Living Down To Expectations

I have another column this week on why space stuff costs so
much--because we expect it to:

http://www.foxnews.com/story/0,2933,98228,00.html

--
simberg.interglobal.org * 310 372-7963 (CA) 307 739-1296 (Jackson Hole)
interglobal space lines * 307 733-1715 (Fax) http://www.interglobal.org

"Extraordinary launch vehicles require extraordinary markets..."
Swap the first . and @ and throw out the ".trash" to email me.
Here's my email address for autospammers:

On Thu, 25 Sep 2003 18:24:42 CST, in a place far, far away, rk
made the phosphor on my monitor glow
in such a way as to indicate that:

Rand Simberg wrote:

I have another column this week on why space stuff costs so
much--because we expect it to:

http://www.foxnews.com/story/0,2933,98228,00.html

Interesting, as always.

First, with regards to complexity, are you familiar with David
Bearden's work at The Aerospace Corporation? He is modelling
complexity, cost, and schedule, and using them to understand success
and failure.

No, I'm not. Is there anything available on line?

Secondly, with respects to the assumptions in the three cost
estimates used, did they compare the internal cost estimates with
those for a research and demonstration job organized by, say, DARPA?
The assumptions over what is to be done and delivered can
drastically effect the final cost (e.g., how much paperwork, how
many reviews, etc.). Not knowing the details in the cost estimates
and the assumptions it's hard to understand. For instance, if the
Air Force and NASA models were for a deliverable vehicle to be used
in operations vs. a simple demonstration you might get quite
different answers. For examples, you mentioned that the NASA
numbers were based on Shuttle experience. Doing work on that
program would be far different than say a one-shot demonstration
job. So, perhaps you can expand on why the numbers are different.
For an analogy -- one always needs analogies in Usenet -- note that
the cost for writing the code for a software job is just a small
part of the overall cost, with maintenance, documentation, etc.,
making up a large part of the pie. I'm sure Kevin would have the
latest breakdown on things like that.

I don't know the answers to any of those questions--you'd have to ask
Steve Hoeser.

--
simberg.interglobal.org * 310 372-7963 (CA) 307 739-1296 (Jackson Hole)
interglobal space lines * 307 733-1715 (Fax) http://www.interglobal.org

"Extraordinary launch vehicles require extraordinary markets..."
Swap the first . and @ and throw out the ".trash" to email me.
Here's my email address for autospammers:

In article ,
says...
For an analogy -- one always needs analogies in Usenet -- note that
the cost for writing the code for a software job is just a small
part of the overall cost, with maintenance, documentation, etc.,
making up a large part of the pie. I'm sure Kevin would have the
latest breakdown on things like that.

Despite the claims of the XP'ers, I haven't seen anything that suggests
the latest breakdowns are any different from the breakdowns that have
been used over the last few decades. (That's a real disappointment.
You'd think we'd be doing a better job of building software as we learn
new things. We aren't. We are building ever more complex software, but
there is no evidence that it is of higher quality. The true high-quality
creations (e.g., the Shuttle flight software) are notable exceptions,
not the general rule.)

The standard rule of thumb is that the QA phase takes 1/2 the cost. The
actual coding takes 1/6th of the cost. Maintenance is two to ten times
the development. So the actual coding is at most an order of magnitude
less than the total development budget.

(These are, of course, broad averages. Individual products may have very
different budgets. I'm ignoring the non-R&D costs (e.g., marketing,
advertising, customer education, legal), yet these often dominate the
R&D costs.)
--
Kevin Willoughby lid

Imagine that, a FROG ON-OFF switch, hardly the work
for test pilots. -- Mike Collins

I have another column this week on why space stuff costs so
much--because we expect it to:

http://www.foxnews.com/story/0,2933,98228,00.html

That's good. The story of the misplaced decimal point is particularly
amusing.

One thing you might consider for a future column is to highlight
possible solutions to the problem. Fixed price contracts might seem
to be one (and have been used, e.g. in the TDRS procurement from the
90's), but the limiting factor here is that the government still will
only pay costs plus some fee, rather than letting the contractor
innovate and keep the savings (as long as they come in ahead of the
other bidders). Dan Goldin tried tilting at this windmill, but didn't
get anywhere. Don't know if anyone could, given the political
problems of wanting it to appear that no money is wasted and the
difficulty of distinguishing a sham competitive procurement from one
where there is really strong competition.

The concept which is probably doing the best job in this area is
Discovery and the other programs modeled after it (ESSP, Mars Scout).
Here you clearly have healthy competition. Here costs are a factor in
selecting missions. And the success of this kind of procurement
(among other factors) has led to a renaissance across most of the
uncrewed side of NASA.

In article , Jim Kingdon wrote:

The concept which is probably doing the best job in this area is
Discovery and the other programs modeled after it (ESSP, Mars Scout).
Here you clearly have healthy competition. Here costs are a factor in
selecting missions. And the success of this kind of procurement
(among other factors) has led to a renaissance across most of the
uncrewed side of NASA.

The fact that there's been a swift kick up the proverbial after two
high-profile and embarrassing failures can't hurt, of course :-)

Here's a question about Discovery & such - I know there's a solid cost
cap (~$350m/project?); does this involve the booster, or is it merely
that when I see a cost breakdown they've already earmarked $x for the
launch? In other words, does the Discovery mission pay for the Delta II
(or similar), or does it come from a different budget? (On a tangential
note, would they be permitted to use non-US launch vehicles?)

Just curious, as always, I do agree that it seems one of their more
successful - or sensible - ideas.

--
-Andrew Gray

Here's a question about Discovery & such - I know there's a solid cost
cap (~$350m/project?); does this involve the booster, or is it merely
that when I see a cost breakdown they've already earmarked $x for the
launch? In other words, does the Discovery mission pay for the Delta II

For a real answer, you'd need to look through the AO (Announcement of
Opportunity) linked from http://discovery.larc.nasa.gov/discovery/ .
The AO's are relatively readable, but there is a lot of text to look
through.

My vague recollection is that the launch comes from a different budget
(provided you want to launch on Delta II or maybe one or two others;
if you want to launch on something random you need to pay). But
that's a dim recollection from the first Discovery AO or two; it could
have changed. One thing I wouldn't expect to change is that I
wouldn't expect the Discovery proposers to want to go through the
trouble of procuring a launcher.

(or similar), or does it come from a different budget? (On a tangential
note, would they be permitted to use non-US launch vehicles?)

I'm pretty sure you couldn't spend NASA money on a non-US launcher.
If there were some non-NASA money involved, maybe (there are rules for
that too, and of course the big risk is what if one of the two funders
cancels their funding and the other wants to proceed).

In article , Jim Kingdon wrote:
Here's a question about Discovery & such - I know there's a solid cost
cap (~$350m/project?); does this involve the booster, or is it merely
that when I see a cost breakdown they've already earmarked $x for the
launch? In other words, does the Discovery mission pay for the Delta II

For a real answer, you'd need to look through the AO (Announcement of
Opportunity) linked from http://discovery.larc.nasa.gov/discovery/ .
The AO's are relatively readable, but there is a lot of text to look
through.

Thanks!

My vague recollection is that the launch comes from a different budget
(provided you want to launch on Delta II or maybe one or two others;
if you want to launch on something random you need to pay). But
that's a dim recollection from the first Discovery AO or two; it could
have changed. One thing I wouldn't expect to change is that I
wouldn't expect the Discovery proposers to want to go through the
trouble of procuring a launcher.


From:

http://discovery.nasa.gov/program.html

"Discovery seeks to keep performance high and expenses low by using new
technologies and strict cost caps. The cost for the entire mission
(design, development, launch vehicle, instruments, spacecraft, launch,
mission operations, and data analysis) must be less than $299 million.
The development time from mission start to launch can be no more than 36
months. The intent is to have a mission launch every 12 to 24 months."

So, it looks like your $300m doesn't include the launch; a Delta II is,
what, $60m? Hmm. Hmm hmm.

(Wandering on the site, I notice the Dawn design review was pencilled in
for mid-2003; their website hasn't been updated since September. Any
idea what's happening with it?)

There's a brief section on the Discovery program to date; ten missions
and one "piggyback" (on Mars Express), of which one has failed, one's
been cancelled, two have finished successfully and two more are in
progress. The others are under development, or in one case en route...
pretty good so far, that.

I'm pretty sure you couldn't spend NASA money on a non-US launcher.
If there were some non-NASA money involved, maybe (there are rules for
that too, and of course the big risk is what if one of the two funders
cancels their funding and the other wants to proceed).

Yeah, I was idly wondering about the status of "barter deals" - one
partner provides the spacecraft, one the LV, and the second partner gets
to fly instruments on the s/c...

--
-Andrew Gray

http://discovery.nasa.gov/program.html writes (emphasis added)

"Discovery seeks to keep performance high and expenses low by using new
technologies and strict cost caps. The cost for the entire mission
(design, development, ***launch vehicle***, instruments, spacecraft, launch,
mission operations, and data analysis) must be less than $299 million.

Andrew Gray writes:

So, it looks like your $300m doesn't include the launch; a Delta II is,
what, $60m? Hmm. Hmm hmm.

Looks like $299 million does include the launch (see emphasized words
above). I don't remember whether the cost is computed the same way
for the cap (which this describes) as for the scoring of proposals,
though.

(Wandering on the site, I notice the Dawn design review was pencilled in
for mid-2003; their website hasn't been updated since September. Any
idea what's happening with it?)

According to
http://www-ssc.igpp.ucla.edu/dawn/ne...html/20030822/
they had PDR for 4 of 5 instruments already (as of August), with the
fifth planned for September. As for the spacecraft and such, they
are having some trouble with financial reserves and adding a fifth
solar array and such. Among other things, this may change the timing
of when they spend their money. Hopefully this is just a little blip
and not the start of the downward spiral of rescope, spend money on
redesign, rescope to deal with that money just spent, etc. I'm
assuming they need to resolve this before PDR.

There's a brief section on the Discovery program to date; ten missions
and one "piggyback" (on Mars Express), of which one has failed, one's
been cancelled, two have finished successfully and two more are in
progress. The others are under development, or in one case en route...
pretty good so far, that.

Two were piggybacks (Missions of Opportunity, in Discovery-speak).
NetLander, the cancelled mission, was to be part of a French mission.
It was apparently cancelled to save money, apparently by the French,
although I guess NASA first made the announcement (concerning their
part of it).

NetLander**
NASA's decision in March 2003 to cancel its participation negatively
impacted NetLander, primarily a European mission. (Tillman began as,
and still is, a member of the Finnish NetLander team.) Our European
NetLander status is now best described by "The [European] Netlander
Steering Committee will have a meeting this month to discuss the
measures to be taken now that the old mission scenario has
vanished. The goal is to revive the objectives of the NetLander by new
type of mission arrangements with ESA and/or Russian space agencies."
http://www.atmos.washington.edu/~mars/

The CNES planned to spend 100 million on the project, which also
involved Finland, Germany, Belgium and Switzerland. In March, NASA
shelved plans to contribute $35 million to the project.
"France purges space programme in bid to survive budget crisis",
Nature, 8 May 2003, page 103.

Yeah, I was idly wondering about the status of "barter deals" - one
partner provides the spacecraft, one the LV, and the second partner gets
to fly instruments on the s/c...

There are separate criteria. Either you are a regular Discovery
mission (in which case you can get up to 33% or whatever it is from
non-US funding), or you are a Mission of Opportunity, which is
basically the reverse (some other nation is the lead, and you are
providing an instrument or something). Again, the AO goes into details.

In article , Jim Kingdon wrote:
http://discovery.nasa.gov/program.html writes (emphasis added)

"Discovery seeks to keep performance high and expenses low by using new
technologies and strict cost caps. The cost for the entire mission
(design, development, ***launch vehicle***, instruments, spacecraft, launch,
mission operations, and data analysis) must be less than $299 million.

Andrew Gray writes:

So, it looks like your $300m doesn't include the launch; a Delta II is,
what, $60m? Hmm. Hmm hmm.

Looks like $299 million does include the launch (see emphasized words
above). I don't remember whether the cost is computed the same way
for the cap (which this describes) as for the scoring of proposals,
though.

Um, that's what I intended to say. Er. Perhaps if you parse it as "you
are given $300m but not given a launch" what I said might make sense...
but it does look like it reads "you are given $300m, which does not
include the money for the launch". Mea culpa. ;-)

(It's a vague phrase, now I think about it; either way of saying it can
be read either way...)

According to
http://www-ssc.igpp.ucla.edu/dawn/ne...html/20030822/
they had PDR for 4 of 5 instruments already (as of August), with the
fifth planned for September. As for the spacecraft and such, they
are having some trouble with financial reserves and adding a fifth
solar array and such. Among other things, this may change the timing
of when they spend their money. Hopefully this is just a little blip
and not the start of the downward spiral of rescope, spend money on
redesign, rescope to deal with that money just spent, etc. I'm
assuming they need to resolve this before PDR.

Thanks... I'll keep an eye out for anything further. It's certainly a
worthwhile mission, and it'll be very interesting to compare its results
to that of NEAR.

Yeah, I was idly wondering about the status of "barter deals" - one
partner provides the spacecraft, one the LV, and the second partner gets
to fly instruments on the s/c...

There are separate criteria. Either you are a regular Discovery
mission (in which case you can get up to 33% or whatever it is from
non-US funding), or you are a Mission of Opportunity, which is
basically the reverse (some other nation is the lead, and you are
providing an instrument or something). Again, the AO goes into details.

Yeah, it's in the list of things to read in depth (when I find time...).
Thanks again!

--
-Andrew Gray

Jim Kingdon wrote:

...they had PDR for 4 of 5 instruments already (as of August), with the
fifth planned for September. As for the spacecraft and such, they
are having some trouble with financial reserves and adding a fifth
solar array and such. Among other things, this may change the timing
of when they spend their money. Hopefully this is just a little blip
and not the start of the downward spiral of rescope, spend money on
redesign, rescope to deal with that money just spent, etc. I'm
assuming they need to resolve this before PDR.

Not all of this, I suspect, is entirely unique to the Dawn mission. In
fact, a few months ago it was realized that another Discovery Program
mission, MESSENGER, is going to exceed its cost caps by, I believe,
about ~$12 million. For that reason, and others, I hear that NASA is
going to stipulate that proposers for future competitively selected
missions, at least those in the Discovery, Mars Scout, and New Frontiers
arenas, retain a 25% funding reserve through Phase B.

As for MESSENGER, from what I understand, NASA is going to issue of
waiver of the cost cap in this particular case due to the uniqueness and
unforeseen nature of the problem (i.e., delamination of circuit boards),
and because the MESSENGER Project appears to have the matter well under
control. Also, a science review panel advised NASA to make an exception
*in this case* because of the compelling nature of the science. The cost
overrun will, however, require more testing with a resultant slip of the
MESSENGER launch date from its primary period in March 2004 to, at
least, the backup launch period in May 2004, or possibly later, in the
July-August 2004 period.

--


Alex R. Blackwell
University of Hawaii