now, especially when it comes to Artificial Intelligence in aviation?
Artificial Intelligence software
Let’s leave hardware out of our considerations for the moment. As staggering as the development leaps of smartphones or tablets have been in the last years, let’s take a look at the software instead. Its breathtaking progress has had an impact on airlines and airports. Due to constantly new platforms and algorithms, our daily life, our jobs and our leisure behavior has changed drastically and probably will change even more in the future. The hardware is what supports Artificial Intelligence, but it’s the algorithms that are Artificial Intelligence. What are airlines doing with this great progress?
To give a rough sketch of an answer to this question, we need to narrow down the topic even more. Instead of talking about the aircraft itself, being in the skies on its way to the next destination, let’s talk about everything that needs to be done to make flights happen. After all, aviation starts on the ground. One aspect where Artificial Intelligence in aviation is getting used more and more frequently is maintenance prediction. Airlines want to use AI to foresee potential failures of maintenance on aircraft. Artificial intelligence frameworks today anticipate when maintenance is required even before something gets broken, considering brisk fixes and limited time on the ground. Supposed “wellbeing monitoring” of aircraft enables data to be dissected more rapidly and precisely, empowering preventive activities to be quickly passed on to airlines.
Want to learn more about predictive maintenance? Check out last year’s blog text. Now, let’s move on!
Another use of AI is in Smart Logistics where machine learning algorithms are being applied to data to help automate airline operations. This is big and aviation industry leaders are starting to use it. One adopter is Southwest Airlines. Until now, the airline has shown only limited effort to implement Artificial Intelligence, but there is some evidence of the airline using machine learning to improve their operations. Jeff Hamlet, former Director of Air Operations Assurance at Southwest Airlines stated that he and his team used time series analysis and pattern recognition to enhance their data mining capabilities. Data mining is the art of digging through huge amounts of data to recognize data patterns – that’s basically what Google does. Hamlet says that these approaches enabled his team to identify potential flight glitches found in pilots’ data reports.
Enhanced passenger experience
Today, Artificial Intelligence in aviation is also frequently used to enhance the customer experience on different levels. One common example is the previously mentioned check-in before boarding, which is a vital task for an airline. Artificial intelligence helps to do this easily. This same technology can be also used for identifying passengers as well as to assist customers in the airport. AI can help a company reduce its operational costs and labor costs at the same time. To date, we can see Artificial Intelligence in aviation being implemented by airlines for facial recognition, customer Q&A, factory operation optimization and aircraft fuel optimization. However, AI has much more potential than these few use cases and can completely revolutionize the way airlines do business. Considering that, it’s no surprise that more than 52% of airlines companies across the world have planned to install tools based on Artificial Intelligence to improve their customer service functions in the next five years.
But despite having a bright future ahead, Artificial Intelligence in aviation is facing some serious pitfalls. First of all, it needs to work flawlessly. Aviation is a business where often the lives of human beings are dependent on the fault-free operation of the respective solutions. Also, implementation and familiarization with these solutions take time. There’s no big-bang go-live on a certain day that instantly brings change. Rather, it takes years of detailed multi-step work, many iterations, wrong tracks and retries. The airlines and airports have to find out what works and what doesn’t, what the customers will embrace and what they’re going to give the cold shoulder. It will also be expensive – a tough condition for a business with notoriously narrow profit margins; for example, a simple chatbot will cost an airline more than $15,000. Thus, AI can be unattainable for smaller companies which could create a barrier between small and big airlines in the future.
Baggage loss reduction
One more application: What about baggage loss reduction? Can Artificial Intelligence help reduce the number of unhappy passengers sent to the Lost and Found? What are the latest IATA developments in this regard? Well, that’s the topic for my next blog. So, to learn more, stay tuned!
What’s your experience both as a customer and as an employer in the aviation industry with Artificial Intelligence? Have things gotten easier for you or even more complicated? Tell us what you think!