Big data analytics is already at the forefront of the aviation sector, and will continue shaping the way airlines manage customers and business processes. In a recent study, 61 percent of aviation companies cited big data analytics as their top business priority.
Yet, a familiar conundrum voiced by many aviation players—especially those making their first forays into big data—is figuring out the best way of managing and analyzing massive quantities of data. Indeed, airlines have much to learn still from other industries, such as retail, in unlocking the value of big data to develop innovative business practices and forge competitive advantages in a fast-paced business environment.
Germany’s Otto Group has big ambitions for integrating innovative, data-driven business models into its existing processes. The retail giant implemented predictive analytics software to capture 300 million data sets from social media, every single week.
By integrating this information with 135 GB of historical data on customers and their preferences, Otto managed to improve demand forecasts for over 100,000 items per day. The IT department can now provide procurement staff with more than one billion predictions, of which 89 percent are accurate—a remarkable transformation, considering the previous forecasts for 63 percent of Otto’s products were off-the-mark by a margin of 20 percent or more.
Destination: Greater efficiency, more satisfied customers
The good news is: the aviation industry generates huge amounts of valuable data across many different processes on a daily basis—with the potential for airlines to reap myriad benefits out of the vast data volumes. Here, I would like to focus on how a data-driven approach can enhance strategic decision-making for airlines in two key aspects: optimizing core operational processes, and reinventing customer engagement.
First, let’s consider how airlines can leverage big data to gain greater business visibility and improve operational efficiencies.
The latest generation of aircraft, such as the Boeing 787 Dreamliner, produces several terabytes of data on each flight that could be potentially tapped to optimize flight paths, streamline maintenance operations and reduce fuel consumption. Data analytics can also be used to make more accurate prediction of flight arrivals, which is crucial as flight delays can result in cost overruns for airlines and airport operators who are already operating on very thin margins.
Secondly, major airlines are exploring how to integrate and make use of data collected across disparate systems to stay competitive and create more dynamic passenger experiences. Price optimization is an important market strategy for aviation players, considering how flight reservations can be prone to seasonal demand fluctuation. Demand volatility remains a prevalent challenge in the travel sector, and airlines can leverage aggregated customer data—captured across the travel value chain, from online reservation systems to social networks—to deliver tailored promotional airfares.
In the face of growing competition, real-time consumer insights are critical to enable airlines to deliver more personalized customer services based on passengers’ past preferences, as well as maximize revenues from ancillary services such as priority seat bookings and car rental.
Implementing a successful big data strategy
Big data certainly isn’t just hype, and airlines need to start adopting new tools and fresh ways of thinking about their big data implementation. The key here is to derive better predictions, faster decisions—in order to streamline operations, better understand customers, and ultimately boost profitability.
To achieve this, airline businesses first need to embrace a more holistic view and build a complete big data “ecosystem”, encompassing high-performance storage, a dynamic cloud infrastructure, and in-memory computing capabilities.
Challenges associated with managing big data solution typically come down to concerns about having the bandwidth capacity and analytics tools to handle multiple data streams, gathered from a variety of touch points. At the same time, you need to take into consideration that there are many data silos in aviation, sitting on both the operational and customer-facing sides, and these disparate systems and data sources need to be effectively harmonized.
By providing both the cloud platform and advanced data analytic solutions on a scalable ‘pay-as-you go’ model, T-Systems can support aviation businesses to develop a solid foundation for data mining, without the need for upfront investment.
Our approach also involves developing the capabilities to process massive data volumes in the fastest and most efficient manner, thanks to the rapid maturation of technologies such as Hadoop. In the retail space, for instance, T-Systems supported Otto in the implementation of Hadoop clusters alongside its existing data-warehousing solutions, enabling the retailer to transform their front-end processes and pave the way for unique customer experiences. What’s more, Otto has outlined plans to tailor product offerings to each customer, based on the Hadoop platform, to deliver personalized suggestions to online shoppers the moment they log on.
 Wikibon, The Industrial Internet and Big Data Analytics: Opportunities and Challenges, September 2013