Re: NASA Space Shuttle - The Real Problem
- From: Andy Resnick <andy.resnick@xxxxxxxxxxx>
- Date: Wed, 03 Aug 2005 09:14:15 -0400
jmfbahciv@xxxxxxx wrote:
<snip>In article <dco8qj$rtg$1@xxxxxxxxxxxxxxxxxxx>, Andy Resnick <andy.resnick@xxxxxxxxxxx> wrote:
Maybe, maybe not. Look at military vehicles- that's a more fair comparison to the shuttle.
Not really. There are only a couple of shuttles and thousands of military vehicles. Thus, warehousing replacement parts for the vehicles would make sense. Or have I messed up again?
I think the comparison makes sense, not because of the number of vehicles- 7 shuttles versus a fleet of F-15s, but because of the timescales and econoscales involved: the project lifetime, number of people involved, and dollars involved. That is, compare the shuttle program with, say, the Abrams tank program. And both of those are dwarfed by navy programs like the Ticonderoga class destroyer program or the Ohio class nuclear sub program.
.. What happens is that over time, replacement parts are more and more difficult to purchase; that's why the government purchases (for example) computer chips in huge volumes. Things like pistons, valves, hoses, cables, fittings, screws, etc. are constructed to standard sizes, pin layouts, and threads; thus, replacements can always be found. Or at least manufactured fairly easily.
One of the problems with warehousing a gazillion foobars is that the manufacturer will move on (nothing to sell) and the knowledge of how to make the foobar will disappear. I never have figured out how to solve this problem; I've spent some thinking time on it.
It's hard, I agree. But, it seems to be limited to computer chips- even things like resistors and capacitors are reasonably stable. The EEs I worked with had these big giant encyclopedia things of pin-out diagrams of every logic chip ever made. They seemed to be excited about these new programmable logic arrays- I guess it makes the hardware more flexable.
<snip>
Right- and that's the problem. Let's say we have an experiment that's going up in December 2006. At least that's when it's expected to launch. We'll come back to that. A launch of December 2006 means the hardware has to be delivered to Kennedy Space Center by June 2006. Which means that the completed part needs to be done for final testing around January 2006. Which means construction has to begin around June 2005. That means the prototype initated tests around June 2004, so the tests can be run and evaluated in time to start construction of the flight unit. Now, if the prototype has to be done by January 2004, that means any breadboard construction had to be done around January 2002, to allow testing and redesign work for the brassboard and protoype.
What's a brassboard? I don't think I've heard of that word.
Oh... sorry. I picked it up at NASA, and it's actually an intelligent to 'do design'. First is the breadboard. That means 'slap a bunch of whatever together any way possible and get something that works'. Then, take a step back and approach the mess rationally: if some custom parts need to be fabricated, or if there are other, better components out there that can be ordered, etc. Then, re-assemble using the new parts. That's the brassboard. Third, clean it up and pretend you are making a production piece: detailed assembly drawings, instructions, etc etc. Do the packaging. Quantify the performance. That's the prototype. Then the actual part gets made.
<snip>
Wow. Your project lifetime is a lot longer than one of our
major OS projects were. I suppose your schedules are similar
to our hardware product lifetimes.
There is a software version of the above, I'm not sure; CMM or something like that? It doesn't have a lot of respect, tho.
Now, if the experiment is due to launch December 2006, it's nearly a sure bet that it won't actually launch until sometime 2009, assuming the shuttle fleet sticks to the schedule and doesn't drop/stop flights. So the high-end multi-million dollar experiment which begins taking data in 2010, to be analyzed in mid to late 2011, was built using 11 year old technology.
And that's just a single measly experiment. Scale upwards appropriately for durable equipment designed to operate for 20 years.
This is an unfair question so you can just tell me to shutup :-)
if you don't want to deal with it. What is your biggest bottleneck? To help you answer in the spirit I'm asking, it
seems like the biz is similar to the old days when people
waited for their IBM card deck to be read in and processed.
I forgot to add that the *science* being done is also 11 years old- that's why so little cutting-edge research is done in space.
But, to answer your question, the biggest bottleneck NASA faces is access to orbit. The satellite and unmanned missions seem to be more hip and with-it because there is much easier access: regular launches, which allows for faster project development.
<snip>
.. One way favors the science and engineering, one favors the accounting. Maybe NASA should be treated the same as any other federal agency, maybe not.
Not knowing anything, my immediate reaction is no :-).
.. Maybe it should be militarized. I can't say, and it's not up to me.
I don't see the advantage of militarizing it. They seem to create as many bells and whistles as anybody.
It can be argued either way. The facts on the ground are that key NASA agency posts (high level stuff) are being filled by Admirals and the like. At least they are used to handling multi-billion dollar programs.
--
Andrew Resnick, Ph.D.
Department of Physiology and Biophysics
Case Western Reserve University
.
- References:
- NASA Space Shuttle - The Real Problem
- From: Morpheal
- Re: NASA Space Shuttle - The Real Problem
- From: Andy Resnick
- Re: NASA Space Shuttle - The Real Problem
- From: Andy Resnick
- NASA Space Shuttle - The Real Problem
- Prev by Date: Re: PLASMA JETS ON EARTH
- Next by Date: Re: T.J.'s Something_is 18 cents Wake is moron
- Previous by thread: Re: NASA Space Shuttle - The Real Problem
- Next by thread: Teleportation
- Index(es):
Relevant Pages
|
