AI and Data science in airlines: why you should become an early adopter

By Hélène Dubos | Trends & Innovation

Sep 10

Artificial Intelligence (AI) is among the most exciting and disruptive technologies in industry today. And there’s a growing awareness of the possibilities around AI in airlines too. Indeed, the market for AI in airlines is expected to reach $2.2bn by 2025, showing just how much the industry is investing in this technology.

Like many new technologies, the spoils of AI will go to the early adopters. The airlines which begin using AI today will gather serious experience and expertise about how it works and how it can be used to improve everything from operations to customer service. That will give them an important edge over late adopters who may well struggle to catch up. 

Let’s look at how AI is being used in the airline industry today, some of the challenges that need to be overcome, and where you can start. 

5 use cases for AI in airlines

Since AI is still a relatively recent technology, its potential use cases in the aviation sector are only beginning to be explored. Nonetheless, the following 5 examples of AI in airlines are indicative of the way the industry is increasingly turning to this technology.

  • Better customer service with chatbots
    Today’s airline customer service solutions are still largely manual, where human operators must interact directly with passengers in person or over the phone. Although AI will not remove the need for humans (at least in the foreseeable future!) it might be used to give customers more options and ways of resolving issues
    For example, SnatchBot or Mindsay are one of a number of companies which uses natural language processing (an AI application) to answer questions, take bookings and manage reservations. Not every customer wants to talk to a bot, but having the option is ideal for some - making it more convenient and faster to solve issues.
  • Enhanced maintenance
    Airlines gather vast amounts of information on airplanes and must generate maintenance plans for every aspect of each individual aircraft. At present, this is largely done through manual scheduling, which means upgrades to hardware and software can be missed or must be made reactively, when something goes wrong.
    Low-cost carrier EasyJet has begun using AI to tackle this. They have begun using a deep learning solution which is able to analyze hundreds of pieces of information about the software used in all of the carrier’s fleet to ensure it is always up to date. This avoids delays to security and software updates - and reduces risk.
  • Less labor-intensive processes
    There are many steps in a passenger’s journey through the airport and these are often perceived as frustrating and inefficient. After all, no one enjoys standing around in queues.

    Delta has showcased how artificial intelligence might help improve this experience. The airline is currently trialing a bag drop and check-in solution which uses facial recognition technology. Rather than queueing up at a desk for a member of the flight crew to check their passport and process luggage, the system instead lets the individual scan their passport and the AI analyzes the passenger’s face to ensure they are who they say they are. This makes the bag drop process a lot faster and means crew can spend time on more valuable activities.
  • Integration with personal assistants
    Amazon’s Alexa is now commonplace in people’s homes the world over. And, a recent collaboration between Amazon and United Airlines showed how this can be really beneficial to users. Passengers of United can ask Alexa about the status of travel and tell the voice assistant to check them in to flights. It’s early days yet for this kind of technology, but shows the direction that airlines are taking.
  • Smarter ticket pricing
    Airlines have long used dynamic pricing to encourage customers to invest in tickets, yet these price variations have tended to be based on fairly limited factors. However, with AI, airlines can expect much more sophisticated dynamic pricing, based on Willingness to Pay, a metric helping to fix the appropriate market price over time. This can help airlines improve revenue and meet customers at the price point they’re willing to pay.

Potential challenges for AI in airlines

While using AI in airlines has enormous potential benefits, it’s valuable to be aware of some of the possible drawbacks and obstacles too:

  • Cybersecurity & Privacy laws: It is essential to ensure that passengers’ personal information is kept safe and treated in compliance with privacy laws, especially when airlines outsource data treatment, use cloud solutions or feed it through new apps.
  • Costs: AI and data science raise new costs for technology implementation and human resources since it creates the need for new skills and jobs on the IT and business side. On the other hand, it should bring new businesses and optimize operations. Airline executives will need to be judicious and be driven by an ROI approach, keeping in mind that slowing innovation initiatives can expose fatal weaknesses in the long term.
  • Proving value: Like any technology, the value of AI needs to be proven. Airlines will need to find their own use cases and use specific measures to assess whether the technology is really providing the value it promises.

Being future-ready

While AI and data science in airlines is still at a relatively early stage of development, the future looks promising - for revenues, safety and customer experience. So where should airlines start?

AI depends more than anything on good quality data - and so a key step to becoming AI-ready is to begin preparing your data to ensure it can be used by current and future AI systems. Conztanz provides a data platform dedicated to the airline industry, providing the foundation to start in this new domain. The platform structures your legacy data and ensures it can be integrated with modern software solutions - contact us today to learn how we can help you become AI-ready.