It’s Scandalous to Keep Using Black Boxes in Airplanes Like MH370
The Australians have predicted that the search for Flight MH370 will cost $234 million. Given the likely duration and current lack of results, that’s probably a low-ball number. But I have a suggestion for where to send the bill:
The International Civil Aviation Organization
999 University Street,
Montreal, Quebec H3C 5H7
The ICAO is a United Nations agency. It is supposed to represent the interests of the world’s airlines—and the interests of everyone who flies on those airlines. It sets “the standards and recommended practices” for the airline industry. Within the ICAO is the Air Navigation Commission, charged with “the safety and efficiency of international civil aviation.”
You might think that with that term “air navigation” would go a concern that in the 21st century it would be impossible for a 330-ton airliner to disappear without trace.
But since Flight MH370 vanished there been no sign that the ICAO is ready to deal with what should be the most urgent air safety issue of the day: Why we are still dependent on a system that allows all the crucial data about the whereabouts and condition of an airliner to disappear along with the airplane?
The ICAO had plenty of time in which it could have shown an understanding of the problem—five years, in fact. It was the search in 2009 for the wreckage of Air France Flight 447 that first drew attention to the obsolescence of the black box (flight-data recorder) in the new age of real-time data streaming.
That search, which ended up costing $160 million, was nowhere near as demanding as the one for Flight MH370. The flight path of the Air France Airbus A330 was known with some precision, as was the time of its disappearance. As a result, 640 pieces of floating wreckage and 50 bodies had been found within three weeks. And yet it would still take another two years before the flight data recorder was discovered.
It also turned out that, by chance, the French safety regulators, the Bureau d’Enquetes et d’Analyses (BEA), had the benefit of something that intimated the value of real-time data streaming. In the last minutes of its life, Flight 447 had automatically transmitted 24 separate bursts of data called fault messages that were received via satellite at the airline’s maintenance center in Paris. These indicated that for some reason the airplane’s flight control computers had shut down, leaving control in the hands of the pilots.
This made the French investigators look harder at the vulnerability of large airplanes flying over remote and deep oceans without an instantly retrievable record of their progress.
In 2010—while the search for Flight 447’s black box was still going on—the BEA, using data from 44 previous accidents, simulated 597 crashes spread around the world’s major air routes in which real-time streaming took the place of the black boxes. (Interestingly, this test was conducted with the help of the London satellite operator Inmarsat, whose role has been critical in the search for Flight MH370.)
As I have reported earlier, the results were astonishing. In 85 percent of the crashes, the streamed data would have given as complete a picture as the data stored in the flight recorder—and in 82 percent of those cases, investigators would have been able to pin the location of the wreckage to within a four-mile radius.
In spite of this compelling proof that the whole principle of the black box—that all the vital evidence in a crash goes down with the airplane—had been made obsolete by technology that was ready to be adapted to airliners, the industry displayed no sense of urgency to do so.
There were, to be sure, many challenges in the path of winning worldwide acceptance for the change. Data-streaming technology was advancing rapidly, and there was an understandable reluctance to freeze it at a stage that might be well short of its potential. Much of the argument centered on bandwidth—how big a “pipe” was needed to handle the same amount of data that is stored in the memory chips of the black box.
There are at least 200 discrete pieces of information, called parameters, required to give a picture of what’s happening to an airplane’s vital systems and its behavior—including its position, altitude, speed, stability, course. (All of the things we need to know but don’t about Flight MH370).
But there is no reason any longer to use the fear of premature adoption as a reason for inaction: The capacity is there.
Another and more frequently invoked concern was the cost.
In the case of the conventional flight data recorder, the international standards for the amount and quality of the data to be stored were constantly improved as the capacity of chips to store data made quantum leaps. The costs of doing so were marginal. It involved simply upgrading a device that already had its place in the airplane—really like downloading an upgrade for your computer.
In contrast, introducing a completely new system like data streaming was not going to be cheap—at least by the lights of airline bean counters. But if the airlines saw only red ink, the owners of global satellite networks saw a lucrative new market. Companies like Inmarsat and Iridium Next were touting a new generation of satellites and they targeted aviation as a huge opportunity to build markets that would last for generations.
In the end, nothing could happen without an international agreement on standards, and the only instrument capable of delivering that is the ICAO.
To be fair, given the number of players and the involvement of many nations, even a nimble-footed, and aggressive regulator would be seriously tested by this problem. The ICAO, a bureaucracy with 191 members, is neither of those things. Indeed, the influential industry magazine Aviation Week, editorialized rather carefully that data streaming “looks increasingly necessary” and lamented that “the ICAO is never able to move quickly.”
Like many other agencies of its kind, the organization holds many seminars and conferences around the world, junkets that are very popular. It also issues annual reports on airline safety. The latest, released on April 10, recorded what is a highly creditable statistic: Worldwide deaths in commercial air travel fell from 388 in 2012 to just 173 in 2013. ICAO Secretary General Raymond Benjamin said, “The aviation community is always grateful to see its commitment to safety rewarded,” and then added, “but as recent events in Malaysia remind us, even one fatal aircraft accident is too many as far as we are concerned.”
The concerns raised after Air France 447 for the industry supervised by the ICAO did more to describe the problems of switching from the age of VCR to the age of Netflix than they did to inspire anyone to find a solution. And so, eventually, the issue seemed to go away.
And then came Flight MH370.
That changed everything in a way that the Air France accident did not, probably because the Malaysian Boeing 777 has become a spooky phantom fueling many wild theories. And it has flamed into a worldwide issue. People are angry. They just can’t understand how it is that a FedEx package can be tracked to the far corners of the Earth but an airliner apparently can’t. Will this anger (and the fear behind it) become a sufficient force in itself to pressure the industry finally to stop depending on the black boxes?
The ICAO doesn’t actually need to make the change mandatory to every airline at once. The priority should be—and can be—given to those airplanes flying over water. Many domestic flights all over the world never venture where a wreck would be hard to find.
There is no doubt that eventually the whole globe will be covered by a satellite network able to know where every airplane is in real time, all the time. Indeed, the shape of those networks is already visible.
The shameful thing is that while the airlines have been dragging their feet on this fundamental safety issue, they can’t wait to give passengers what they call “the fully connected experience.” The same satellite resources—and companies—that would be used to live-stream vital performance data are bidding to offer Wi-Fi and Internet services to every seat in the cabin. In the U.S., United and JetBlue already have some airplanes equipped with an Internet connection via satellite, and others will quickly follow.
Disgracefully, there are parallel universes here. In one, the adoption of new technology involves intractable problems and unwelcome costs (actually no more than $1,500 a month per airplane). In the other, a wonderful new world of cabin amenities beckons everybody to be an early adopter. No worries here about costs. The talk is all of new revenue streams, not data streams.