Attention All Passengers (20 page)

• In 2010, NPR broadcast a report about the Aeroman repair station in El Salvador, which was responsible for what was termed “a mistake that could have potentially been catastrophic.” In January 2009, a US Airways aircraft en route from Omaha to Phoenix was forced to land in Denver after the pressure seal around the main cabin door started to fail in-flight. Later it was discovered that Aeroman mechanics had installed a key component on the door
backward
.

An Outsourcing Outbreak

The latest numbers are hard to uncover, but there's no question that outsourcing keeps increasing. According to the Aeronautical Repair Station Association, the maintenance repair and overhaul (MRO) market annually exceeds $50 billion worldwide. Within the United States, ARSA claims the industry has grown to 200,000 employees at 4,200 firms, but small- or medium-sized enterprises constitute 85 percent of this total.

In September 2008 the DOT's inspector general issued a detailed report on aircraft maintenance outsourcing and noted that nine of the ten largest U.S. carriers were outsourcing 71 percent of their “heavy” maintenance and repairs, up dramatically from 34 percent in 2003. Also, 907 repair stations were performing heavy maintenance for the largest domestic carriers. Of these outside facilities, 661 were located in the United States, while 81 were in Central America, 79 in Asia, 69 in Canada, and 17 in Mexico. (The United States and the European Union have a reciprocal oversight agreement—a “transatlantic NAFTA”—so European work is not included.) Overall, the FAA had certificated a total of 4,159 domestic and 709 foreign repair stations by July 2008, an avalanche of new outposts.

But even these numbers don't tell the whole story. Incredibly, regulations allow U.S. airlines to
voluntarily
provide such data. Also, these voluntary requests cover only their “top ten substantial maintenance providers,” even if an airline has hundreds. The IG found that reporting system to be “inadequate” and bluntly stated that the FAA “still does not have comprehensive data on how much and where outsourced maintenance is performed.” There's no question the outsourcing percentages are climbing even higher today.

The IG redacted airline names, but among its other findings was that over a three-year period, FAA maintenance inspectors assigned to one airline visited only four of its fifteen most substantial maintenance providers. In another case, a major foreign engine repair facility was not visited by the FAA until
five years
after the FAA approved it for repairs; such an inspection should have occurred immediately. And one maintenance vendor continued performing work despite failing to respond to a total of
sixty
red-flag open corrective action reports that had exceeded the required response time windows. Yet despite agreeing to a specific recommendation in 2003, by 2007 the FAA still was not properly identifying 33 percent of outside repair stations in its database.

As for oversight, it's mostly the scout's honor system. The FAA relies heavily on air carriers' oversight, and audit programs are not always effective. The regulations are quite loose about when problems need to be addressed, and violations can be resolved simply based on written pledges from the outside repair stations! In fact, the FAA does not require its maintenance inspectors to conduct on-site inspections of outside repair facilities prior to or even
after
approval of those stations—even when problems are found.

What's more, there's plenty of blame to go around. The IG report stated: “At one heavy airframe repair station,
all three types of oversight failed
—FAA, air carrier, and repair station.” In other words, the system is not working from top to bottom. In this case, two internal airline audits and two external FAA inspections failed to detect “significant weaknesses” that eventually shut down the outside facility for more than a month.

Loopholes, mushy language, omissions, communication gaffes. The regulations are Swiss cheese, the airlines exploit them, and the FAA cannot or will not enforce what it clearly could enforce. In addition, there's a tremendous amount of undocumented work being done on the nation's airline fleets, which undermines the big-picture maintenance trend analysis continually conducted by the FAA, airlines, and manufacturers; in other words, if problems with a specific airplane aren't reported at Airline A, how can Airline B follow up?

Has any good come from all this scrutiny? Well, in January 2008 the FAA issued a directive requiring every U.S. airline to simply
list
all the certificated and noncertificated repair stations it uses. A positive first step at getting a handle on this mess, no? Bad news. The FAA promptly rescinded this policy, because the airline industry “expressed significant concerns” with the undue burden of reporting this.
1
Consider that. The FAA merely
asks
the airlines who is performing all this work and even just
listing
their vendors is too hard for Fortune 500 companies in the digital age. Couldn't someone just ask accounts payable where the checks are cut?

In fact, several government entities have been warning about the dangers of outsourcing for years, in report after report. Consider that in June 2005, the IG concluded: “FAA inspectors were not able to effectively use the oversight systems to monitor the rapidly occurring changes.” Translation: electronic surveillance is being employed just when the airlines are offshoring tons of maintenance work. And in December 2005, the IG released a scathing report that concluded: “Neither FAA nor the air carriers [we reviewed] provided adequate oversight of the work that non-certificated facilities performed.” It found that as many as 1,400 noncertificated repair facilities—including 104 foreign facilities—had
never
been inspected by the FAA. Translation: none needed. And in September 2006, the House Transportation Committee held hearings and noted that the FAA had addressed only one of nine IG recommendations made in 2003: “According to the IG, the FAA has not made much progress on those recommendations.” Translation: what else is new?

The inspectors I've spoken to in recent years are based throughout the country. All of them are livid and most request anonymity, for fear of reprisals from FAA management. One said that “All inspections outside the U.S. must be coordinated with the State Department. Nothing is a surprise to them. A red carpet is laid out for you.” Another reported that “Our inspectors just can't get [to Latin America]. Down there they've got twenty unlicensed mechanics with one licensed guy from the airline looking over them.” Yet another told me, “You may have twelve to fifteen laborers working under the license of one mechanic. . . . It used to be if there were twelve mechanics working on an aircraft they were all licensed.”

Dangerous Data Uncovered

Among the many programs the FAA oversees is the Service Difficulty Reporting site, whereby industry professionals are required to electronically log in all manner of discrepancies; eventually the SDRs are available to the public. But after a visit to Massachusetts, I was stunned to learn that the FAA's entire airline database is capturing only a small fraction of the serious incidents and events occurring in the airline industry every week. And a systemic problem with smoke in domestic airline cabins may be avoiding detection.

John King is a veteran mechanic who was licensed in 1963 and eventually fired by Eastern Air Lines as a whistle-blower in 1987 when he reported on that carrier's maintenance improprieties. I visited him and he supplied me with reams of data—so much so that I later told a former inspector that King is doing the work of ten people at the FAA, and the inspector corrected me: “One hundred people.”

King's methodology is decidedly low-tech, but without remuneration he performs a critical and patriotic service by continually Googling key aviation terms:
diversions
,
emergency landings
,
flight returns
,
unscheduled landings
. He then gathers as much information as possible through news reports of all U.S. airline in-flight anomalies. Finally he periodically compares his private database with SDRs collected by the FAA. The results are stunning, and riddled with omissions:

• 2008: 112 of 148 incidents (76 percent) not filed; 132 of 148 incidents (89 percent) not in compliance

• 2009: 96 of 156 incidents (62 percent) not filed; 128 of 156 incidents (82 percent) not in compliance

• 2010: 118 of 183 incidents (64 percent) not filed; 153 of 183 incidents (84 percent) not in compliance

The compliance numbers are so low because additional incident reports either omitted a required cause or omitted required documentation (part numbers, etc.). But two other points are critical. One, King's database reflects only the forced landings and air turnbacks captured by news reports. Two, such diversions are but one symptom of a wide range of potential mechanical problems. In other words, as disturbing as these findings are,
they truly represent just a fraction of existing problems.

Even more chilling, a significant number of these incidents were due to smoke in the cabin: 32 percent in 2008, 24 percent in 2009, and 26 percent in 2010. And overwhelmingly these events were not entered into the SDR database, with only 30 percent correctly filed in 2008 and 38 percent correctly filed in both 2009 and 2010. Consider that no major media outlet in the United States has reported on the fact that our nation's commercial aircraft fleet has had a widespread and systemic problem with smoke in airplane cabins—and the FAA has not been effectively tracking this problem.

“It's dysfunctional,” says King of the SDRs. “It's financed by the government and it's the only mandated program. Without an accurate SDR database, there's no way to gauge the voluntary reporting programs.” And he adds, “This doesn't cover a whole spectrum of other problems.”

Last year, before he resigned following a DUI arrest, I asked Administrator Babbitt about the SDR system, and he responded: “We just want the data. We want a very concrete example of where we can identify that data.” He cited the FAA's Aviation Safety Information Analysis and Sharing (ASIAS) system, and suggested, “Delta may not want United to know about its engine life.”

But whistle-blower Bruno is not surprised by King's findings: “The margin of safety keeps getting worse. It's not a safety agency. . . . Today they are a facilitating agency, to facilitate the airlines' profits.”

The rapid expansion of maintenance outsourcing introduced new risks into the FAA's safety oversight. These missing data underscore both a troubling increase in maintenance failures
and
deficiencies in FAA surveillance.

Double Standards?

At the FAAC, John Conley raised what came to be known as the “single standard.” The administrative vice president for the Transport Workers Union, Conley repeatedly asked if the FAA oversees all maintenance equally, whether in-house or outsourced, in the United States or overseas? FAA officials said yes. But the differences can be striking.

First, there's the issue of certifying mechanics with airframe and powerplant licenses. At one meeting, I noted that the historical model was for ten FAA-licensed mechanics to be supervised by one licensed supervisor, but now we've seen that at many outsourced facilities, ten
unlicensed
mechanics can be supervised by that same licensed mechanic. To the FAA, it's all about the certification—that is, the work is being certified by a licensed representative of the facility and/or airline. But common sense indicates these are two very different models. As Brantley says, “It's one thing if the people doing the work are licensed—but they're not. That is huge. Let's be honest—the only way to get your labor costs down is to use less qualified people.” But Sarah MacLeod, executive director of the Aeronautical Repair Station Association (ARSA), dismisses the value of a license: “I know a lot of educated idiots. The certificate doesn't guarantee anything. . . . Will it not lead to an accident? We can't judge that.”

Then there's drug and alcohol screening; the regulations require periodic testing of mechanics inside the United States, but this rule is waived overseas. At the FAAC, we were told that this is because America can't impose its values on a sovereign nation; true enough, but if such testing is illegal in another country, why is that country repairing U.S. aircraft? Bourne is dismissive of these arguments: “They say we can't make other countries do what we ask—well, we damn well do it with cars. If you want to sell cars in the U.S., you have to adhere to our rules.” Meanwhile, ARSA has been waging a legal battle on this front since 2006, to prevent such regulations from being enforced at more than twelve thousand facilities nationwide. MacLeod says, “ARSA is not against drug testing, it's the level of testing.”

And there are security concerns as well. “We were under armed guard for three days,” says an airline mechanic who visited the Aeroman facility in San Salvador (servicing JetBlue, Southwest, and US Airways). “Guys with machine guns were guarding the parts bin. That's a security problem.” A former air marshal calls the lack of security at maintenance facilities “a huge misdirection of resources.” Bourne points out that while a sixty-year-old unionized ramp worker at Continental is subject to a background check that extends back to his sixteenth birthday, “bin Laden himself could have gotten a job fixing U.S. planes in El Salvador since the background check goes back one day.”

Old-time mechanics recite an industry bromide: If there's a mechanical problem, you can't pull an airplane over to the side of the road. As for outsourced work, one mechanic for a major carrier says problems “run the gamut,” citing improperly rigged cables, miswired and transposed cockpit instrumentation, and drill shavings trapped in wiring bundles. All could lead to catastrophic failures and fatal accidents.