Author Topic: JWST: Albatross of SMD now $8.8 billion  (Read 130801 times)


Offline Blackstar

  • Veteran
  • Senior Member
  • *****
  • Posts: 11646
  • Liked: 3204
  • Likes Given: 1
Re: JWST: Albatross of SMD now $8.8 billion
« Reply #381 on: 10/26/2013 12:15 am »
I know someone who dressed up as Hubble once. It was much easier. Just need a big tube and you can make some solar panels for your arms.

Online JohnFornaro

  • Not an expert
  • Senior Member
  • *****
  • Posts: 9235
  • Delta-t is the salient metric.
  • Planet Eaarth
    • Design / Program Associates
  • Liked: 657
  • Likes Given: 360
Sometimes I just flat out don't get it.

Online JohnFornaro

  • Not an expert
  • Senior Member
  • *****
  • Posts: 9235
  • Delta-t is the salient metric.
  • Planet Eaarth
    • Design / Program Associates
  • Liked: 657
  • Likes Given: 360
Re: JWST: Albatross of SMD now $8.8 billion
« Reply #383 on: 10/30/2013 12:49 pm »
So the annual budget has already peaked?
No.  The budget in the last post was not adequate to meet the 2014 launch date so it was moved to 2018, and new budget put forward in 2013. So to readjust for the new launch date,

found the actual data for the FY13 JWST budget from 2013 to 2018 Launch
Per the link:
Quote from: jwstbudget

"As a result of the rebaseline, the launch date moved from 2014 to 2018, and the development cost
increased from $2.581 to $6.198 billion. The FY 2013 Presidentís Budget Request officially establishes a new baseline for JWST (as shown in the Budget Summary table above) consistent with direction in NASAís FY 2012 appropriation to cap JWST formulation and development costs at $8.0 billion.

Cost: Bottom-up review resulted in a revised life cycle cost estimate of $8.835 billion. This estimate is consistent with the 66 percent joint confidence level with a cost confidence level that is significantly higher than the 80 percent recommended by the ICRP.

Schedule: Bottom-up review resulted in a revised
development schedule, with launch in October 2018. The revised schedule incorporates 13 months of funded schedule reserve."

So the $8.9B in the thread title could be 6.2B, but including confidence and 13 months of schedule is 8.9B.  This does not include operations (FY18 and beyond).

So now what is the big news?


Some big news is coming wrt JWST

Quote
This estimate is consistent with the 66 percent joint confidence level with a cost confidence level that is significantly higher than the 80 percent recommended by the ICRP.

What?

This makes absolutely no sense.  These people are just talking.  66% is higher than 80%?

Or is that the "biog news"?
Sometimes I just flat out don't get it.

Offline vulture4

  • Full Member
  • ****
  • Posts: 1007
  • Liked: 317
  • Likes Given: 90
Re: JWST: Albatross of SMD now $8.8 billion
« Reply #384 on: 11/18/2013 07:54 pm »
I agree. These estimates of "confidence levels" appear to be almost meaningless. Either we have an accurate estimate of the remaining cost or we do not.

Offline beancounter

  • Full Member
  • ****
  • Posts: 1252
  • Perth, Western Australia
  • Liked: 106
  • Likes Given: 167
Re: JWST: Albatross of SMD now $8.8 billion
« Reply #385 on: 11/19/2013 04:47 am »
I agree. These estimates of "confidence levels" appear to be almost meaningless. Either we have an accurate estimate of the remaining cost or we do not.
Yes it does make sense.  It simply means that they now have less confidence than before.  This is quite simply 'polispeak'.  Thought everyone on this thread would understand that!!  :P
Beancounter from DownUnder

Online JohnFornaro

  • Not an expert
  • Senior Member
  • *****
  • Posts: 9235
  • Delta-t is the salient metric.
  • Planet Eaarth
    • Design / Program Associates
  • Liked: 657
  • Likes Given: 360
Re: JWST: Albatross of SMD now $8.8 billion
« Reply #386 on: 11/19/2013 02:00 pm »
To me, that sounds like a confirmation that "polispeak" is the verbal equivalent of "almost meaningless".

The takehome is that we do not have an accurate estimate of the remaining cost.  They are even less confident in their own estimate.
« Last Edit: 11/19/2013 02:03 pm by JohnFornaro »
Sometimes I just flat out don't get it.

Offline joek

  • Senior Member
  • *****
  • Posts: 3128
  • Liked: 874
  • Likes Given: 444
Re: JWST: Albatross of SMD now $8.8 billion
« Reply #387 on: 11/19/2013 10:08 pm »
I agree. These estimates of "confidence levels" appear to be almost meaningless. Either we have an accurate estimate of the remaining cost or we do not.

There are two confidence levels referenced which are related but not the same:
Quote
... Estimates that include combined cost and schedule risks are denoted as joint confidence level; all other confidence levels reflect cost confidence without necessarily factoring the potential impacts of schedule changes on cost.  For the new JWST baseline, the associated cost confidence level is significantly higher than the 80% recommended by the ICRP.
So yes, they appear to have high confidence in the cost estimate (> 80%); not so much for the joint schedule + cost estimate (66%).

Offline BeanEstimator

  • Full Member
  • ***
  • Posts: 357
  • Pray for Mojo
  • Taxation without Representation
  • Liked: 3
  • Likes Given: 1
Re: JWST: Albatross of SMD now $8.8 billion
« Reply #388 on: 11/19/2013 10:49 pm »
I agree. These estimates of "confidence levels" appear to be almost meaningless. Either we have an accurate estimate of the remaining cost or we do not.

There are two confidence levels referenced which are related but not the same:
Quote
... Estimates that include combined cost and schedule risks are denoted as joint confidence level; all other confidence levels reflect cost confidence without necessarily factoring the potential impacts of schedule changes on cost.  For the new JWST baseline, the associated cost confidence level is significantly higher than the 80% recommended by the ICRP.
So yes, they appear to have high confidence in the cost estimate (> 80%); not so much for the joint schedule + cost estimate (66%).

yay...somebody gets it! 

no comment on the accuracy of the $8.XB or any upcoming news.  was simply stopping by to try and detangle what is actually very simple....

the difference in statistical confidence levels is nothing magic - on the one hand you have a cost-only confidence level of the budget/value, and on the other hand you have a joint cost and schedule confidence level.  if you read between the lines what they are telling you is how they are managing the plan.  they have a high cost confidence level >80% (lot o money in the budget), but they have a low schedule confidence level, hence the reduced joint value (66%).  again, it all ties back to the plan.  they are 66% joint confidence level of achieving the plan for cost and launch date.  they have purposely set the plan with high cost confidence in order to compensate for the results of the risk analysis that yielded a low schedule confidence level.  clear as mud? 

it certainly isn't meaningless, and as i just pointed out - if you read between the lines, it tells you quite a bit. 

now if you want to get into the whole "g-d cost growth, stupid estimators cant estimate, stupid nasa can't manage" then i return you to that regularly scheduled program...all analysis can suffer from GIGO and of course sales/unbridled optimism.  just because it's "joint cost and schedule" doesn't change the need for a solid basis of estimate  but it isn't meaningless or "polispeak".  i'm sure most of you know exactly what a confidence level is in stats anyway...(i hope)

a single point estimate value, is 100% the wrong value.  if you want an eac on a contract you can go fetch from one from the dime a dozen evm systems in play.  quality analysis and estimation will incorporate risk and uncertainty and instead of producing a single point estimate will provide a range of statistical values that will then allow you to determine the "confidence level" of your plan budget/schedule/whatever.   https://acc.dau.mil/adl/en-US/316093/file/46243/AF_Cost_Risk_and_Uncertainty_Handbook_Jul07.pdf  sorry i don't have more time, you can read more about risk analysis, uncertainty, simulation modeling and statistical confidence levels at that link.  its all about moving away from a point estimate, which is based on heaps of assumptions, to a quantifiable risk analysis that produces a range.  set your budget in the range, set it for success, and then go off and execute and manage performance.

here is a link to a table of current nasa projects and their associated confidence levels.see page 13 (M&P 203) http://www.nasa.gov/pdf/754125main_12-NASA_FY14_M&P508-pt3.pdf
Note:  My posts are meant to discuss matters of public concern.  Posts and opinions are entirely my own and do not represent NASA, the government, or anyone else.

"Balancing Act: Public Employees and Free Speech"
http://bit.ly/Nfy3ke

Online JohnFornaro

  • Not an expert
  • Senior Member
  • *****
  • Posts: 9235
  • Delta-t is the salient metric.
  • Planet Eaarth
    • Design / Program Associates
  • Liked: 657
  • Likes Given: 360
Re: JWST: Albatross of SMD now $8.8 billion
« Reply #389 on: 11/20/2013 12:50 am »
I agree. These estimates of "confidence levels" appear to be almost meaningless. Either we have an accurate estimate of the remaining cost or we do not.

There are two confidence levels referenced which are related but not the same:
Quote
... Estimates that include combined cost and schedule risks are denoted as joint confidence level; all other confidence levels reflect cost confidence without necessarily factoring the potential impacts of schedule changes on cost.  For the new JWST baseline, the associated cost confidence level is significantly higher than the 80% recommended by the ICRP.
So yes, they appear to have high confidence in the cost estimate (> 80%); not so much for the joint schedule + cost estimate (66%).

Thanks for clearing that up.  In real life, they can only deal with joint schedule + cost estimate.  The other permutations are a smoke screen meant to deceive the policymaker.
Sometimes I just flat out don't get it.

Online JohnFornaro

  • Not an expert
  • Senior Member
  • *****
  • Posts: 9235
  • Delta-t is the salient metric.
  • Planet Eaarth
    • Design / Program Associates
  • Liked: 657
  • Likes Given: 360
Re: JWST: Albatross of SMD now $8.8 billion
« Reply #390 on: 11/20/2013 12:58 am »
it certainly isn't meaningless, and as i just pointed out - if you read between the lines, it tells you quite a bit.

Thanks for clearing up the semantic differences between the several statistical methods.

From a pragmatic viewpoint, the statements are functionally meaningless.  There is no launch date, nor guaranteed cost figure.  The statistics about con-fidence levels tell us nothing.

The history of the program, as is well known, began with a deliberately misunderestimated cost of $500M.  Those of us who tire of the continued non-performance of this telescope view these tweaked and highly technical statistics as means to obfuscate, not as means to inform.  There is no accountability for their accuracy.

Of course, if you like your space telescope, you can keep it.  Period.
« Last Edit: 11/20/2013 01:00 am by JohnFornaro »
Sometimes I just flat out don't get it.

Offline baldusi

  • Senior Member
  • *****
  • Posts: 7439
  • Buenos Aires, Argentina
  • Liked: 1459
  • Likes Given: 4525
Re: JWST: Albatross of SMD now $8.8 billion
« Reply #391 on: 11/20/2013 01:08 pm »
it certainly isn't meaningless, and as i just pointed out - if you read between the lines, it tells you quite a bit.

Thanks for clearing up the semantic differences between the several statistical methods.

From a pragmatic viewpoint, the statements are functionally meaningless.  There is no launch date, nor guaranteed cost figure.  The statistics about con-fidence levels tell us nothing.
Can you give it just a bit of thought before writing? There's no way to know before you do this things. Taking a statistic approach is the only pragmatic approach. If you can't understand decision taking in stochastic environment, I would recommend that you go and study a couple of introductory econometric books. The Gujarati is pretty good.

Online JohnFornaro

  • Not an expert
  • Senior Member
  • *****
  • Posts: 9235
  • Delta-t is the salient metric.
  • Planet Eaarth
    • Design / Program Associates
  • Liked: 657
  • Likes Given: 360
Re: JWST: Albatross of SMD now $8.8 billion
« Reply #392 on: 11/20/2013 01:28 pm »
it certainly isn't meaningless, and as i just pointed out - if you read between the lines, it tells you quite a bit.

From a pragmatic viewpoint, the statements are functionally meaningless.  There is no launch date, nor guaranteed cost figure.  The statistics about con-fidence levels tell us nothing.

Can you give it just a bit of thought before writing? There's no way to know before you do these things. Taking a statistic approach is the only pragmatic approach. If you can't understand decision taking in stochastic environment, I would recommend that you go and study a couple of introductory econometric books. The Gujarati is pretty good.

Parsing out the unnecessary ad hominem and focusing instead on the current "functional meaninglessness" of the current statements to policymakers who, like me, do not have intimate familiarity with the Gujarati method of pricing, we find that policymakers still have not been given a launch date nor a cost, but that the JWST program managers admit that their confidence level on cost and schedule is only 66%.

It is not my ignorance of the math involved which will threaten JWST with cancellation.  It is the decade of Gujarati mumbo jumbo which to date has not resulted in a functioning telescope, despite increasing heaps of scarce dollars, and which may result in program cancellation.  As more or less human beings, policymakers don't like being told of their ignorance while the program managers demand more money and time.

This is part of the "between the lines" meaning that BeanEstimator, in my mind, could be touching on.  JWST program managers are still saying, in effect, "rocket science is hard".  They simply, at 66% confidence level at this late stage in the program, still do not have a handle on cost or schedule.

Perhaps if cost and schedule could be actually managed, rather than handwaved away in inscrutable math, they'd have less to be concerned with about the growing sense, yet still a minority sense, in Congress, that this program should be cancelled.
Sometimes I just flat out don't get it.

Tags: