My last blog said we should stand by for the rise of AI in aviation. I underlined that as an industry we’d better fasten our seatbelts, as AI’s traction in aviation is becoming overwhelming! So why is that the case (particularly for an industry with a reputation for being change and risk adverse)? Well, we’re clearly seeing three major triggers at play:
- First, companies today can now leverage comprehensive data (Big Data generated through digital transformation). They can then access the powerful algorithms and cheaper GPUs required for the unprecedented processing power to harness that Big Data.
- Second, the capacity clock is ticking. In 20 years’ time the number of passengers will double to reach 8 billion, says IATA. The number of aircraft will double too. But the number of airports will increase by only 7%, creating an imbalanced supply and demand ecosystem with congestion on the ground and more disruptions in the air.
- And third, digital transformation has arrived. It’s identified as the most valuable project for the air transport industry. It creates a lot of Big Data that needs to be mined as part of the digital transformation objectives of organizations.
AI is no longer a ‘nice to have’ … it’s becoming mandatory
Forward-looking airlines, airports and ground handlers are embracing AI and embarking on the ‘road to optimization’ so they can leverage their data assets and remain relevant and competitive. This will help to ensure passengers will still have a seamless journey, despite the upcoming congestion. And it will enable industry players to achieve the objectives of their digital transformation. These objectives include efficiency in disruption management, industry collaboration, security and cybersecurity, as well as the delivery of a great passenger experience.
With these challenges and opportunities in mind, AI is no longer a ‘nice to have’ option. It’s becoming mandatory; a business imperative for every player in aviation. It will ensure they remain relevant, with personalized offers and services for extremely fickle customers who are ever more demanding and tech savvy.
AI in aviation is a breakthrough that will enable digital transformation of all players in the air transportation ecosystem. As I said before, we’ve identified seven key areas where AI applications will change our industry, from the passenger and bag journey, to border management and aircraft operations. (See below.)
A story from the frontline: Optimizing turnaround with Machine Learning, Computer Vision and Operations Research
Let me now dive into a great example of an AI use case in aviation. During our SITA Innovation Forum earlier this year, delegates in our Operational Intelligence workshop identified turnaround time as a key business challenge where they foresee great potential to apply AI.
Why turnaround? Well, we all know that the turnaround process can adversely impact on-time performance of airlines, airports and the overall passenger experience – often at great cost, both monetary and reputational. Today, the process is managed by a turnaround manager who follows a rules-based list of activities that need to take place, and ensures that all these activities are performed.
At SITA, we’re prototyping an AI solution that leverages Computer Vision to capture and explain the turnaround process, from on-block to off-block. More importantly, it aims to provide real-time recommendations to proactively manage disruption.
We’re using Machine Learning first to understand and explain video captions of what’s happening around the aircraft during turnaround. This will complement existing features to better explain turnaround activities and anticipate disruption during the process. We’re also using advanced optimization algorithms to provide the best sequence of activities that should take place under some constraints to effectively manage turnaround and disruption.
Airports: a ‘playground’ for AI and OR
Our goal is to provide airlines and airports with best recommendations based on Machine Learning and optimization models to efficiently manage and optimize turnaround time. An airport is an extremely highly structured environment. It’s a ‘playground’ to leverage AI and OR. Together they can harness all the data available to predict and optimize turnaround, and to effectively manage disruption.
The outcome of turnaround optimization with AI will be increased collaboration between machines, humans and robots. This will efficiently manage all the activities to proactively handle disruption, and to optimize turnaround operations and related activities. At the same time, we’ll be achieving operational excellence, thanks to collaboration and coordination between the resources involved in the turnaround process.
So, from where you should you start your AI journey? The AI IMPACT Cycle …
So that’s one use case. But from where should you start your AI journey? I would say for most aviation industry players, you must first identify your key business challenges and priorities. Think hard about all those business challenges where AI could be applied to help you to achieve your goals. To simplify this overall exercise, think about the AI IMPACT Cycle (see diagram ).
Take a ‘baby steps approach’
To remain relevant, you need to make a start. As you may have seen in my last blog (‘Fasten your seatbelts! AI is taking off in aviation’), organizations are making significant investments into in AI now. You can’t afford to wait for the technology to mature, or have all the data you would need, and be fast follower. Start your AI journey leveraging a ‘baby steps approach’: begin with small projects to test and quickly learn and adjust.
AI will play a key role in about every aspect of the aviation industry. The industry is at a pivotal point where AI’s capabilities are a competitive differentiator, thanks to processing power, great algorithms and comprehensive data. Data is the crude oil for our industry. It is the most valuable and powerful asset in today’s globally connected world. You need to leverage AI to mine this data and extract the most value for your business.
Look out for my series of blogs on AI in aviation
This is the second in my series of blogs. Over the coming months, I’ll be producing a blog for each of the seven areas of AI applications in aviation that we’ve identified:
- Passenger Journey
- Bag Journey
- Airport Operations
- Airlines Operation
- Ground Handlers
- Border Management
- Aircraft Operations
Read my last blog: ‘Fasten your seatbelts! AI is taking off in aviation’