A critical reading of the news: “FAA Criticized In Report on Airline Parts”
Original story on the Washington Post: FAA Criticized In Report on Airline Parts
At first glance, this story sounds pretty scary.
Passengers have flown on jetliners built with “substandard” parts, some of which may have been made in foreign countries, because the Federal Aviation Administration lacks an adequate system for checking the quality of airplane components, according to a federal oversight report.
Wow! That’s quite an opening paragraph. Next time I get into an airliner I just know it’s going to corkscrew directly into the ground. Let’s see what they back it up with.
The parts for commercial airliners such as the Boeing 727 and 737 were once manufactured almost exclusively in the United States. But the parts on today’s big jets, such as Boeing’s 777 and its planned 787, are made in such countries as China, Japan, Brazil, Italy, France and Australia, in addition to the United States. Boeing, Pratt & Whitney, GE and other plane manufacturers buy parts made overseas largely because they are cheaper.
Ok, so parts are made overseas. So what?
But the bargain-hunting has come at a price, according to a new report by the Transportation Department’s inspector general.
“Neither manufacturers nor FAA inspectors have provided effective oversight of suppliers; this has allowed substandard parts to enter the aviation supply chain,” reads the report, dated Feb. 26. The agency released the report yesterday after it was made public by the Project on Government Oversight, a nonprofit organization that focuses on government accountability.
SourceWatch: Project On Government Oversight
The report cited four engine failures in 2003 — three on the ground, one in flight — that were traced to “unapproved design changes made by a . . . supplier” of speed sensors on engine fuel pumps. It did not cite any more recent incidents, nor did it specify the degree to which continuing problems with parts threaten to cause similar failures.
Even the journalist sounds unconvinced at this point. They didn’t cite anything more recent than 2003, and they didn’t specify the possible danger of unapproved design changes.
And, frankly, only four engine failures in all of 2003? According to the National Air Traffic Controllers Association, there are more than 87,000 flights a day, and an average about 64 million flights a year. Only four failures? Sounds like these parts suppliers are doing pretty damned good to me.
During a visit to one parts supplier, the inspector general’s office observed an employee who “used a piece of paper, scotch-taped to the work surface, as a measuring device for a length of wire on an oil and fuel pressure transmitter.”
And here’s the line that prompted me to write this blog post, after seeing coverage of the article at The Consumerist blog.
My immediate reaction was: so what? The article obviously implies that measuring the lengths of wire with a piece of paper scotch-taped to a work surface is bad in some way, but it doesn’t tell us why it’s bad!
- It doesn’t say whether the wires were the correct length after being measured this way.
- And even more important, it doesn’t even tell you why the wire needs to be cut! If it’s just being cut so it’s short enough to fit in the casing of this particular component, does it really matter how precise the cut is?
Sure it sounds scary, until you actually engage your brain and realize that paper can actually be cut to a specific length! And it can be marked, too! It’s a miracle material! Well, since the article doesn’t explain it, I can only assume it’s there merely to sound scary to people who don’t make use of their critical thinking skills or have never seen paper before.
So far, no airline accidents have been attributed to faulty overseas parts, the FAA said. “There are absolutely no imminent safety issues raised by the report,” FAA spokeswoman Alison Duquette said.
A confirmation that these horrible systemic problems you should be terrified of haven’t actually caused any accidents whatsoever, directly from the FAA.
The report identifies 17 major components of commercial airliners made by Boeing, including the wings, rudder, nose and engine nacelles. When the Boeing 727 was introduced in 1964, all 17 of the components were made in the United States.
By contrast, of the 17 major components of the Boeing 787, which is scheduled to make its first test flight this year, 13 were made exclusively or partially overseas.
“FAA’s process for supplier audits should be designed to address newer manufacturing business models, which have expanded the number of foreign suppliers, locations where parts are assembled, and the degree of independent manufacturing responsibility suppliers now have,” the report reads.
“Exclusively or partially overseas” is a handy statement. Since you bucket “exclusively” and “partially” in the same bucket, there’s no way to know whether all 13 of those components are made entirely overseas, or if all of them merely contain one small screw made in China. It would take a lot of effort to be more vague.
So what is this article really saying?
Ignoring the parts designed to scare you, like the idiotic paragraph about (gasp! shock! horror!) measuring wire with a piece of paper, every other part of this article is basically saying: “getting airline parts from foreign suppliers is bad.” Of course, it never explains why it’s bad. (It does, however, explain one way it’s good: it’s cheaper.) It also doesn’t mention airliners made by the entirely-foreign Airbus cooperative at all.
I’m a pretty patriotic guy, and I even make an effort to buy American whenever I have the chance, but this article is nothing but scare-tactic propaganda. Airliners are perfectly safe, whether we cut the length of wire in Mexico City or in Dallas, specifically because companies like Boeing, backed up by government agencies like the National Transportation Safety Board, have expended great effort to make sure they’re as safe as possible. And yes, that’s not 100% safe… nothing is.
And the part that bugs me most is that blogs like Consumerist are sharing and re-printing this article, with the scary wire cutting quote in the headline, all over the Internet without even stopping to think about it. If you run a blog that reprints stories, please at least spend 5 minutes thinking about the story first.
Not saying you’ll be convinced, but check out the report itself:
http://www.pogoarchives.org/m/tr/faa-supplier-20080226.pdf
http://www.oig.dot.gov/StreamFile?file=/data/pdfdocs/WEB_FILE_Final_Supplier_report.pdf
Have you? Because there are far more examples cited than the ones we or the Post highlighted. Also–and this is really important and I think you should consider this–the IG–which cannot, nor is meant to, replace the FAA and the manufacturers who are supposed to be the on-the-ground everyday presence that ensures quality control and thus safety–found problems everywhere in the system when they showed up to check out what was going on. That, even with the very likely forewarning given to those they audited and inspected.
You can Sourcewatch POGO, my organization, all you want, but it’s not our report, we just made it available (once I got a hold of it and found out it wasn’t public, I thought it should be and put it online).
Frankly, the outsourcing to suppliers abroad thing wasn’t the big issue in my opinion (though the fact that we rely on many suppliers from countries where we lack bilateral aviation safety agreements is significant; that’s a sign that we don’t believe they have a robust aviation safety regulatory apparatus). My main beef was that there seemed to be issues with quality control. I urge you to read through the report itself. Sure, most parts made by suppliers, even with the problems in quality control, may be fine, but if a plane goes down because quality control systems are found to be inadequate–and those inadequacies are known and fixable–then that’s criminal negligence, if not criminal itself. You might want to read the part of the report that cites the new 2003 FAA system which one FAA manufacturing district office representative said led to the reduction of supplier audits by half.
For an example of something not cited in the Post article, but in the report itself (and in our POGO press release):
“Effective oversight of suppliers is essential to ensure that substandard parts do not enter the aviation supply chain. For example, in February 2003, 1 supplier
released approximately 5,000 parts that were not manufactured properly for use on
landing gear for large commercial passenger aircraft. At least one of these landing
gear parts failed while in service. While FAA became aware of this large-scale
breakdown at this supplier in 2003, it has not performed a supplier audit at this
facility in the last 4 years.” (page 11 http://www.oig.dot.gov/StreamFile?file=/data/pdfdocs/WEB_FILE_Final_Supplier_report.pdf)
Me: You’d think a high risk supplier that did something like that, should be audited at least once after 4 years???
Nick,
First of all, no I haven’t read the report, nor do I intend to. I have better things to do with my time and, frankly, the entire point of reading a secondary source like the Washington Post is that they summarize the report so I don’t need to waste my time on it. If the Washington Post report is horribly inaccurate, well, then frankly good: because otherwise the report would be entirely worthless.
Secondly, despite the amounts of words typed here, you still haven’t explained why it’s so shocking and horrible that someone’s using a marked piece of paper to measure the length of a wire. I’m assuming the Washington Post did, in fact, get that out of the report and didn’t simply make it up.
In fact, screw it, I’m going to make a liar of myself, open up the report and find it. The bulletpoint says, in its entirety:
So, nobody can explain why this little gem of knowledge is so scary, not even the report itself?
Thirdly, in response to your second post, one failure in 5,000 parts doesn’t sound that bad to me. (“At least one” is just a weasel-phrase for “one”.) If my car’s parts were that reliable, I’d jump for joy. Did the one failure cause an accident, or endanger anybody in any way? That might be shocking if so, but then again you’d include that in the report if so.
So what is the report really about? It’s about people with way too much free time trying to make themselves feel good by making the rest of us pay more to fly– who do you think pays for the 250 extra pages the FAA adds to their inspection checklists? Us. I’m all for making things safe, but if you’re going to worry about unsafe things, why not crusade against red light cameras, school buses with no seatbelts, and all the other things that are orders of magnitude more dangerous than flying? Flying’s safe enough.
The problem here is we’re talking about precision tools and complex machines in unforgiving environments where minor imperfections can be fatal hence the need for stringent specifications and tools and quality control systems for ensuring them. Though I don’t know the full back story to the 5,000 defective landing gear parts, one failing while in service means possibly hundreds or up to 4,999 had to be replaced or fixed after they were installed on planes–costing a whole lot more than if they were replaced at an early point in the production process, or better yet, prevented which robust quality control systems tend to do.
I don’t know if you know the history of the deHavilland Comet, but it’s an important one in the history of aviation safety. The bottom line there is that a minor, almost negligible issue in design and installation turned out leading to or playing a major part in numerous crashes. The culprit: the window’s square shape exarcerbated by cutting corners in the installation of it–it was punch riveted instead of drill riveted and glued. From wikipedia (and yes, internally sourced):
Investigators began considering fatigue as the most likely cause of both accidents and further research into measurable strain on the skin began. Stress around the window corners was found to be much higher than expected, “probably over 40,000 psi,” and stresses on the skin were generally more than previously expected or tested. This was due to stress concentration, a consequence of the window’s square shape.
The problem was exacerbated by the punch rivet construction technique employed. The windows had been engineered to be glued and riveted, but had been punch riveted only. Unlike drill riveting, the imperfect nature of the hole created by punch riveting may cause the start of fatigue cracks around the rivet.
The principal investigator concluded, “In the light of known properties of the aluminium alloy D.T.D. 546 or 746 of which the skin was made and in accordance with the advice I received from my Assessors, I accept the conclusion of RAE that this is a sufficient explanation of the failure of the cabin skin of Yoke Uncle by fatigue after a small number, namely, 3,060 cycles of pressurisation.”
(http://en.wikipedia.org/wiki/De_Havilland_Comet#Comet_disasters_of_1954)
Yes. I know about the Comet. Was its wires measured using paper instead of a ruler? And that’s why the windows were poorly designed? Is that the point you’re trying to make?
I want an explanation of why measuring the length of a wire with paper is bad or dangerous in some way. If you’re not going to post that explanation, please don’t bother to post another reply.
Also, as you’d probably guess I’d say, I don’t think robust quality control costs us money on balance, but saves, while making us more competitive economically. It’s something the Japanese learned a few decades ago, while Detroit stubbornly refused to embrace–to its peril.
Plus when a car crashes, typically the death toll is no more than a half dozen, with the loss of say a vehicle worth $5-30,000 typically. One plane crashes, you’re talking hundreds of deaths and a machine that costs tens of millions that’s meant to last a decade or longer. I’ll grant you the irrational fear factor that accompanies flying–which statistically has a better track record versus automobiles in terms of deaths–but that irrational fear isn’t going away. Is it worth it to have a plane crash because a supplier didn’t meet their contractually-mandated specs (and they are almost always contractually-mandated)? “Aw man, so you’re saying we shouldn’t calibrate a fuel pressure transmitter to the millimeter using a paper ruler we marked up?”–They’re not marking the increasing heights of little kids on your kitchen wall, but building precision machines. Nor are these shitty consumer items built with planned obsolescence in mind. Our aviation track record is great and it should stay that way.
Yes, an airliner is a “precision machine,” and yet the last one I got aboard was delayed for an hour because of a stopped drain in the bathroom. My point being, not every part of the airplane must be, or even should be, “precision.”
If the wire was cut because of some sub-millisecond timing issue, or to guarantee that it has a particular resistance, then yes, maybe a more precise cut would be warranted. But if it’s only cut so that it fits in some particular physical casing, then it doesn’t matter at all.
My computer is a “precision machine” as well, but you know what? It doesn’t matter how long the SATA cable in it is, as long as it’s short enough to fit the case and long enough to reach the drive. There are thousands of wires on an airliner exactly like this.
Since the report, and the Washington Post, don’t bother to tell me *why* it needs such precision, I’m assuming that the only reason for the factoid is to scare stupid people who aren’t capable of making the above leap of logic.
Yes, there’s an irrational fear factor that accompanies flying– and this study you dug up contributes to it! How can you acknowledge it exists, and is a bad thing, then in the same breath encourage it with pointless fear-mongering, like the paper claim?
Now, for the last time, explain to me why it matters that this wire was cut using a piece of paper. Or, be honest with yourself and everyone else, and admit that that statement in the report is nothing but baseless propaganda.
The fact that there are irrational fears of flying, does not mean there are no legitimate safety concerns. It’s not fear mongering to say that defective parts–some of which failed during flight–got on planes because of quality control deficiencies. It’s not fear mongering, it’s accurate, and I’d rather work to fix the problem. And I can still admit that commercial flight has a great track record, in large part due to regulations, because I like to think that I’m intellectually honest and will admit that things aren’t black and white.
Actually I’ve emailed the OIG asking the question you’ve asked. Why don’t you send them an email too? Feel free to copy and paste.
———- Forwarded message ———-
From: Nick Schwellenbach
Date: 4 Mar 2008 00:53
Subject: Question re: recent audit
To: OIGPublicAffairs@oig.dot.gov
The recent report released on Feb 26 states:
“At another supplier, an employee used a piece of paper, scotch-taped to the work surface, as a measuring device for a length of wire on an oil and fuel pressure transmitter.”
I was wondering about the full significance of this anecdote. I’m assuming the suppliers should have used a more accurate measuring device, but didn’t know what the implications of minor imperfections were in this instance–wrongly calibrated fuel pressure transmitters?
Thanks,
Nick Schwellenbach
Project On Government Oversight
Wow it took a long time for you to say “I don’t know.” Next time you don’t know something, just admit it without the essays about defective jetliners and save everybody’s time.
But congratulations, by sending that email, you are now 10,000 times better at journalism than Frank Ahrens at the Washington Post.
Yay blogs sticking it to traditional media!