The FAA is not alone in its conviction that Boeing has solved its battery problem. In Japan, JTSB spokesman Mamoru Takahashi told my researcher Takeo Aizawa that the safety board was not surprised or concerned by subsequent cases of cell venting. This was similar to what the board’s Koji Tsuji had told me. “We cannot create the situation where no short circuit happens, but we have reached the point where we can control the heating, even if venting occurs,” he said. “We can consider it a minor problem.”
“Boeing has made all the conceivable improvements for their 787s to ensure the resumption of normal flight,” including more than eighty modifications that “are desirable, but may not be necessary,” Tsuji told me. It was eerily evocative of what was said about the Comet before it was returned to service with the source of its problem still a mystery.
1 Now known as British Airways.
2 Now Yangon, Myanmar.
3 Queen Elizabeth had already taken the throne, but her coronation was delayed to observe a period of mourning for the death of her father, King George VI, in February 1952.
4 “Report of the Public Inquiry into the Causes and Circumstances of the Accident Which Occurred on the 10th January 1954, to the Comet Aircraft G-ALYP, http://lessonslearned.faa.gov/Comet1/G-ALYP_Report.pdf.”
5 Tony Booth, Admiralty Salvage in Peace and War 1906–2006: “Grope, Grub and Tremble” (South Yorkshire, U.K.: Pen and Sword Maritime, 2007).
6 This was the first of four modifications to the Parker Hannifin design.
7 USAir was renamed US Airways in 1997. In 2012 it merged with American Airlines.
8 John Boudreau, “In Praise of the 787’s Emotional Experience,” San Jose Mercury News, September 25, 2012.
9 “Aviation Investigation Report A98H0003,” Transportation Safety Board of Canada, http://www.tsb.gc.ca/eng/rapports-reports/aviation/1998/a98h0003/a98h0003.asp.
PART FOUR
Humanity
An accident sequence is like someone slipping down a knotted rope. The pilot’s decision may be the last knot in the rope, but there are many other events which set up the accident sequence. Pilot error is increasingly seen as far too simplistic.
— MAURICE WILLIAMSON, NEW ZEALAND’S
MINISTER OF TRANSPORTATION, 1999
The Right Stuff
Sometimes in the summer I go kayaking with my friend Pete Frey, a pilot for a U.S. carrier whom you met earlier in this book. For a man in his late fifties, Pete’s in great shape. Still, it’s hard for men of a certain age to pull off the baggy shorts and windblown hair look. Truth is, Pete doesn’t even try to look hip.
In his pilot’s uniform, however, Pete is a babe magnet. It’s not just the navy jacket with the gold braid; it’s his confidence, his completely unselfconscious “you’re safe in my hands, baby” attitude. Pete the airline captain is all authority and competence.
Around the world, tens of thousands of airline pilots make a similar transformation. That harried blonde in line at Starbucks, the guy pumping gas into his pickup truck—when these suburban parents and weekend anglers don their uniforms, airlines trust them with multimillion-dollar airplanes and the companies’ reputations. This is why the making of airline pilots is serious business. Airlines screen them before they hire them, test them once they’re in, and train them repeatedly throughout their careers. The airlines want seemingly contradictory qualities in their pilots: decisive but open-minded, vigilant but flexible, experienced but constantly learning, adherence to standard procedure but with an ability to improvise when required.
Creating a “pilot-type person” is so important that 95 percent of the cadets hired by Lufthansa have not flown at all when they arrive at the company-owned Airline Training Center in Arizona, fifty-seven hundred miles from company headquarters in Frankfurt, Germany. The thinking at the carrier is that if they can find people with the right personality, they can “grow their own pilots,” as Matthias Kippenberg, president and CEO of the training center, explained. I saw this process up close in the fall of 2010, when I joined class number NFF380 for a week.
Lufthansa might as well be looking for astronauts; that’s how hard it is to get selected. And with good reason: at the time, it cost thirty-five million dollars a year to run the school that turns cadets into airline pilots. “Students are selected by the airline, trained and sponsored by the airline,” Kippenberg said. “They have a job guarantee,” he said. “All they have to do now is learn to fly.”
Kippenberg graduated from ATCA in 1977 and has been running the program since 2002. He may not have appreciated how hard his students work to get accepted until his own daughter Lisanne applied at the age of nineteen to take an Introduction to Aviation course offered by the Swiss government. She qualified to apply because her mother is Swiss, but, like everyone else, she would have to take a series of aptitude exams. In preparation, she and her dad went online for guidance—only to be confounded by the very first practice test. On the screen were six three-dimensional cubes. An X was placed on one of the six sides of each cube. Lisanne had to listen as a voice instructed her to imagine the cube rotating up or down, right or left, back or front. Lisanne’s job was to keep track of where the X ended up.
“My dad and I were looking at it, and then we looked at each other; we didn’t know what was going on,” she told me. The “nine clocks test” was equally baffling. I’ll spare you the details.
The students in my dorm at ATCA told me they’d performed similar tests, and others in which they had to listen to long series of numbers and repeat them back in reverse order. It sounded like torture to me, and I had the chance to experience it myself when CTC Wings of New Zealand put me through my very own pilot aptitude evaluation.
This differed from the kind of screening conducted by airlines because applicants at CTC are paying their own way, with no assurances they will find a piloting job. The aptitude test is necessarily less selective and less intense. Even so, having to keep the wings of my simulated aircraft straight and level while flying through a series of yellow rectangles appearing on my computer screen, I was tense and sweaty even before the mental processing exam began.
My overall score suggests that as a pilot, I make a great writer. I could process information okay and even acquire data under time pressure, judging by how quickly I was able to make sense of an image of shattered glass. (Don’t ask.) When it came time to handling a joystick while following a flight path and working math questions involving counting backward, I was fried.
Viktor Oubaid, head of the German Aerospace Center in Hamburg, told me that all these challenges are designed to test working memory. One could practice and get better, as Lisanne and the Lufthansa students did, “but the maximum possible performance depends on your original abilities,” Oubaid said. “In other words: many people can learn to fly, but only some are able to work as airline pilots.”
After several months spent practicing for the tests, Lisanne went to Dübendorf, outside Zurich, for her entrance exam. The setting was quiet, and she had a good feeling. “I did a lot better than I thought I was going to do,” she told me. Sure enough, Lisanne was accepted. After two weeks of flying lessons, she said she was even more excited about a career as a pilot and more appreciative of what the tests were trying to determine about her cognitive abilities.
Lisanne’s dad knows very well this aspect of her experience. Call it multitasking or workflow management—pilot aptitude tests are designed to detect this and other things because so much more than epaulets is riding on the shoulders of the men and women on a flight deck. This was made most evident five years later, when one of the students at Kippenberg’s school passed right through the airline’s tightly woven web of diligence.
Andreas Lubitz arrived in Phoenix right after I’d left, when he was twenty-three years old. He had aced five days of testing and had interviewed with confidence back in 2008, when he was selected to begin ground school in Bremen, Germany. Yet Lufthansa’s notably tough standards may have proven to be too m
uch, because after two months, he took a leave of absence. From January to October of the following year, he underwent psychiatric treatment for reactive depression that a German medical examiner told the FAA had been triggered by excessive demands.
By 2010, Lubitz was considered fit to continue his training, and so he did: ground school in Bremen, and then flight school in Phoenix, followed by jet training back in Bremen and a stint as a flight attendant. In 2013 he became a first officer on Lufthansa’s low-cost carrier GermanWings.
In the spring of 2015, Lubitz would commandeer his own flight from Barcelona to Dusseldorf and fly it into a mountain, killing himself and 149 others. The thirty-four-year-old captain, Patrick Sondenheimer, had left the cockpit to go to the bathroom after leveling the plane at thirty-eight thousand feet. With Sondenheimer gone and the cockpit door locked, Lubitz put the Airbus A320 on an autopilot descent to one hundred feet, a path that would take the plane directly into the high terrain in the French Alps.
Lubitz overrode the captain’s attempts to return to the flight deck and did not reply to radio calls from controllers. For eleven minutes the plane descended, until finally it hit a mountain near Prads-Haute-Bléone. It would soon come out that Lubitz’s depression had returned, and that in the weeks before the event, his physicians had advised him not to work, according to notes found in the trash in the young man’s home.
Suicidal and/or homicidal airline pilots are a special kind of scary, even though this is an exceedingly rare occurrence. During its investigation into the crash, the French air accident bureau reported six similar events in which commercial pilots had deliberately crashed planes with passengers aboard. Japan Airlines in 1982, Royal Air Maroc in 1994, SilkAir in 1997, EgyptAir and Air Botswana both in 1999, and Mozambique Airlines Flight 470 in 2013. These crashes were all believed to have been purposely initiated by one of the pilots. With more than 717 million flights since 1980, you can see how infinitesimal this particular flight safety risk is.
It is an entirely different story when you get to unintentional acts. Then the mistakes pilots make that contribute to disasters are everywhere you look.
Sole Responsibility
None of the six people on the private jet from Samoa to Melbourne on November 18, 2009, will ever forget the night they ditched into the choppy waters of the Pacific off the coast of tiny Norfolk Island. Capt. Dominic James and First Officer Zoe Cupit were piloting the medical evacuation flight, with patient Bernie Currall, her husband, Gary, and a medical team riding in the back. As the Israel Aircraft Industries Westwind jet approached the island where the plane was to refuel for the last leg to Melbourne, the weather was so bad that neither James nor Cupit was able to see the runway. After four missed approaches, James decided to put the plane down in the ocean before it ran out of fuel.
The jet split in two on impact with the rough sea. Only three of the six people on board had life jackets, but all got out of the plane. For ninety minutes, James played shepherd, swimming around them in a circle and keeping them together. Finally, a search party on a charter fishing boat spotted the light from the small flashlight James was waving toward shore, and the group was rescued.
“It gives me goosebumps still thinking about it,” said Glenn Robinson, an island resident and one of the crew members on the boat that rescued the survivors. “They’re all alive. You know they’ve ditched that plane into a rolling ocean in the middle of the night, and here they are.”
Brainy, articulate, tenacious, and the spitting image of the actor Tom Cruise, James was a hero and a celebrity when he arrived back in Australia. “Gold standard” is how he and Cupit were described by John Sharp, the chairman of Pel-Air, the company whose airplane the pilots were operating that night. By Christmas Eve, however, the two had fallen from grace. The Australian Civil Aviation Safety Authority (CASA) suspended their licenses, claiming the two had demonstrated bad airmanship. The accident was entirely the fault of the captain, said John McCormick, CASA’s chief.
A man who puts a jet aircraft into a dark and stormy sea on a moonless night and then, without a life preserver, keeps his passengers together for an hour and a half like some kind of aquatic sheepdog is not the kind of man to allow himself to be made a scapegoat.
In planning for the flight from Apia, Samoa, James said he had been hampered at every turn. Unable to get WiFi on his phone or at the hotel, he did his flight plan on his cell phone in the hotel parking lot. He fueled up the jet with the assumption that it would fly in reduced vertical separation minima airspace (RVSM), a horizontal slice of the sky between twenty-eight thousand and forty thousand feet that requires planes with highly calibrated altimeters and special certification, both of which this Pel-Air plane did not have. Because flying in RVSM airspace can reduce fuel consumption, pilots routinely beg their way in by explaining that they are on a medical flight. James said it was a company practice he had complained about the year earlier, but nothing had changed.
When weather deteriorated on the way, James did not have enough fuel to make it to an alternate airport. He was not required to, under the work rules that applied to air ambulances. This was a loophole that national safety authorities had been trying to close for years, but CASA had failed to take action.
Years later, when James got his license back and started flying for other operators, he realized that many tools were available that could have changed the outcome that night. “I had access to flight planning software wherever I went. I had access to performance data, so I could look at a destination and calculate weights, speeds, and options,” none of which he had access to while flying for Pel-Air.
These shortcomings and others were noted in a special audit of the airline that CASA conducted right after the ditching. Thirty-one safety deficiencies were found. CASA noted a conflict between “the commercial objectives of the company and safety outcomes.” That’s bureaucrat-speak for saying the operator was “more worried about profits than safety.” Nevertheless, the Australian aviation authority’s position was that the operation of the airline was not relevant. The accident, it stated, had been “caused by poor fuel planning, poor decision-making” by the captain.
Two years before the Pel-Air mess, George Snyder wrote in an article for the Flight Safety Foundation’s AeroSafety World magazine, “The assignment of blame artificially and prematurely restricts the investigation process” and can even stop the investigation in its tracks. It was a prescient bit of writing, because that’s exactly what happened in the Pel-Air case. Even knowing the airline had safety lapses, the head of CASA, John McCormick, did not share them with the accident investigators because the ditching, he said, “was entirely the fault of the captain.”
Captain James admits he made mistakes, but adds that none of the support that would have helped him do his job was there. “I didn’t operate in a vacuum. I operated as a pilot who belonged to a company that was overseen by a regulator,” he said. “You can’t isolate one from another and say that’s a fair appreciation of what took place.”
An Australian television documentary program called Four Corners presented the full story in 2012, just as the ATSB was issuing its probable cause report. The program prompted several parliamentary hearings into the way the aviation agencies were doing their jobs. This intrigued me, because issues of safety can be nuanced, and that isn’t always appreciated in politics. Yet here was a case where the politicians sounded reasonable while the aviation professionals were looking no further than the pilot.
“It is surprising and dismaying both,” said John Lauber, a research psychologist in the field of human factors—essentially, trying to understand why people do the things they do. He had been a member of the NTSB, and he’d spent most of his career trying to improve support systems for pilots. “All human performance takes place in a context, defined by the technology, procedures, and training.”
On a bright morning in October 2014, my sister Lee and I were on vacation in Darwin, the capital of Australia’s Northern Territory.
We’d booked a seaplane flight into the aboriginal territory called Sweets Lagoon. Outside our hotel, the bus to the airport was waiting, and out of the driver’s seat bounded a tanned and attractive man who looked vaguely familiar.
“Christine Negroni?” he asked with a smile, extending his hand. “Dom James.” By then I had recognized him. My surprise was seeing him driving a tour bus. When he’d told me his story over lunch at a restaurant in Sydney half a year earlier, he was dashing off for a pickup job he had flying corporate jets. By then, CASA had restored his pilot’s license. He was hoping for full-time work, but he had been vilified by the nation’s top aviation official, and that had taken a toll. He would go for job interviews, but it was always the same story. “All these people would say, ‘the crash, the crash, the crash.’ They didn’t know about the senate inquiry. All they know is I’m some guy who crashed a plane,” he told me. “You can’t make someone be educated on the nuances of the accident.”
Pilots err for many reasons, Lauber told me. “To say a pilot made a bad decision is not a reflection on that pilot, but a reflection on the overall design of the system that he is tasked with operating.”
The Prevention System
There’s a story about an airline captain who, having landed at the airport and parked the jet at the gate, announced to the departing passengers, “Welcome to your destination, ladies and gentlemen. The safest part of your journey has come to an end.”
Flying is so safe that we can appreciate the joke. From the very first plane crash in 1908, attention has been focused on finding out what went wrong and how to fix it. For decades this meant modifying the airplane or engines, and sometimes both, as was the case with the Comet; or coming to grips with new technology, as with the Dreamliner. Dana Schulze had spent more than a decade as an air safety investigator when she started the two-year project to understand what happened on the 787. She told me it was rare to have a case where her team was concentrating solely on the machine. By the time the probe was finished, though, human errors had been discovered in quality control during the manufacturing of the battery cells and in the assumptions made by engineers during the certification of the airplane.
The Crash Detectives Page 17