I agree with Bob Zubrin that the numbers coming out of Aerospace on development costs are highly suspect:
Following retirement of the Shuttle, Aerospace’s cost estimates have ground operations cost triple to $900 million by 2012, and then continue to rise to $1.8 billion by 2022. This sixfold rise in ground operations cost would be difficult to explain in any case, but in the absurdity of this instance is outstanding since during the entire ten year 2012-2022 period in question, there are NO heavy lift flights at all for the ground operations to support. In other words, the Aerospace Corp’s estimates have NASA’s ground operations costs rising sixfold over Shuttle flight support requirements, spending $15 billion over ten years, in order to launch nothing.
Rather than basing their projections on actual grounded estimates of development costs for different types of hardware, what the Aerospace Corporation appears to have done is to regard each program element as an “activity” which each need to be funded continuously at multi-billion levels per year. The program is then arranged so that no flights beyond LEO can take place before around 2023. So, with a budget of about $3 billion per year (equivalent to 30,000 employees on payroll) the Ares 5 development program is allowed to run for 12 years, bringing development costs to the spectacular $36 billion level. Why, in this day an age, a launch vehicle development program needs to run 12 years (or require 30,000 people) is left unexplained. In contrast, the Saturn V development, done at a time when much more still needed to be learned about launch systems, took only 4 years to complete (Contract awarded in 1962, first flight in 1966.) Summing up all such activities the net result is a program which costs $14 to $20 billion per year (140,000 to 200,000 employees) and which does nothing at all for a decade.
I disagree with this, though:
Americans want and deserve a space program that is actually going somewhere. In order for that to happen, a radically different methodology to that being accepted by Augustine Committee needs to be employed. Rather, a real goal, worthy of spending serious money on, if necessary, needs to be selected. That goal can only be humans to Mars. Then a minimum cost, minimum complexity, and, critically, fastest schedule plan needs to be selected to achieve that goal. In order to minimize schedule and cost, such a plan should avoid advanced propulsion, on-orbit assembly, or other futuristic ideas, and instead get the job done in the manner of the Mars Direct and Semi-Direct missions by employing a strategy of direct transportation to Mars of required payloads using an upper stage mounted on the heavy lift launcher.
I don’t agree that the goal “can only be humans on Mars,” at least as the focus for the program, though it may be a useful long-term ultimate one, as the Augustine panel has stated. And the notion that on-orbit assembly is a “futuristic technology” is quite amusing, seeing that we’ve been doing it with ISS for over a decade.
[Update late afternoon]
There are a lot of good comments by “Red” over at Space Transport News. Also, I would add that while I consider the numbers suspect, this isn’t meant to be a criticism of Aerospace or its methodology so much as the NASA inputs and assumptions.
Incorrect, first flight of Saturn V was November 9, 1967.
This sixfold rise in ground operations cost would be difficult to explain in any case, but in the absurdity of this instance is outstanding since during the entire ten year 2012-2022 period in question, there are NO heavy lift flights at all for the ground operations to support.
This statement is misleading. First, Zubrin is quoting numbers from the wrong chart. These numbers are not for Ares V or Shuttle C but for the “Ares Lite”, which isn’t even discussed in his article.
If you look at the Ares V chart, ground costs don’t simply “triple to $900 million by 2012, and then continue to rise to $1.8 billion by 2022,” as Zubrin states. They go from $500 million in 2012 to $1.9 billion in 2022, then down to $1.7 billion in 2024.
Significantly, the costs go up and down during this time. They actually peak at $2.3 billion in 2021.
Why would ground ops costs for Ares V peak before the rocket even starts to fly? The only logical explanation is that Aerospace Corp. expects NASA to be building new launch pads, vehicle assembly buildings, and other ground facilities during that time.
That explains the great mystery Zubrin was unable to figure out. Ground costs are independent of flight rate in a given year. Zubrin ignores that fact.
In fact, he seems to ignore ground costs entirely when he says SpaceX offered to develop a heavy lifter for $2.5 billion and Lockheed Martin estimated the cost at $4 billion. (And does anyone believe that $4 billion estimate wouldn’t go up, just as the initial estimate for Ares I went up?)
The chart for Shuttle-C shows lower ground operation costs during the 2012-2022 period — only $400 million in 2012, peaking at $1.5 billion in 2021, dropping back to $1.0 billion in 2022. Those costs do rise again later, though, climbing all the way to $2.0 billion in 2027. All three scenarios have roughly the same annual cost by 2027.
The Aerospace Corp. numbers may or may not be accurate, but they clearly do have some analysis behind them. They did not simply “regard each program element as an ‘activity’ which each need to be funded continuously at multi-billion levels per year” as Zubrin states.
>== Ed says
>Why would ground ops costs for Ares V peak before the rocket even
>starts to fly? The only logical explanation is that Aerospace Corp.
>expects NASA to be building new launch pads, vehicle assembly
>buildings, and other ground facilities during that time.
>
>That explains the great mystery Zubrin was unable to figure out.
>Ground costs are independent of flight rate in a given year. Zubrin
>ignores that fact. ==
Agreed. I don’t know where Zubrin is getting his numbers from, but I strongly expect hes not allowing for the heavy cost growth of a gov program, vrs a commercial program.
Also see comments at hobby space:
http://www.hobbyspace.com/nucleus/index.php?itemid=14894#c
All I need to know about Aerospace is that they hired the jsc cs, upon his retirement, that spoke of nothing but aliens (greys, etc) and conspriracies…
Zubrin shows his familiar mix of ignorance and arrogance.
Obviously, given an Apollo era willingness to vastly increase peak year funding, we could do it much sooner. That was then, this is now.
And there are also real costs to not funding an activity continuously, as the commercial HLV option shows. If you want to stop funding NASA ground operations, you need to consider termination costs for the existing staff and the cost of mothballing or scrapping the physical infrastructure.
Shuttle fixed costs are something like $2B a year, excluding the Orbiter. How can ground operations (whatever that includes) be less than that? And if you don’t keep the standing army around, how could you possibly resurrect it without enormous cost later?
In general terms, you can go examine the cost estimates at the beginning of just about any NASA program and then go see what they actually cost and find a wide discrepancy. A NASA program that comes in on time and budget is the exception (and a rare one at that), not the rule. The same can be said for US military and NRO space projects. I wish it wasn’t the case but wishes aren’t a good way to determine procurement policy. Aerospace recognizes that but it appears Zubrin doesn’t.
As I recall, that Aerospace Corp study was festooned upon its introduction with so many caveats that the Auggies had to complement the presenter on his ability to keep his lawyers happy. Little wonder then that the whole exercise drifts up to the fiscal “orbit of Pluto”. Earlier studies of Mars Direct ranged in cost from $50B to $100B, yet (to my knowledge) nothing was said to rebuke those efforts before launching into the bloated exercise under discussion.
As to the lower estimates of SpaceX and Lockheed Martin, those perhaps do not indeed factor in the “standing Army” or new facility cost. Yet they fall within the “factor two” spread that seems a fairly common range for careful analysis. Why are completely new facilities needed anyway, and why should we be so foolish as to fail ourselves by allowing a break in the continuity of heavy lift expertise? Why are inflating factors of six and seven over normal project costs allowed to slip in here so completely undefended?
Clearly we must await the final report and see what true use is made of the Aerospace Corp effort, but with Dr. Zubrin, I must say the whole thing has a pretty strong “skunk factor”.