Human error

Quote: Nareed

Since there were no other Air Canada planes running out of fuel on a regular basis. We can assume any such errors were caught and corrected somehow.

But that is the point. Perhaps 10% of the time people miscalculated. Perhaps 5% of the time, when the plane was turned on, the Captain could see from the fuel gage that there wasn't enough fuel, and he had to call out the crew to top off the tank. None of these mistakes might have been noted and written up. Then after perhaps dozens of times, the gage was broken so that the mistake was never caught,

Generally it is human nature to not make a big deal out of an error that is caught before anything bad happens. It is quite possible that the gages rarely break. No one envisioned both problems happening at once.


History documentary of Ethiopian hijacking

Quote: Pacomartin

We had an interesting case at work.
We were told to charge our plane tickets on our federal credit cards, and then itemizing them as an expense. We would include the receipt from the plane ticket as proof that we spent the money (just like a hotel receipt or a car rental receipt).

Sadly, I knew people who would charge expensive refundable tickets to their credit card and expense them back to the company. The scam was they would also buy a cheaper ticket and return the expensive one. They would make hundreds of dollars each time they did this.

Quote: DRich

Sadly, I knew people who would charge expensive refundable tickets to their credit card and expense them back to the company. The scam was they would also buy a cheaper ticket and return the expensive one. They would make hundreds of dollars each time they did this.

How did the pool light work out?

Quote: DRich

Sadly, I knew people who would charge expensive refundable tickets to their credit card and expense them back to the company. The scam was they would also buy a cheaper ticket and return the expensive one. They would make hundreds of dollars each time they did this.

That is a standard scam, but it doesn't really work with federal travel. The control of federal aviation rates is pretty careful. In the nationwide investigation, they only found one person who was really milking the scheme.

With my colleague his defense was that his secretary was filling out the forms identical to the way they always were filled out. Next to "cost of airfare" they would put in $235 . The receipt was clearly included which used to say that it was charged to a credit card, and after the change the receipt said that it was prepaid by travel agency. The reimbursement system should have been smart enough to know when to reimburse since it was all in the computer.

The investigators response was, naturally, that once he realized that he was being reimbursed, he should have alerted the authorities. His only defense (a legitimate one) was when you fly 50-80 times a year, you don't even look anymore.

I've been fascinated by air disasters for years, and I read and watch everything about them that I can. There are two things that keep coming up:

1) So often the culture in the cockpit contributes to a disaster. A copilot or flight engineer sees something, but doesn't feel it is his place to say anything. Whereas if he would have acted, everything would have been fine. This has actually become part of pilots' instruction, that they create a culture where anyone who sees something can speak up. Just google "cockpit culture".

2) What I call a cascade of failure: one thing goes wrong, that causes another thing to go wrong, that causes someone to make an assumption, which makes everything go to hell.


I highly recommend the mother of all airline books: Fate is the Hunter, by Ernest K Gann. The final chapter is riveting.

Quote: Mosca

1) So often the culture in the cockpit contributes to a disaster.

That was the cause of the disaster at Tenerife in the 70s.

Quote:

I highly recommend the mother of all airline books: Fate is the Hunter, by Ernest K Gann. The final chapter is riveting.

Thanks! I was going to ask for recommendations. Speaking of which, do you have any other recommendations?

Come to think of it, a chain of events also contributed to the disaster at Tenerife. The bomb threat which closed the other airport, and the fog.

Quote: Mosca

What I call a cascade of failure: one thing goes wrong, that causes another thing to go wrong, that causes someone to make an assumption, which makes everything go to hell.

Most of the time single point failures are easy to predict and safeguards cover them. Almost all catastrophe's are caused by multi point failures.

AFAIK, nobody bothers to build a computer prediction of remaining mileage on your gasoline reserve in a car. It is sufficient to know what the current level is, because 99.9% of the time you simply pull into a gas station to refuel.

Obviously such a predictive tool is more important in an airplane so that you have plenty of time to set up alternative options if you might run out of fuel (including getting a higher priority for landing in backups). But it seemed as if having the fuel gages was not considered a problem worth cancelling a flight. The predictive instrumentation was sufficient. But nobody anticipated somebody manually entering the wrong amount of initial fuel on takeoff.

Quote: Pacomartin

AFAIK, nobody bothers to build a computer prediction of remaining mileage on your gasoline reserve in a car. It is sufficient to know what the current level is, because 99.9% of the time you simply pull into a gas station to refuel.

My 2011 Corolla displays and estimate of how much distance I can cover with the current amount of fuel. It changes over time, naturally, as fuel is consumed, and/if fuel economy suffers major changes (ie stuck in traffic, the estimate goes down; making good time on the highway, it goes up).

That's not a precise measure of fuel remaining, but it's the next best thing. If you see a sign that the next gas station is 50 km away, and the car's display says you can go another 40 km, then you're likely in trouble.

Quote:

Obviously such a predictive tool is more important in an airplane so that you have plenty of time to set up alternative options if you might run out of fuel (including getting a higher priority for landing in backups).

Oh, it's absolutely essential. Flight plans include alternate airports. You have to know how far you can fly before diverting to another airport is no longer possible. That was a factor in the Avianca crash in NYC.

Quote:

But it seemed as if having the fuel gages was not considered a problem worth cancelling a flight. The predictive instrumentation was sufficient. But nobody anticipated somebody manually entering the wrong amount of initial fuel on takeoff.

Some reading I did indicates it should have been reason to ground the plane. There is a Minimum Equipment List for every type of airliner. It states a long list of equipment you can't fly without. But the 767 was a new plane and the list changed fairly frequently at first.

Suppose they had taken enough fuel, but a fuel leak had developed in flight. Depending on the size of the leak, they might not have made it as far as Gimli. Worse, they might have passed it and Winnipeg, and run out of options.

Fuel leaks are rare, yes, but they do happen. There was an Adam Air which had to glide to a landing at the Azores due to a leak. That was a great deal of luck, not running out of fuel in the middle of the Atlantic.

Quote: Nareed

I don't think any civilian airliner ever takes off with full tanks, other than on manufacturer tests. If then.

Military planes get full tanks, and strap-on additional tanks, almost all the time. Many even refuel in mid-air shortly after take off to top off. that's how come they can fly such long distances.



I know of exactly two crashes involving running out of fuel. The Air Canada I mentioned, and one where a Colombian airliner crashed after much delay near landing. I don't know if there have been others.

Great thread.

A very famous out-of-fuel crash, one which ultimately changed the industry, was UAL173 near Portland, OR in 1978. The Captain was working a landing gear issue, having requested and received holding to do so, the FO and FE were too intimidated by the Captain and company procedure to tell him directly they were getting low on fuel, and they ultimately ran out at low altitudes, crashed on 7 mile final to PDX on/near Burnside Avenue, with many fatalities. It's worth looking up the details via wiki, or even better, on the NTSB database, and reading not just an overview; the CVR transcript is particularly telling in hindsight. This accident did 2 things; strengthened the practice of honest examination of crash causes for the greater good, and forced the change in the cockpit from Captain as dictator to Crew Resource Management (CRM) which empowered the entire team as contributors without retribution from command authority.

I wrote a master's analysis on Fate is the Hunter during my degree course - I really like that y'all recommended it as good reading. It's also a good movie.

There has been a huge industry movement towards "break the chain" since the 70's, because most of the mechanical and other major issues have been identified and either built out of the industry, standardized a work-around, or an improvement has been made. Almost all issues now are either small multiple events chaining into something significant, and/or human error. So safeguards and redundancies are built into routines that seem like time and money-wasters, but are collectively more desirable than the 1% or whatever miniscule amount of times they prevent the elevation into major issues they were designed to stop.

One of the most interesting and useful publications is NASA Callback, published ~ monthly (and free, either in hard copy or electronic by subscription), which is almost entirely anecdotal. It is a forgiveness mechanism, where no regulatory punitive action is allowed, in return for a completely honest report of personal human error in flights. The database is compiled and looks for patterns of error, whether in design flaws, procedures, training, phraseology, whatever the reporter experienced or observed, along with their best guess of why. The NTSB, FAA, and NASA collaborate in developing recommendations and changes from those reports. It's been in publication for about 25 years, and back issues are available on the website. Its most frequent contributors are airline pilots, but pilots of all types, controllers, and other aviation professionals often contribute as well. http://asrs.arc.nasa.gov/publications/callback.html

The NTSB Reporter is another outstanding publication, which reviews accidents and incidents for cause on a monthly basis. This is a paid subscription service, at $24/year, which sorts through the database as reports become available and provides synopses of causes and recommendations, both of individual incidents (the majority of content) and of identified trends, general recommendations, and updates on media-heavy incidents. Since they're essentially a culling service, though I was a subscriber for many years, I suggest you might simply look at the NTSB database itself if you enjoy digging out information without paying.

http://www.ntsbreporter.us/index.htm

The NTSB database itself has things listed in such a way that you can look up anything they reported on from their inception, or simply browse by month or year and read more (from synopsis to details to supporting evidence to conclusions) as you find something that interests you. You can also investigate by aircraft type, airline, type of problem, fatal or not, or many other search parameters. The NTSB often gets invited even to foreign investigations, especially if an American manufacturer is involved, so many international accidents are included in their database.

http://ntsb.gov/_layouts/ntsb.aviation/index.aspx

The majority of accidents in the US are not actually NTSB investigations; they're done by FSDO arm of the FAA. That's because the NTSB does commercial aviation and high-media private, and the majority of accidents involve General Aviation aircraft; you can read about causes and conclusions on those (also a searchable database) at the following link. The two databases are not mutually exclusive; the FAA also covers most, if not all, of the domestic incidents investigated by the NTSB.

http://www.asias.faa.gov/pls/apex/f?p=100:1:

The first definitive textbook I know about was the one I used in my Human Factors class (Master's study level). It's title is Human Factors in Aviation, authors Wiener/Nagel, published in 1988. Still considered the gold standard in the field. I listed the Amazon link below, which also contains some other recommendations by them (and later editions of this book by other editors), but all are pretty expensive. If one of the regulars wants to borrow my copy instead, I'm willing to loan it to them for the cost of postage. (PM me at WoV and we'll set it up; I don't get over here as often as I like).

http://www.amazon.com/Human-Factors-Aviation-Cognition-Perception/dp/0127500308

Another book I don't personally think was presented impartially, but contains some good information, is Flying Blind, Flying Safe by Mary Schiavo, c1997. She made some biased and self-serving conclusions in the book, but it's a pretty interesting snapshot of the last of the era of major accidents, their causes, and trend identification, after some major steps had been taken but before their effects could register (in what is now about 15 years of shocking statistical improvement over the years Ms. Schiavo reviews). Again, I'll loan it for postage if someone wants to read it. However, you'll note on the Amazon link that they can't even GIVE it away, so chances are it's cheaper for you to order it if you want it. (lol...sorry for the disdain, but she earned it from us)

http://www.amazon.com/Flying-Blind-Safe-Mary-Schiavo/dp/0380975327/ref=sr_1_1?s=books&ie=UTF8&qid=1439834013&sr=1-1&keywords=schiavo+Flying

There are also a series of books called, "I learned about Flying from That", which are compilations of columns from Flying Magazine over the last 40 years or so from GA pilots self-reporting a bonehead move they made and its consequences. Again anecdotal, still some great stories among them. And the column itself, which is likely archived somewhere on the Internet, so probably 500 or more stories of that sort available, possibly many more.

Hope some of this is useful. Good thread!