Other News

AI’s ‘transformative role’ in commercial aviation

AI’s 'transformative' role in commercial aviation
The stakeholders already utilising artificial intelligence highlight advantages abound (Image credit: Savvapanf Photo/Adobe Stock)

Artificial intelligence is revolutionising advancements in technology, and experts believe it is poised to “transform” all areas of our lives. William Hallowell explores the potential of AI in commercial aviation as industry professionals hint at its ability to play a major role in the future of flying

According to Caltech, the California Institute of Technology: “AI and machine learning-enabled technologies are used in medicine, transportation, robotics, science, education, the military, surveillance, finance and its regulation, agriculture, entertainment, retail, customer service and manufacturing.”

Indeed, artificial intelligence is creeping into our everyday lives increasingly – and some industry leaders believe this new technology has the ability to globally revolutionise commercial aviation and the way we fly. From the online travel agent (OTA) that the passenger books with, to the airport they fly from and airline they fly with, there is seemingly great potential for AI in the global commercial aviation industry.

A revolutionary new technology in aviation

Aviation is notoriously a slow industry where the adoption of new technology is concerned. But the stakeholders already utilising artificial intelligence highlight advantages abound. Not only will AI improve services for the passenger, they say, but it can dramatically enhance operations airside and in the air.

From greater safety to maximised efficiency, tech companies are already providing new and innovative AI solutions to improve stakeholders’ operations across the industry – with some already reaping the benefits of this new technology.

Safety is paramount in aviation, and tech solution providers argue that AI would greatly reduce the risk of accidents and enhance safety procedures due to the technology’s ability to make predictions and decisions based on past events. This includes, but is not limited to, predicting flight paths and assisting air traffic controllers, for example.

According to Bartłomiej Rusiniak, chief technology officer at PhotoAiD, a Polish software development company that uses artificial intelligence to create biometric passport photos: “Predictive models will be the most revolutionary AI systems for commercial aviation. These machine-learning systems can provide substantial support in decision-making for air traffic controllers, enhancing overall safety in the skies.

“Advancements in AI also play a crucial role in enhancing air traffic safety through predictive maintenance by assisting controllers in decision-making.

“Additionally, AI, through the use of predictive maintenance, can reduce the costs of aircraft servicing and help in the early detection of potential defects, which can increase safety in the long run.

“It’s fascinating to think that AI [solutions] might also have an important function in airport and in-flight security by smartly identifying passenger behaviour patterns.”

Rusiniak explains that “the potential of this technology is too compelling to ignore”, adding that “I envisage a radical change in air travel in the future” through the ability of AI to perform a multitude of operational benefits.

Daniel Logvin, CEO of blockchain solutions company LedgerByte, argues that one other way artificial intelligence can enhance airlines’ efficiency is through the reduced risk of human error, which ultimately benefits passengers.

He outlines: “Passengers and airlines would be the ones that mostly benefit from AI applications in commercial aviation.

“The reason is because AI would minimise human error, which prevents delays, overlooking problems and a better management of the resources that each aircraft has, providing more accurate information, as well as a more accurate administrative [procedure].”

Replacing legacy technology: AI and blockchain?

A report published in August last year valued the global market for AI in aviation at US$686.4mn in 2022. It projected that by 2032, the market would be worth more than $4.04bn, suggesting a CAGR (compound annual growth rate) of more than 20 per cent. Such growth could encourage a move away from decades-old legacy systems which are at end of life.

One aviation technology expert predicts this transition is only a matter of time. Brendan McKittrick is the CEO and founder Aerobloc, which he envisions as a commons platform that brings together the potential of AI solutions with blockchain technology.

McKittrick tells ARGS: “48 per cent of all tickets reserved globally go through touchpoints with software written in the 1960s and 1970s. All the aircraft in Europe were grounded [last summer due to air traffic control failures].

“And what happened? A legacy system failed because two destinations had the same flight code. This is an early indication of what’s to come. These systems are at end of life.”

McKittrick determines that today’s legacy systems will be replaced by AI on a global scale as early as the next few years because this new technology not only dramatically reduces the hazard of human error, as Logvin points out, but also provides substantial economic efficiency savings.

Improving the customer experience: Rebuilding airlines’ relationships with passengers

While tech experts highlight the opportunities for the role of artificial intelligence in commercial aviation, the passenger will also feel the benefit, according to McKittrick. He says AI and blockchain, like no solutions that have come before, can bring together all parties – from the OTAs to the airports and airlines – to ensure a better customer experience.

However, one challenge to improving the customer experience today is the lack of will to collaborate with competitors and other parties, the CEO believes. But he says a commons platform like Aerobloc would remove this burden.

McKittrick asks: “How many people do you know have been to an airport, had a problem, the airline picks up the ticket and tells them they need to contact their travel agent?

“The passenger then says ‘Well, I booked it online’ and the airline says ‘Yes, you booked with an online travel agent, so you need to ring them because you’re flying on our plane but you’re not our customer’.

“That disconnect is a direct result of these legacy systems not allowing airlines to combine and give customers a single offer automatically using AI or blockchain.

“Airlines have lost that relationship with the customer that they need to regain, and they use loyalty systems to do this, but the problem is that the customer is getting further and further away from the airline.”

That’s the customer-facing element. But how do airlines themselves consider the role AI can play in improving operations?

airBaltic’s AI rollout

ARGS exclusively spoke with airBaltic in January to understand how the Latvian flag carrier has implemented the technology since introducing it in June 2023.

Lauris Mikelsons, airBaltic’s vice president of compliance and safety, highlights how the airline has adopted AI algorithms to analyse and process safety-related issues through automated processes and simulations that can help to predict potential hazards in the future.

“AI’s role is transformative, focusing on continuous learning and data analysis to identify patterns in operational safety reports, which helps in detecting risks and hazards,” Mikelsons outlines.

“This technology enables the faster detection of anomalies in reports that precede more serious occurrences, aiding safety analysts in developing timely preventive measures.”

Further: “AI-driven simulations mark a significant advancement in predictive safety analytics, allowing airBaltic to model various hazard scenarios and gain insights into potential future occurrences.

“This facilitates proactive policy, training and procedure adjustments to mitigate operational safety risks,” airBaltic’s compliance chief continues.

“AI also enhances the productivity of safety analysts by automating the processing of large volumes of safety reports and categorising them by ‘criticality’. This automatic triage helps the safety team to prioritise critical safety reports, focusing their efforts more effectively.”

Mikelsons shares McKittrick’s optimism in AI’s ability to improve industry collaboration for the benefit of the passenger and industry stakeholders alike. He says AI has the ability to “revolutionise” commercial air travel for “airlines, airports, online travel agents, and most importantly, airline customers”.

The most significant impact AI could have in commercial aviation, Mikelsons explains, is its capacity to enhance all data-based processes, starting from the customer experience during online flight booking, up to and including safety analytics based on a multitude of data inputs.

“By analysing vast amounts of data from various sources, AI can identify potential risks and hazards with unprecedented precision, thereby enhancing the overall experience and safety of air travel,” Mikelsons expands.

“This increased safety is not just a benefit for passengers but also for all industry stakeholders, including airlines and airports, as it will lead to more efficient operations, fewer disruptions and fewer incidents.”

He adds that, even from the start of the process, AI can offer customers a better experience at the point of booking through OTAs leveraging the technology’s opportunities.

Receiving results ‘in minutes’

The US and UK-based tech company Aerogility provides model-based AI solutions to allow companies to simulate real-world scenarios to generate insights for forecasting, planning and decision-making, through the use of digital twins. The company works with the British low-cost carrier easyJet to optimise the airline’s maintenance operations.

According to Simon Miles, Aerogility’s head of AI, airlines’ adoption of the technology is about “planning the optimal use of maintenance facilities so they can ensure their fleet is always available whenever they need it to be, that they don’t maintain things too early, but they always maintain them within the window that’s required”.

He adds: “As common with most of Aerogility’s customers, what they were doing before was spreadsheet-based planning and forecasting. The problem is that this just doesn’t meet the complexity of what you need to do when airlines are trying to work out what is the best maintenance schedule.

“It can take days, weeks or months to get a good projection because airlines want to try out different possibilities to say, ‘What if we do this, what if we do that’. The benefit of having a digital twin, especially an AI-driven one, is that airlines can get things done in minutes.”

Miles says machine-learning technologies can be adopted by stakeholders to take sensor data off aircraft and use it to predict when it might next require maintenance – and how long that maintenance might take.

It can also be used to provide information on what height to fly at to reduce aircraft emissions or, through generative AI, extract information from documents and provide data on issues that need to be resolved. But, Miles suggests, AI’s use can differ on a case by case basis.

Improving ground services offerings

In the Summer 2024 edition of ARGS, ground services provider dnata outlined the steps it’s taking to enhance operational efficiency through AI. According to CEO Steve Allen, the company wants to spearhead an approach focused on new technology and innovation through automation and artificial intelligence because “the potential for AI is huge”.

Already, dnata has trialled autonomous, AI-led technology in the ground handling and catering arms of its ground services business. Artificial intelligence may also prove a vital tool for data collection in the future.

“On AI, we already use tools which track an event and monitor [aircraft] turnarounds, and then we use that data to predict what might happen to a turnaround [in future scenarios] and how we speed turnarounds up to meet punctuality targets,” Allen says.

“We’re using data to make ourselves more efficient in terms of managing turnarounds, and how we improve efficiency.

The CEO added: “We use tools to identify waste in the kitchen, [which means] using data to model how much waste we’re producing on an individual flight and then trying to minimise the amount of waste going forward using machine-learning techniques.”

On the catering side, dnata is also using this AI-led approach to inform its airline customers what they should sell. The company now gathers data to show carriers their least popular and best-selling products, for example.

Does the technology need to develop further?

In spite of AI’s advantages, however, there are shortcomings. The recent failure of Air Canada’s AI chatbot is perhaps one instance.

In February, Canada’s Civil Resolution Tribunal ordered the airline to honour a refund policy its website’s chatbot invented when a passenger used the service to seek a ticket refund.

Despite Air Canada arguing the chatbot was a separate legal entity and bosses claiming the bot was “responsible for its own actions”, one member of the tribunal declared that “while a chatbot has an interactive component, it is still just a part of Air Canada’s website … It should be obvious to Air Canada that it is responsible for all the information on its website. It makes no difference whether the information comes from a static page or a chatbot”.

The court concluded there was no reason why the passenger should have known that one section of the airline’s page was accurate while another was not.

Indeed, even dnata’s CEO acknowledges the ground services provider must be “careful in the way we use AI because it still makes a lot of mistakes, and this is why we currently don’t deliver AI directly to the customer”.

“But,” he says, there is “no doubt it’ll get better and more accurate – and eventually we’ll be able to deliver it directly to customers.”