This story was available to PaxEx Premium subscribers first. Sign up now and get early access to the news shaping the industry.
“Inflight connectivity doesn’t just create revenue, it could save the airline industry US$15bn a year.”
That’s a bold claim from Inmarsat and the research it commissioned from the London School of Economics (LSE). Much of the savings comes from better weather forecasting and the associated effects: reducing delays and fuel burn. Part of the forecast savings comes from predictive maintenance opportunities, allowing the plane to track its own performance and use on-board connectivity solutions to report back to headquarters when operations are less than nominal. The so-called Internet of Things for Aviation (IoT/A) has long been held up as the financial savior of the connectivity platforms, delivering the necessary financial support to justify installations. What will it take to realize the $3-46bn in annual savings the research revealed? A lot of work, and it is unclear which connectivity vendors are truly committed to that effort.
Read More: Gogo 2020 program aims for massive cost cuts
Airlines today feel compelled to “keep up with the Joneses” in delivering connectivity for passengers, but are often unable to make the business case on that alone. Inflight connectivity is a cost center for the airlines. And, despite estimates from similar Inmarsat/LSE research suggesting that passengers will be willing to pay once on board, data from connected aircraft today tells a different story. Take rates are woefully low where passengers must pay and the revenue is insufficient to justify the airlines’ costs. Or, as Inmarsat Aviation’s SVP Market & Business Development Fredrerik van Essen recently summed it up:
A lot of airlines have bought in to the idea of inflight connectivity from a high-level perspective or competitive pressures. They have been sucked into this and figured they could make the business case internally based on selling sessions to users. They’ve run into the problem of not being able to close the business case on that. They need a wider scope, bringing in the operational aspects. They’re still struggling to quantify what that would exactly mean.
That struggle includes figuring out how to realize the promised operational cost savings on the fleet. Just talking about it at high levels does not actually deliver the goods.
Nodes on a network
Bringing the aircraft into the digital world of aircraft maintenance, making them “nodes on the network” is a common rallying cry for this work. Just connecting the aircraft remains only part of the problem, however. Additional, significant effort is needed to integrate services and data, taking advantage of the pipe but also knowing which data to move and when to move it.
David Bartlett, Panasonic Avionics‘ new CTO, believes that his company’s pivot to use cloud-based resources avoids many of the data challenges. “With the cloud we’ve taken away all the limits. We can do the heavy ingestion of data for every flight, for every route, to drive better results… You’re going to see an explosion of new features and apps coming out of this architecture.” Like many grand plans in this industry this, too, may come to pass. But it requires the right pipe to get all that data off the plane. Currently the market has too little capacity to do so on the thousands of connected aircraft and the price for the data transfer remains too high.
Airlines and service providers must optimize costs and avoid just “dumping everything” off the plane all the time. That approach mostly works on the ground where connectivity costs are an order of magnitude lower. But some processing and filtering must remain in the sky for the economics to play out correctly.
On-board server hardware manufacturer Kontron believes its kit will help deliver the necessary on-board processing, regardless of the connectivity provider or which systems are being monitored. Improving the systems capacity on the planes allows that “Large chunks of [processing] could definitely be done on board. It makes sense to do it real time on board and then filter that,” according to Product Manager RJ McLaren. “If there’s data that needs to be highlighted and messaged off, you can prioritize that and send smaller packets of information with your connectivity system and do that intelligently.”
Fortunately most service providers use the hardware from Kontron (or similar kit) to manage some processing on board rather than dumping all the data. Adding that intelligence to the process remains a necessary part of the development cycles for service providers and application developers. Expecting sufficient bandwidth to dump the full raw data stream will leave vendors and airlines woefully ill-equipped.
One major challenge on this front is that the connectivity vendors are often not the application vendors. This requires more advanced collaboration to share the server resources on board or the installation of extra servers on the aircraft. The latter is especially cost-prohibitive while the former represents more of a logistical challenge, especially given the number of airlines seeking multiple connectivity providers across their fleet. Developing for the different portal and server platforms of the various providers makes life difficult for the application providers. If the complexity can be solved the net outcome should be positive, but getting there is a long and challenging road.
Baby steps forward
Some progress is being made on these efforts. Gogo describes its goal as “building the framework, it is not just the pipe to the plane.” With 60,000 crew tablets online on Gogo-serviced aircraft the operational data definitely flows. But this effort remains in its nascent stages. It is more about service recovery – flight rebooking or letting passengers better manage delays – than optimizing the operational performance. The recently announced Gogo 2020 plan raised some questions about just how much the company will focus on internal developments of the framework versus just the pipe. In general the company is “not developing those applications but we’re delivering the connection for them.” And the company recently lost its Director of Connected Aircraft TJ Horsager to APiJET a company more focused on the analytics and application side of the business.
Global Eagle‘s Chief Strategy Officer Walé Adepoju sees the necessary confluence beginning today, “A number of the airlines are looking at this right now and bringing us together with their other application providers. This is happening with the EFB, navigation, flight operations and maintenance.” That’s good news, but it is still a piecemeal effort on the part of these vendors, “[O]n adoption there are bits and pieces that are working but there is no one that has become a comprehensive digital airline yet.” Adepoju was the most optimistic of the vendors, suggesting that the effort to hit the integration targets is nearing a 20% milestone. There is still much work to be done.
Multiple vendors believe the discussion around operational benefits of connectivity is finally reaching the C-suite level of the airlines. These are the people who can make the necessary budget allocations or change management commitments to the efforts. Or they can scuttle them. Just getting in the door is a good first step but vendors must also show solid RoI numbers and what the platforms look like, not dreams of 20 years down the line.
The LCC Conundrum
Spirit Airlines is adding inflight connectivity to its fleet. Perhaps that alone is enough evidence that LCCs must adopt the technology or fail in their markets. In the US, where connectivity is far more ubiquitous and mature, that argument alone might play. Elsewhere, however, it is a harder sell.
Across the market in general, but especially for LCCs outside the USA, the “book away” phenomenon remains relatively low across most segments because enough of competition also lacks connectivity. Or because the flights are sufficiently short. For segments under three hours many passengers can forgive the lack of connectivity, even when their survey results state otherwise. Spirit flies an average of 1,040 miles per sector while EasyJet comes in at 686 and Ryanair at 770; AirAsia hits 800 miles per sector average. The extra 30-45 minutes of flight time can be the difference between passengers feeling compelled to connect and stay in touch or to disconnect and relax for a short flight.
Also, when travelers connect on those shorter flights they are increasingly using lower bandwidth and lower cost options. This could be seen as a great excuse for an airline to install a lower CapEx system like the SITAONAIR SwiftBroadband solution but the per byte costs remain terribly high; there is no good option to scale that up for operational usage or to eventually deliver better service for passengers who want more than just the occasional SMS message.
Inmarsat’s van Essen believes that the LCCs will eventually adopt connectivity specifically because of the operational value it presents, “Their angle seems to be even more than what we expected not on the ancillary revenue side but on the operational performance side. That is driving the case. Providing airlines with a ready-to-go ecosystem where they can get these savings is what they’re looking for.”
Alas, van Essen also acknowledges the major problem the industry faces on this front, “The ecosystem isn’t really there yet.”
Perhaps it is unfair to pin the responsibility for this work on the connectivity providers. After all, they’re working hard enough to deliver uninterrupted bandwidth to planes jetting around the globe. But it also appears that just being the “pipe” to the plane is not enough to convince airlines of the larger value play. More work must be done at the value-added layer to bring about the benefits airlines desire. This does not mean that the connectivity providers must become fully vertically integrated, delivering all the services end-to-end. But integrating those efforts into the other work around building the inflight connectivity ecosystem is critical. Airlines need to know it is part of the plan. Inmarsat’s van Essen touches on this demand, “We’re now enabling it in the pipes but the apps are in a very embryonic phase. I see a lot of intention, but not the results.”
Those results need to happen a lot faster than today’s pace shows for the savings to arrive and the industry to finally deliver the long-promised value-proposition in this area.