- NASA Watch
- May 28, 2023
GAO Sees Through NASA's SLS/Orion Smoke and Mirrors
Actions Needed to Improve Transparency and Assess Long-Term Affordability of Human Exploration Programs, GAO
“- The SLS estimate is based on the funding required to develop and operate the initial 70-metric ton variant through first flight in 2017 but not the costs for its second flight in 2021. NASA is now incurring some costs related to the second flight, but it is not currently tracking those costs for life cycle cost estimating purposes. Furthermore, the estimate does not include costs to incrementally design, develop, and produce future 105- and 130-metric ton SLS variants which NASA expects to use for decades. NASA is now funding concept development and analysis related to these capabilities.
– The Orion estimate does not include costs for production, operations, or sustainment of additional crew capsules, despite plans to use and possibly enhance this capsule after 2021. It also does not include $4.7 billion in prior costs incurred during the approximately 4 years when Orion was being developed as part of NASA’s now-defunct Constellation program.
– The ground systems estimate excludes costs to develop or operate the ground systems infrastructure beyond 2017, although NASA intends to modify ground architecture to accommodate all SLS variants.”
Anyone surprised? Not me. Truthful accounting would turn out scary numbers. We have to keep the real costs hidden, then ask for more money later.
In other words, the costs of the SLS/Orion system could be a lot higher, it’s just that NASA has declined to release accurate figures to let the bean-counters figure that out.
While proudly proclaiming the programs to be “on schedule and on budget” every time they get near a microphone.
Uncannily on target about the accounting trickery. Well informed and smart GAO auditors. Now the next report…chances anything SLS Orion EVER adds up for exploration of any kind such as Lunar or Mars, on the expected budgets.
From what I’ve seen the number one lesson learned management learned from Constellation being canceled was to control cost information better, more tightly. In effect, lie better next time.
The only surprising thing about this is that it has taken this long for someone in government to bring the matter up. In the blogosphere, everyone has known this for years. What is still telling is that it was an outside organization that brought it up. I would have expected the internal NASA groups to have identified the issue and call for correction first. SLS and Orion are not the only issues but they would be a good start. Amazing that NASA management is this out of hand.
Internal NASA groups that would have done this were disbanded after Constellation was canceled. The teams awaited the tasks for SLS and Orion. Word went out early-no cost numbers would be looked at further…disband.
I am curious who these groups were and what they estimated for the cost of Constellation.
There are things called standing review boards. Also there were independent cost analysis of large scale projects. Back in the, pre pc 80s, these type of cost analysis were often parametric, but also hampered by lack of data from past or current projects. The 90s saw a surge of demand for cost data from past projects with which to inform future projects. Unfortunately a tragedy of the commons ensued. Collecting data for the benefit of others in the future, potentially leading to negative criticism of your current project, was always a hard sell. As boards became common for large scale projects, alongside many an independent analysis (albeit an internal NASA civil servant) things went sour in the last decade.
Why assist an internal analysis that would only catch projects doing things on the sly? Honesty is not rewarded, so why ask or support someone providing it?
Oddly enough, as bad as cost estimation had been way back, it was improving quickly in the cost community just as these type analysis fell out of favor. We failed to predict that, ironically. As the ability to crunch numbers got valuable insights, peoples experience had matured, and data had been collected to support estimates, we should have seen the collision ahead in a culture and system based on only short term thinking.
“Amazing that NASA management is this out of hand.”
We call that lack of leadership.
This is just one example of many that NASA managers knowingly lie to the general public and others about the true internal estimates for the real cost of a program, hoping that it will be too late or politically unfeasible to cancel a project that everyone working on it knows is spiraling upwards and sliding to the right out of control when the truth becomes known.
You got it.
Does anyone still believe in this mess of a program? SLS was forced on NASA and they’ve had to defend it. But everyone should be aware by now that it’s always been a jobs program. NASA never had any intention of flying the thing except perhaps once and then retire it like Ares1-X. Just another Cx disaster.
Orion was going to be cancelled outright but Obama had to compromise to keep CC funded. Oh well! And the vehicle isn’t being designed for reuse, only expendable. Been significantly de-scoped as well.
Personally, I’m guessing two flights – the uncrewed EM-1 and the crewed EM-2, which I’m expecting to be whittled down to a bare minimum on cost grounds.
I say two because they’ve ordered enough European Service Module parts for two spacecraft and there has been no real talk about what to do after that. From this, I suspect that even NASA’s internal planners expect the program to be cancelled after EM-2.
Just to emphasise: I am not happy about this; I don’t even feel a degree of schadenfreude. This could prove fatal for NASA HSF and I get the impression that, at least for some of the factions involved, that was always the game plan.
The MPCV is also a vehicle with limited capabilities. For instance it does not, as currently designed, have a rendezvous sensor. It could not visit the ISS even if they decided to do that as a test (as designed right now). It could not rendezvous with a small asteroid either of course. That is just one example of avionics that have been taken out of MPCV. They could put one in – but that would add cost.
I just don’t understand how they can go public with a plan for a 130mt rocket and not release the true cost. Transparency will yield better results.
Great headline, Keith.
I will be surprised if this changes any attitudes inside the agency. Many will see this as a political attack by their enemies in the administration.
The deception starts at the top. The top guy is basically clueless when it comes to systems engineering so he get duped into supporting these schemes. Then at the next level they basically work through a bunch of constructs to obfuscate the truth. Its pervasive, I am sure a little poking around about the plan to “refurb” the test stand at MSFC, a little bit of poking around about ARM, and other stuff that is in work will show that there is a direct effort to keep the truth from the public. I don’t blame the hard working people that work on these projects, they believe their management, problem is that their management is no longer believeable
You mean to tell me that the top people don’t have experience in the areas they are expected to manage?
How could that be?
Civil service rules require experienced, educated, knowledgeable people. You mean that people without experience or education or knowledge are placed in the leadership positions?
There is little in the way of requirement that a CEO in private industry have hands on experience with the hardware. Mary Bara is the first GM CEO in many years who actually knows how to build a car.
The CEO, yes that is true. Their job is politics, marketing and finance. It is not technical and it is not flying Shuttles or airplanes.
However the people resoonsible for designing and developing vehicles which are most of the people under the CEO should have some knowledge of design and development.
Apparently in the human space program they take a very cavalier attitude that perhaps was learned from the astronaut office over the last few decades that anyone can do any job; no knowledge or experiece required. It has not worked out too well. In case no one has noticed we don’t have a Shuttle, we don’t have an Orion despite a decade and $16 billion, and we don’t even have any NASA developed prospects in the plan for any time in the forseeable future.
Only one or two senior managers at KSC have any design and development experience. Nearly all of the middle managers and most of the line managers also have no experience in development. Most of the workforce doesn’t either, or if they do it is minimal sustaining engineering experience from Shuttle, not clean-sheet design. The whole place is in shambles and management thinks we can do no wrong. “League of morons” comes to mind. Cost estimating is just the tip of iceberg, one indicator of complete incompetence. And I mean complete.
The people that work on projects have the responsibility of informing the public if their taxes are being wasted. NASA’s reliance on systems engineering itself is part of the problem. It asserts that analysis is equivalent to experience. It’s not, by a long shot.
SLS/Orion is a little unusual in that the Administration wanted to cancel it but was prevented from doing so by a Congress intent on wasting tax dollars. So who should we hold accountable?
Just MHO but for the amount of money NASA spent on the J-2X, NIH could award roughly 600 level-O grants, major research programs leading to significant advances in treating disease. NIH awards in some areas are a bit cliquish but most are fairly objective. NSF might fund 4000 significant $250K research projects leading to advances in human knowledge. Some parts of NASA such as the smaller (nonflagship) science missions are in most cases efficiently run. Some large DOD R&D projects have been more wasteful then NASA, while DARPA has a good reputation for efficiency.
I can answer some of your questions. RE cost differentials between private industry and government in general, the rule of thumb we used back when I was in private industry was that doing a cost-plus government contract would at minimum cost two-to-three times fixed-price commercial.
Beyond that, individual government departments have their own additional cost multipliers. Most of what I might say on that is pretty subjective, but NASA actually quantified that for their rocket-building bits a few years back. They ran a study of how much SpaceX F9 would have cost NASA to develop through first flight under standard NASA launch-vehicle development procedures, using a standard NASA project cost estimating model. Their answer was $3.977 billion.
SpaceX’s actual costs for F9 through first flight were $300 million. Add on the $90 million they spent on F1 because a lot of that work applied directly to F9, and actual F9 cost was $390 million – 1/10th NASA’s estimate for F9 done the same old NASA way.
Worse, GAO around that same time released a report on a half-dozen recent major NASA projects. The average actual major NASA project cost was about 50% higher than NASA’s initial project estimate.
So, for MSFC doing a major launcher development, as of 2011 we’re talking a 15X multiplier over the commercial equivalent. And this was known when Congress decided to go ahead and insist on SLS. And I would guess it’s only gotten worse since. Weep.
Have people ever complained before? Yes, for decades now. It took a loooong time to get any traction at all.
Take a look online for what the other nations spend on their space programs. The CIA databook has some realistic numbers. NASA spends as much as all others combined. Money does not tell the entire story. Numbers of workers would be a more accurate measure. I am convinced that if NASA management were better they could accomplish far more with the same dollars. Politis has an effect but you don’t need to reduce the dollars going to a particular district. You just need to do more with it.
We have generally two types of specifications in my field and in most of construction. One is Performance Specs: these simply outline and describe what the final product is to look like; with a few exceptions, these specs leave the ‘how’ to the contractor, who has been pre qualified and known to be capable of the work. Oh, sure, there’s a products list, but the how-to is left to guys who do this work day after day.
A more detailed set of specifications are also possible. These answer the “how”, written by folks who feel they are qualified that if they say “how” and the instructions are followed, a predictable result will occur.
These aren’t the only two options but they do represent the totality of looking at specs. My own choice is inevitably the first approach, for several reasons. Although I’ve been a practicing consultant for more than 30 years, I do know that the qualified contractors have much to teach me. I simply find those who can do the job and ask for prices.