D A T A A N A L Y T I C S
Loosely defined as the use of computer
systems to complete tasks usually
done by humans, the use of artificial
intelligence technology within aviation
is growing in parallel with the rise in
use of big data and data analytics.
“Big data refers to data sets of such a large size and
complexity that they present a challenge for traditional
computer systems to cope with in regard to storage
and analysis,” says Noel Trout, chief technology officer
at Avinode Group, the world’s largest third-party private
aviation charter company.
Data analytics, meanwhile, is “the science of analyzing
such data sets to draw conclusions and test a hypothesis,”
he says. “You often hear about artificial intelligence in
conjunction with big data, because many modern AI
techniques require large data sets to be trained on.”
According to Trout big data, analytics and AI will have a
significant impact on almost all industries going forward, and
business aviation is no exception. He says, “There is truth
to the saying knowledge is power. These techniques are all
about systematically collecting and analyzing data with the
aim of producing knowledge and efficiencies.”
Data sources
In pursuit of this knowledge there is a wealth of data sources
to tap into within business aviation. For aircraft operators,
data can be parsed into two distinct subsets, data generated
internally and externally-sourced data.
“If I am an operator there is data from my operations
software with regards to quoting, flown flights,
aircraft maintenance and pilot hours. This can be
used to improve your operations and drive process
efficiency increases.
“The most common external data sets we see
are airport databases, historical flown flights, future
capacity demand and flight tracking.
“Flight tracking has historically been a manual
activity, but more and more companies are trying
to trigger automatic processes based on it. For
example, a push notification sent to the lead
passenger to tell them that their aircraft is delayed”.
It’s possible to go a layer deeper in the analysis than
tracking flights. Amadeus, the Spanish IT provider for
the global travel industry, has a product which tracks
passengers on board flights.
“It tracks what we call Origin and Destination O
and D traffic,” says Rodrigo Acuna Agost, head of AI
research at Amadeus. “The product estimates all the O
and D combinations in the world. This could be relevant to
18 | BU S INE S S A I R P O R T INT E RNAT I ONA L J U L Y 2 0 1 9
How much data is t oo much?
One of the big imponderables when it comes to
handling data is the question of what data to collect.
Rodrigo Acuna Agost, head of AI research at Spanish
aviation IT provider Amadeus believes it makes sense
to hold on to as much as data as possible because
“even if you don’t know what to do with it today you
may find a use for it in the future”.
This applies particularly to flight data. Agost is
less convinced about the usefulness of retaining
passenger data, which in any case can pose a
security concern if confidential information of
individuals were ever to be hacked. Amadeus, for
example, does not retain any data on individual
travellers from the commercial airline sector. Agost
says, “We only use the data they give us when they
travel. Nothing else.”
This approach is taken partly to avoid laws
around data management and partly to avoid the risk
of mixing data, since Amadeus’ products are used
across multiple channels of the travel industry. It is
also because most of those travelling via commercial
aviation do so infrequently, with nearly 90% taking a
flight only once a year, he says.
In private aviation where flights are more
frequent and the relationship to customers more
personal, there may be more advantages to
leveraging data that helps tailor services to the
individual. “You might want to feed the data through
a customer relationship management tool that
remembers your meal choices next time you come to
book, for example,” he says.
business aviation in terms of understanding where people
are travelling and what potential markets there might be out
there for them.”
Amadeus, provides IT for both customer-facing
applications and aircraft operations. In the latter
group Agost points to a schedule recovery product,
which uses data on flight schedules and status,
passenger information and aircraft maintenance to
mitigate the impact of disruption. He says, “It revises
fleet operations in an optimal way so that you come
back to your operations quickly to minimize delay
and cancellations.”
The area where Amadeus’ AI products have
found the biggest application is in the booking
process. One of their products is a computer model
that predicts travel choices. Agost says, “When
you are looking for flights between two places there are an
overwhelming number of choices. Our choice model can
predict from a range of flights the one you are more likely to
choose and offer it to you.”
AI on the Up
Most of Amadeus’ customers are in commercial aviation.
As of yet AI and big data have found comparatively fewer
L “Flight tracking has
historically been a manual
activity, but more and
more companies are
trying to trigger automatic
processes based on it”