By Joel Achenbach, Scott Higham and Ashley Halsey III Washington Post March 29, 2014
The bizarre tale of Malaysia Airlines Flight 370 comes at a time when flying is safer than ever. Nervous fliers squeeze the armrests for dear life, but most passengers have no problem nodding off as their jetliner cruises seven miles above the Earth. They have internalized the statistical truth that the most dangerous part of an airplane trip is the drive to the airport.
Yet disasters still happen, including this one. Officially declared a plane crash at sea with no survivors, the event remains so deeply mysterious that it seems premature to refer to the people aboard as deceased.
Viewed in the broad context of aviation safety, this weird case actually fits snugly within a recent pattern: Airline disasters now tend to be unprecedented in nature — what investigators call “one-offs.”
In the old days, planes typically went down because of engine failures, wind shear, collisions or some other familiar problem. But turbine engines almost never fail these days. Improved radar helps pilots dodge the lethal downdrafts of thunderstorms.
Collision-avoidance technology commands the pilots of converging planes to make diversionary maneuvers.
Even if Flight 370 is the ultimate one-off, aviation safety advocates say there are lessons to be learned from this tragedy, starting with the need for tamperproof equipment that would stream data in real time to satellites and reveal a plane’s position.
“This could well be the kind of ‘black swan’ event that requires everybody to carry a location device,” said Chris McLaughlin, senior vice president of Inmarsat, the satellite company that helped calculate the likely flight corridor of the missing plane.
“As a ticket-payer, wouldn’t you like to know that the authorities know where your plane is at all times?” he said. “This is not expensive. We’re talking maybe a dollar an hour or less to get that information off the plane.”
Mary Schiavo, former inspector general of the U.S. Department of Transportation, which includes the Federal Aviation Administration, would go even further. She’d like to see the full “black box” data — a massive amount of information about the plane’s performance — streamed to ground locations during flight.
“There’s lots of reasons to have streaming data — not the least of which is to foil criminals — and to solve the mystery of what happened to your plane. It would have helped in 9/11. It would have been immensely helpful here,” Schiavo said.
Star Navigation Systems Group, a Toronto company, has designed black-box technology that streams data in real time. It costs about $50,000 per plane, plus $10 more to transmit the data for each flying hour, chief executive Viraf S. Kapadia said. Only one customer has bought the system, he lamented.
“It’s all about the money for the airlines, even though it’s not a lot of money,” he said.
“We’re using 1970s technologies instead of the data streaming that’s been available for nearly a decade now,” said Paul Hudson, president of FlyersRights.org. “The fact that airplanes can basically disappear over water is not only troubling to passengers, but it also gives a heads-up to those who would do a terrorist act or a murder-suicide. They would be virtually undetected. These planes can fly to almost any place on Earth nonstop.”
Sean Cassidy, an Alaska Airlines pilot serving as national safety coordinator for the Air Line Pilots Association, International, said his organization is wary of such a sweeping change in industry practice.
“How is the data being used? How is it being protected? And how is it really driving up the safety case?” he said. Referring to Flight 370, he said, “If this is such an incredibly, exceedingly rare event, I think it’s a little bit premature to begin discussing the way we transmit some of this data until we understand if that really would have made much of a difference in how this flight ended up.”
Katie Connell, a spokeswoman for Airlines for America, the trade organization for the leading U.S. airlines, said in an e-mail that it is “premature for us to speculate and/or discuss potential changes to safety and security procedures.”
Safety advocates contend that real-time data-streaming technology will become more critical in coming years simply because there is more air traffic and longer routes are being flown over increasingly remote areas, including the North Pole and great expanses of open water.
The industry roughly doubles in size every 15 years. People like to fly.
The civil aviation industry has been working with authorities for years on what is known as NextGen technology. Among other things, the new technology would allow airplanes to be tracked more effectively by satellite. The FAA’s work on NextGen has been sluggish, however. Few airlines are willing to invest billions of dollars until the FAA delivers the final regulations that will govern the system.
The human factor
Flight 370 was a modern aircraft with modern technology. The Boeing 777 had multiple robust mechanisms for communicating with or sending data to the rest of the world during flight. The transponder sent a radar signal and another system transmitted data via satellite to ground stations. But these systems were either turned off intentionally or somehow became disabled. No one in the cockpit radioed a distress signal or activated a hijacking code. No one used the jetphones in business class.
Authorities have released no evidence of wrongdoing or malicious intent by anyone on the flight crew. But in general, as technology improves, the remaining hazards in aviation are often found in the cockpit. Among the issues that safety advocates worry about are fatigue and inadequate training.
“What’s causing accidents these days is humans figuring out ways to crash perfectly airworthy airplanes. That’s the area that we really need to be focusing in on,” said John DeLisi, director of the Office of Aviation Safety at the National Transportation Safety Board.
He cited three recent examples of pilot error: Colgan Air Flight 3407, which crashed on its approach to Buffalo in 2009, killing all 49 people aboard; Comair Flight 5191, in 2006, in which 49 of 50 people died when the pilot used the wrong runway, one meant for small planes, and crashed; and Asiana Airlines Flight 214, which came in too slow in San Francisco in July and hit a seawall, resulting in three deaths.
DeLisi declined to address Flight 370. There is no evidence of pilot error, pilot suicide, hijacking or any kind of terrorist event, nor is there evidence of a mechanical malfunction, fire or decompression. There is, in essence, no evidence of anything other than that the aircraft did not go to China as planned but rather flew in a zigzag fashion into the southern Indian Ocean.
The Flight 370 disaster has some similarities to the tragedy that befell Air France Flight 447, which on June 1, 2009, with 228 people aboard, vanished in the Atlantic Ocean during a thunderstorm off the coast of Brazil. It took investigators five days to find the wreckage and two more years to locate the plane’s black box.
Learning from disaster
After the Air France disaster, the International Civil Aviation Organization, a U.N.-sponsored organization based in Montreal, began to study technologies designed to locate airliners when they go missing. Several task forces examined in-flight streaming technologies and other ways to improve the chances of finding the black boxes, boost their battery power and increase the strength of the beacons they send out.
Recommendations of those task forces have yet to be implemented. This is a complicated process at “technical, political, legal, economic and operational levels,” ICAO spokesman Anthony Philbin said in an e-mail.
But Philbin put in a plug for the industry as a whole, calling it “arguably the most significant cooperative achievement in the history of human endeavour and perhaps our greatest shared example of what countries and peoples can ultimately achieve when they work together toward consensus-driven objectives.”
Any discussion of safety has to acknowledge the progress made in recent decades. Major plane crashes have become exceedingly rare. Disasters are instructional, and airlines and regulators make adjustments. For example, after the TWA Flight 800 disaster off Long Island in 1996, in which the plane exploded when a spark ignited fumes in an empty fuel tank, airlines began filling those empty tanks with nitrogen to prevent any such sparking.
Robert Benzon said the improvement in safety was one of the reasons he retired after 27 years as a lead investigator for the NTSB.
“I got kind of bored because I had no accidents to go out on,” Benzon said. “When I first started in the big-aircraft investigation business, each investigator in charge had two, sometimes three, concurrent investigations they were working on. Nowadays, if you’re really lucky as an investigator in charge, you’re doing one major accident. And even that is pretty damned rare now.”
Cassidy, of the Air Line Pilots Association, said passengers trust pilots for good reasons.
“By the time you get to the captain’s seat, especially with a major airline, you have been checked, you’ve been trained, you’ve been through countless training situations, you’ve been evaluated by many, many instructors. By the time you get to that point, you should have a high degree of credibility and respect,” Cassidy said.
But DeLisi, the NTSB official, noted that some new pilots with small companies are paid shockingly low salaries.
“We love airline travel, and yet the economic model is one where new pilots starting out in their careers typically working for regional carriers are making, pick a number, $25,000 or $30,000, working long hours,” DeLisi said. “The number one thing for safety right now is to have a well-trained, well-rested, non-distracted crew.”
DeLisi said technological innovations will never obviate the need for a human in the cockpit. He cited the 2009 case of Capt. Chesley “Sully” Sullenberger, who famously landed a US Airways jetliner safely in the Hudson River after the plane struck a flock of geese and suffered engine failure.
“He did an amazing job of taking an extremely rare failure mode — impacting birds when you’re flying over the Bronx — and landing a disabled jet that is no longer producing thrust,” DeLisi said. “That defines why we still have humans in the cockpit. Because there’s no way you could have ever programmed the autopilot or had an unmanned vehicle make those decisions he made that day to safe those lives.”