[conspire] 737 MAX story keeps getting more fractally bad

[Tony Godshall]

New York Times audio:
https://dts.podtrac.com/redirect.mp3/rss.art19.com/episodes/b01d6007-c8be-4d9b-8650-b9fa930d4802.mp3

Quote:

In testimony before a House committee on Wednesday, Dennis A. Muilenburg,
Boeing’s chief executive, said, “If we knew everything back then that we
know now, we would have made a different decision.” Congress is
investigating two crashes of Boeing 737 Max jets which killed 346 people,
cost the company billions of dollars and raised new questions about
government oversight of aviation. So what did Boeing executives know about
the dangers of the automated system implicated in the crashes — and when

* Published: 10/31/19 2:54:29 AM

* Episode feed: The Daily - http://rss.art19.com/the-daily



On Tue, Jul 9, 2019, 8:42 AM paulz at ieee.org <paulz at ieee.org> wrote:

> The saga at Boeing is portrays a very different mindset from my experience
> with makers of medical devices.
>
> Every one is very aware that patients lives are effected for bettor or
> worse by the company product. If that isn’t enough, bad outcomes result in
> lots of lawyers. I’m sure you have seen ads on television, “If you or a
> family member was injured...”
>
> First every significant step in the development from original requirements
> to production testing must be carefully documented. Makes it very easy to
> go back and figure out what happened. If the company does not have a good
> documentation system, FDA will not approve the product. The whole design
> process will need to be repeated with better documentation.
>
> Another important part of the process is FMEA, Failure Modes and Effects
> Analysis. Lots of brainstorming to try to list every possible thing that
> could go wrong, the consequences and how severe it is to the patient. Items
> with any small possibility of serious problems to the patient, the doctor
> or the equipment must be analyzed and ways found to mitigate it.
>
> To use a hypothetical example from the airline industry. Some planes have
> a thing called angle of attack sensor. If the sensor just gives an
> indication to the pilot, and the plane has lots of other instruments,
> perhaps pilot training can tell him when to ignore AoA and how to use other
> information to safely fly the plane. And the documentation system will
> track the training requirement to the actual training.
>
> Just speaking, not that this would happen, but suppose the AoA sensor was
> connected to an automated flight control system and a bad sensor input
> could possibly, under remote circumstances, cause the plane to crash. This
> is not acceptable.
>
> Mediation would require multiple sensors, using different technologies.
> There might be an undiscovered weakness in one kind of sensor. Software
> will have a requirement that a bad sensor input will not cause the plane to
> crash. Maybe the requirement includes checking the altitude before pointing
> to the ground. Again the documentation system will track this from software
> requirements to final validation.
>
> _______________________________________________
> conspire mailing list
> conspire at linuxmafia.com
> http://linuxmafia.com/mailman/listinfo/conspire
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://linuxmafia.com/pipermail/conspire/attachments/20191031/b1a27167/attachment.html>