Lufthansa Chief Digital Officer Christian Langer believes that AI is bringing an end to planning as the airline industry knows it, provoking a mixture of fear and excitement in his business and the sector.
Langer is excited about AI because Lufthansa is rich in data. A single Airbus A350 produces roughly 1.5 terabytes of data per day and a single General Electric GEnx (General Electric Next-generation) jet engine produces 150 million data points on an average flight. This data can help Lufthansa improve its current services and develop new ones.
He's worried about AI because it also creates opportunities for more nimble competitors to challenge the established airlines. Langer expects around $48 billion of venture capital funding to be invested in the industry in 2018, and knows of 2,500 travel tech and mobility startups, all of which are monitored at the Lufthansa Innovation Hub in Berlin.
These startups threaten to replace the traditional motor of the airline industry: planning systems.
Change of plans
Lufthansa uses planning systems to manage its staff, passengers, aircraft and spare parts.
"Our customers unfortunately are way more flexible than our assets, so we need planning systems," Langer explained at the AI Summit London. "We plan with very long planning horizons."
Lufthansa schedules its flights 18 months before they take off, its planned maintenance six months in advance, and crew timetables four to five months ahead of the flight. These traditional systems have served the company well but the era of long planning cycles is coming to an end.
Flexible startups are isolating links in the chain between the passenger and the airline that they can target with data analytics to make real-time decisions for each part of the customer journey. Lufthansa has to reimagine its strategy to compete.
"All of our intelligence in our planning systems won't be of value anymore in five-to-ten years," said Langer. "We have to rethink the whole way we plan and operate our business. We still need some assets - aircraft and spare parts and crews and all this stuff - but the way we plan them and operate will change dramatically. This will devalue all the knowledge we have, all the expertise we have in our pricing algorithms, in our network planning algorithms, in our maintenance schedules."
Upstarts are arriving to challenge every element of the Lufthansa business. The traditional model of aircraft ownership and operation is under attack from Amedeo, an aircraft leasing firm that buys old planes and plans to offer them to new entrants such as Airbnb.
There will also be new aircraft on the way to take on Lufthansa's fleet. Lilium is developing an electric plane with vertical take-off and landing capability, which could carry a handful of passengers from the roof of one inner-city building to another.
Other startups are disrupting the ticket marketplace. Hopper analyses billions of flights from airlines around the world to predict prices and offer customers lowest fares.
"Now we have to find the capacity to fill the needs of the customers, so we have to have a way more flexible asset base than we had in the past," said Langer. "To rethink our business in that way is a huge step for our industry."
Planning for the future
Lufthansa has responded in a number of ways. The airline has developed a predictive maintenance platform called Aviatar that analyses data from aircraft parts to continuously assess the condition of each component.
"We have close to 2,000 aircraft in the Aviatar who are constantly monitored," said Langer. "Every second we ingest something like 60,000 data points in our systems coming from all these aircraft in the air, parts in the workshops or aircraft on the ground. We combine all this data and build our models and are able to predict the future behaviour of this specific spare part."
The system is particularly useful for any parts hidden under the exterior panels of a plane, such as hydraulic connections that can leak without regular maintenance. Lufthansa traditionally had to open the panels up to check the condition, but can use sensors to predict when a leakage is likely to occur.
Aviator showed its value just days ago, when the system suggested a hyrdaulic leakage was likely in an Airbus A320 that was due to fly a 'VVIP' to their destination within hours. A mechanic was called who confirmed the fault and made the repair, allowing the flight to take off as scheduled before the start of the World Cup.
The result showed how AI can give Lufthansa an edge. By combining the tech used by startups with the resources and experience of a major airline, the company can add fulfillment to prediction.
"All this prediction is worth nothing without fulfilment," said Langer. "What does it help us if we can predict there's going to be a leakage if we don’t have mechanics at hand who can fix it?
"What does it help if we can predict that in five hours the time ticker system or the turbine condition warning system is going to fail, but we are not able to have a spare part ready at the place where the aircraft is going to land?"
Lufthansa can also use data to improve the onboard experience. If analytics suggest a large proportion of passengers will be Chinese, the airline can ensure that there are enough Chinese-speaking flight attendants. If an individual passenger historically preferred an extra pillow on their seat, Lufthansa can supply it without them having to ask.
"This is why we are feeling fine when it comes to the end of planning," said Langer. "We have the data in our hands. We have good partners and now IT crews to help us with the prediction part. We are able to combine the physical with the digital world and we are able to fulfil on what we predicted."