When you think of machine learning in action, it’s easy to imagine analytics-based marketing research or a matrix-style AI takeover. But there are actually far more grounded and practical applications that benefit hundreds of millions of people in their day-to-day leisure and business activities. Take the travel industry, for example. We’ve now reached a point in human history where over 100,000 flights are taking off every single day. With airports seeing more foot traffic than ever before – Dubai International alone saw over 78 million passengers in 2015 – they need to be run and managed like a carefully oiled machine. Alleviating bottlenecks is central to ensuring as smooth an operation as possible in the travelling experience.
In this blog, we’ll take a look at a few ways machine learning can improve travellers’ experiences and aid the industry as a whole in adjusting its operations to accommodate the new era of global travel.
Airlines and travel sites can help you plan the perfect trip
Nearly every organisation involved in the travel industry has been keeping records of its activities for decades on end – be it airline companies with holiday destination statistics or tourism agents with the popularity of certain seasonal attractions. Until fairly recently, all of this data had little use beyond fact-checking and surface-level insights.
With the rise of machine learning and predictive analytics, however, that data can be put to work to better align passengers with travel routes and amenities. Imagine planning a trip to Shanghai for a conference, but instead of having to wade through an endless stream of flight routes, hotels and activities for the trip, you’re immediately presented with options that are, statistically, best suited for you. It does this by looking at any trends in your previous flight info, such as a tendency to fly at night rather during the day, or to fly direct instead of having a layover. This is combined with information data from your online purchasing habits (those gold clubs you bought off Amazon, for example) to craft a customized trip that appeals to your hobbies and lifestyle. While it may seem eerie for a computer system to accurately predict a suitable trip based on its understanding of you, it’s only a few steps away from the technology behind automatically curated news feeds.
Airport security can be less intrusive
So you’ve embraced the ease and convenience of an automated, customized itinerary, and are now on your way to the airport anticipating the arduous task of getting through airport security. But security measures really needn’t be as intrusive and time-consuming as they currently are. That’s the basic premise behind Silicon Valley-based Qylur’s proposed solution – the Qylatron. It’s an automated luggage scanning device that uses machine learning technology to determine whether the contents of any particular bag are potentially harmful or illegal. To accomplish this, the machine uses X-rays and chemical screening to scan the contents of a bag and then compare each item to an internal library of prohibited objects and substances. The Qylatron is capable of learning about new prohibited items from its operators, as well as from other machines like it .The system can process ten pieces of luggage every minute, which is far quicker than the manual and often intrusive inspections conducted at most airports today. And considering the influx of travellers passing through airports on a daily basis, systems such as these can ease the massive backlogs and frustrations that often characterise the traveling experience for many.
You’re more likely to arrive on time
Perhaps the most important way machine learning can improve business travel and holiday travel alike is by predicting exactly when planes will take off and land – thereby reducing flight delays and runway congestion. It does this by analysing all the variables affecting departure and arrival times, like weather and possible ground issues, and then giving the prediction to technical staff at the airport to update details on all surrounding flights. Accurate prediction of take-off and landing is incredibly valuable in the airline industry, as it’s historically been difficult to account for variables in any meaningful way. This is reflected in the seven-minute average delay time when a plane is off schedule. Knowing when delays are likely makes it easier to avoid a knock-on effect on other flights. Time waiting at the gate is reduced for passengers and for the airlines, the cost-savings is huge: each minute in reduction per flight could also save $1.2 million in annual crew costs and $5 million in annual fuel savings for a midsized airline
Connecting the dots to further elevate the travelling experience
When augmenting travel information with data from hotels, restaurants, conference centres and the like, we start to uncover more opportunities to elevate their experiences by providing travellers with personalised suggestions on amenities, activities and leisure experiences. And thanks to internet ubiquity and mobile devices accompanying people wherever they go, reaching travellers to make personalised suggestions in real-time as they move from point A to B is a simple matter of following the crumbs of personal preference and past behaviour data they leave behind. The Global Travel Ecosystem from Amadeus is a great example, as it uses GPS and real-time algorithms to turn a traveller’s journey into a contextual and tailored experience – from personalised flight and hotel recommendations to suggested leisure activities and shopping. Travel services providers in turn benefit from increased conversion on complementary services or product offerings
Contact us for more information on how predictive analytics and machine learning can and is helping businesses provide products and services that centre on a personalised customer experience.
Image credit: https://www.travelocity.com/
About the Author, Julian Diaz
Julian Diaz is Head of Marketing for Principa. American born and raised, Julian has worked in the IT industry for over 20 years. Having begun his career at a major software company in Germany, Julian made the move to South Africa in 1998 when he joined Dimension Data and later MWEB (leading South African ISP). Since then, Julian has helped launch various South African technology brands into international markets.