What would you tell the head of NASA?

[sevenstrokeroll]

Many of you may have read about the airline safety information report somewhat squelched by NASA...and then NASA released the data along with interviews with the head of NASA sort of putting it all down.

WHAT would you tell to the head of NASA about the state of flying in America...and this is from the cockpit, mx, fa's,and dispatchers. Also ATC.

I think I would just smack him in the nose, but I've thought better of that.

[maui]

Haven't seen or read the report you refer to. However to include references to NASA and safety in the same sentence is a bit of a stretch.

Maui

[Loose rivets]

I think I would just smack him in the nose,

Oh, don't waste time with diplomacy, just tell him what you think

[PBL]

Folks,

here a view from the other side. My colleagues at NASA include some of the best safety-critical digital systems people and human factors people in the world, and they put out seminal analyses; for example the report on HIRF in the TWA 800 docket; the short paper on evidential reasoning about rare events that is one of the two bases of that subject (which is finding itself in a bit of a crisis at the moment, since far too few engineers know of that work despite that it should be in their university coursework), work on ultrareliable distributed systems for electronic control, the dynamically-reconfigurable fault-tolerant flight control system which operates through normal stick-and-rudder control even when your aerodynamic control is all but gone, and of course they pioneered anonymous reporting systems, which are now mimicked all over the world.

But aviation at NASA is largely organised on a contract basis. They got the HIRF contract, for example, because a Harvard english professor speculated about a military accident with HIRF potentially causing the TWA 800 crash in a literary-political journal, the NYRB, and got into a letters exchange with the head of the NTSB. Smart move by the NTSB: I don't know what the story is but I guess money somehow appeared (publicity in a chattering-classes organ can have that effect sometimes) that they could give to NASA to work on it and the NASA people seem to have leapt at the chance; the report makes great reading.

To do more good work, therefore, NASA needs contracts. From outside: NTSB (which has no money); FAA (which has got to figure out whether to give it to NASA or to use it to train a few more air traffic controllers to relieve the crunch); DoD (which is currently using it to pay for a couple of wars); aerospace manufacturers (who usually prefer to develop and keep their skills in-house) or the U.S. Government (Congress, often led by the Executive Branch). And the Executive Branch has told Professor Griffin that his job is to go to Mars, and cut his resources at the same time.

I have no idea where Professor Griffin's sympathies lie. I think anybody in his position has hisher hands so tied that it may not matter.

I think you will find that if money comes from somewhere there are world-class people there who will be leaping over themselves to get to work and it won't be wasted.

In other words, the problem lies elsewhere. Nobody else cares enough about aviation to provide them the resources. That is a pretty odd state of affairs in a country whose commerce is absolutely dependent on it as their primary mode of transportation between population and industrial centers.

PBL

[Intruder]

This morning I just read the first news article on the [partial] release of the study. IMO, the report didn't say what NASA and FAA politicos wanted it to say, so they tried to squelch it. Now the head NASA politico is trying to convince us that the study was flawed in its methodology, so it will inherently cause an inflation of reported incident rates.

The solution, IMO, is simply to release the full report so other independent analysts who want to do so can validate the methodology or show where it is flawed. To keep the full results away from "peer review" will only tend to confirm our suspicions that the politicos have something to hide...

[sevenstrokeroll]

Intruder...you echo my thoughts

[10002level]

I don't know anything about the report, but I would tell him/her that flying isn't rocket science.

[PBL]

ssr,

You say you agree with Intruder. But you said first you would preferably smack Griffin in the nose, and Intruder said he thought that the details should be made public in order to allow independent (pro bono) analysis.

I think the difference is that most senior safety-critical system specialists would refrain from the former and greet the latter suggestion enthusiastically.

Have I read you right or wrong?

PBL

[sevenstrokeroll]

you must understand, flying is serious enough, the complaints are real enough, that if we didn't approach it with a bit of humor, we might just cry.

lost in translation...

[blakkekatte]

The NASA Report

For those who are interested, the report is here:

“NAOMS Reference Report: Concepts, Methods, and Development Roadmap”

PS: NAOMS = National Aviation Operational Monitoring Service

[411A]

Now the head NASA politico is trying to convince us that the study was flawed in its methodology, so it will inherently cause an inflation of reported incident rates.

I strongly suspect it was flawed.
Sometimes crews have a knee-jerk reaction to certain events, and upon further reflection, might tell a different story.

I suspect the concerned NASA study is not worth the paper it's printed on....

[Safety4All]

Maybe it is better to not judge the study by its cover. It is not too difficult to get some meaningful numbers from the data that can be found here:
http://www.nasa.gov/news/reports/NAOMS.html

The most interesting part (for me) are the "Redacted Air Carrier Survey Responses" which list safety-related events during some 2.5 million flight hours.

Using that number the following can be deducted about many safety-related events, for example (always in the period April 2001 till December 2004):

1. On average an air carrier pilot had to perform an evasive action about once every 700 flight hours to avoid an imminent in-flight collision with another aircraft that was never closer than 500 feet including evasive action in response to a TCAS advisory.
2. The same pilot experienced less than 500 feet of separation from another aircraft while both aircraft were airborne about once every 1600 flight hours.
3. Our pilot was unable to communicate with A.T.C. in a time-critical situation because of frequency congestion about once every 110 flight hours (half of which while airborne in the terminal area).

And the list goes on. The data was obtained from what pilots were able to recall from memory during a 30-minute telephone interview about safety-related events they experienced during the most recent 90 days before the interview.

In other words, the real numbers today are probably even more disturbing...

[Intruder]

It could well be flawed, even beyond the point where it is useful if imperfect (as most valid studies are). The problem is, we won't KNOW how flawed it is without the data and the methodology.

Right now, all we know is that some NASA politico has made some public conclusions that we cannot evaluate for validity. Maybe NASA has not reformed it's "old boy" network in the past decades after all...

[TWN PPL]

Pilot complaints

http://www.cnn.com/2007/TECH/space/1...ety/index.html