Push Back on MSL Cuts
Budget Axe Falls on Mars Science Laboratory — Science Plans for the Mission Cut, Planetary Society
“The loss to science on MSL seems out of proportion,” said Wesley T. Huntress, Jr., former NASA Associate Administrator for Space Science and member of the Society’s Board of Directors. “The goal of MSL is to conduct science, and to throw out so much of the mission science objectives for less than 4 percent of the mission cost, and for assurance costs that have net yet been realized, seems penny-wise and pound-foolish,” he added.”
Editor’s note: OK Wes: where’s the additional money supposed to come from? This is a zero sum game. To write a check for a mission that is out of the cost box means taking something away from another mission that planned ahead. Isn’t it about time to instill some cost realism – and responsibility – into this process?
Reader comments – send them to [email protected]. Your comments thus far:
Hi Keith,
I would like to add to the discussion thread, suggesting that the true budget-busting culprit is often times not the obvious party. To the untrained eye, it might seem that the proposing mission team and/or Principal Investigator should ultimately be resonsible for delivering a mission on budget. In this simplistic view, the “provider” proposes to deliver something, and the “customer” (NASA) agrees to fund the project. Afterall, this is how most things work in our life, right?
Well, the reality of a project with NASA is quite different. Imagine you want to buy a house (just like NASA might want to buy a mission). So, you hire a builder, he proposes the design/budget, and you write him a check. But wait, it turns out that you also own a brick company (NASA field center), so you tell the builder that he has to buy your bricks for fives times the going rate for bricks. Now, the builder says he doesn’t want to buy your bricks, so basically you rig the system so that his project gets cancelled unless he buys your bricks (the Mafia kind of work this way…). Oh, wait again, you tell the builder that you now want 3 stories instead of 2 (requirements creep). Now, all of the sudden, the “builder has gone over budget.” So, you flog him in the local newspaper, saying how horrible he is. Heck, you even hire a “manager” on your side of the fence who goes around threatening all the local builders because they seem to always go over budget. Gee, those darn builders, how is it that everytime they build a house for me, they bust the budget. Boy, there must be something wrong with them!
I’ve seen this game on JWST and Sofia. Both involved field centers that spent hundreds of millions of dollars of project money that produced nothing and was required internally within NASA against the wishes of those with a scientific stake in the project. The argument within NASA at the time was that HQ did not have the authority to stop all this waste because they had to let NASA field centers spend down project budgets in order to avoid layoffs in key congressional districts.
Really, this isn’t rocket science.
Name witheld
Keith,
This all comes down to whether or not you have adequate cost reserve built into the initial commitment. History provides ample evidence on which to select levels of reserve that provide whatever confidence you like of both not overrunning and not reducing scope.
However the very same NASA HQ that is demanding scope reductions and blaming projects for underestimating the costs is also pressuring new projects to maximize scope and minimize cost in order to be selected to go forward, whether they are competed or directed. There is a large range of possible cost reserves you can assign to a project and still be considered credible. (More on why that is below.) The pressure to be selected forces the cost reserve to the low end of that range, to be replaced by increased scope to make the project more attractive. In that case, overruns are inevitable. Even if you know that you will be descoped when you overrun, which is the “discipline” that is currently in vogue, you will still take this path, since you’d rather be eventually descoped than not selected in the first place.
NASA HQ is even able to do this directly without the help of a project when they select a payload for a directed mission. The desire to pack as much science as possible in the available mass and budget results in inadequate consideration of the required reserves, for what are often new instrument concepts that promise exciting and compelling investigations, but with little or no flight heritage. Pushing the edge of the envelope is what NASA is all about, but you need adequate resources to do that.
It doesn’t hurt a project to be descoped — a project is only a means to an end. It will do what it can with the resources it has, and will do so prudently. Instead, descoping hurts the sponsor in reducing the bang for the buck that they are getting. When you skim off the top, you’re always taking away the cream. So it is up to the sponsor, NASA HQ in this case, to decide how to incentivize the behavior it wants. If NASA wants as many launches as it can get, with low bang-for-the-buck, descoped payloads, then it is on the right path. If NASA wants to maximize the science return per dollar, then it is not.
The objective should be to avoid late descopes to the greatest extent possible. You end up spending a lot of money on something that doesn’t fly. For the very same reason, you don’t want to move money from one mission to another. That also requires descopes or possibly cancellation, which wastes money in exactly the same way. To avoid all of that, you need better estimates to start with. The key is to mitigate the unfortunate condition mentioned above: a wide range of credible reserves, even after independent review. To narrow that range, you simply need a much better understanding of the job, having done enough work to retire the big surprises. Then you can review the costs without huge inherent uncertainties, allocate your budget to the appropriate set of missions, and hold some of the reserve at the HQ level to take advantage of the statistics of a large number of projects.
The problem is that we are committing to project and instrument costs and scope with insufficient design and prototyping work. Studies have shown a strong correlation between project success, which includes staying on budget, and the amount of money spent in formulation (prior to commitment) as a fraction of the total cost of the project. Good fractions are 5% to 15% of the total cost spent in formulation. What we see too often is less than 1% spent in formulation on, for example, a Discovery mission or on an instrument responding to an AO for a directed mission. The answer is straightforward. Spend the right amount to understand what you plan to do before you commit to doing it. Do the detailed design work and implementation planning, as well as the necessary prototyping work for those aspects that have no flight heritage. Even after doing all that, you’ll still need about 30% reserve, but the uncertainty on that amount is now much smaller.
If we want discipline, then it should be applied to being willing to spend sufficient up-front resources and time on new missions and instruments. Trying to “save” money by short-changing formulation will always cost more in the end.
Mark Adler
(Speaking for himself)
As with your other commenter, I also wish to remain anonymous. In addition, I would like to further add (as also was done by a previous commenter) that the situation is likely more complicated than just some folks went over the cost cap. As is often for missions and instruments like these, a team proposes an instrument with a given design and set of requirements. NASA then selects such an instrument, but imposes additional requirements that were not in the original proposed design. In order to meet such new requirements, costs inevitably increase to meet the new requirements. For NASA to come in and cut off a project for “going over the cap” without necessarily accounting for requirement creep seems to be a bit of a problem … this is all said with the recognition that almost all involved believe that costs should be controlled. However, it should be a two way street with both NASA and the instrument providers being held accountable both for costs and requirements.
Editor’s note: On further thought, I have to say that I feel that it is about time that people who overrun and expect to have all of their overruns taken care of because of their mission’s or instrument’s importance need to be held accountable. Of course, these same people also try and put their little pile of money in context of the agency’s overall budget – as if that excuses their mistakes or overrun.
An overrun is an overrun. A schedule slip is a schedule slip. Trying to stand next to an elephant with your hand outstretched so as to look small by comparison doesn’t take away from the nature of your own overrun or slip.
If people were more serious about the management of their project (perhaps even when they propose it), problems might go away sooner or noticed in a smaller form so as to be more easily dealt with – and dealt with earlier.
And how better to instill that behavior that to threaten – and then implement – changes from HQ when costs do get out of hand? When he arrived at NASA Alan Stern said he was going to do things like this and hold people accountable. Now he has done so. And people are surprised? Pay more attention to the man behind the curtain, folks.
Or should there be no rules – except the one where the project that screams the loudest gets the money? That’s management by megaphone.
I have gotten a few complaints suggesting that since other parts of NASA are (seemingly) not held to strict (or equally strict) cost overrun constraints that science missions should not be held to them either. OK. Let’s just accept that as fact. So just where does NASA start to exhibit some cost accountability? If one part of the agency doesn’t do so, does that mean that it is OK for projects in another part of the agency to overrun as well? Or that it is wrong to at least try and start to try and fix the problem – somewhere?
The Planetary Society has been vocal on this topic. It is good that they are vocal about things – and I sincerely hope that they continue to be. Alas, their blogger Emily Lakdawalla refers to these cuts as “weird” and “strange” but misses the point. The Society’s Executive Director, Lou Friedman, has written a letter to Sen. Mikulski about this proposed action asking that Stern’s MSL decision be stayed until Congress examines the proposed cuts. Fine. See you in six months to a year, Lou – for that is how long it will take Congress to do this – by then the problem being addressed will simply be worse.
If people think that the problem is insufficient Science Mission Directorate funding (I happen to be one of them), then they should go complain to Congress about that. But dragging Congress in to micromanage things done in response to an extant budget environment simply hampers the management of these limited reserves and inevitably causes money to be taken from something else to pay for such micromanagement.
Yet keep in mind that even with a windfall budget increase, funds would still be limited and they will still need to be managed prudently. If NASA does not learn how to exhibit such discipline, it will find it increasingly hard to work within budgets that will always, by definition, be constrained.
Without cost discipline, we will not explore our solar system.
Others will.
Since GAO has long been unhappy with NASA project cost overruns, close controls on all NASA projects with a life cycle cost over $250M are now written into the NASA authorization act in the 109th Congress (see Public Law 109-155). As soon as project development cost exceeds the original estimate by 15% or a milestone is delayed by six months, the Administrator has to notify Congress and report to them within three and a half months. I would guess that’s in Dr. Stern’s mind, and the minds of all Associate Administrators: do I want my boss to have to go to the Hill to explain why we overran?
And if the overrun is 30% or more, starting 18 months after the Administrator’s report to Congress, _no_ funds can be expended on the project unless Congress specifically authorizes them — except costs for termination.
Dear Mr. Cowing,
I wish to respond to your question to Wes Huntress in your “Editor’s Note” related to recent Mars Science Laboratory budget cuts. In the name of full disclosure, I am a member of an MSL science team.
Perhaps the money MSL needs to maintain its budget reserves could come from the same place that funding came from to pay for the “sample cache” added to MSL by a NASA HQ directive. Design and incorporation of the “sample cache”, which is claimed as a pathway for Mars sample return, diverted the attentions of MSL engineers and wasted precious time and effort that would have been directed to solving the challenges that MSL faces. The “sample cache” is, at best, a symbolic entity. It is a hockey puck-sized container that mixes uncharacterized dirt and pebbles together and requires a future mission to revisit the same place MSL will go. If NASA were serious about Mars sample return, they would have funded the Sample Collection for Investigation of Mars (SCIM) Scout mission to return a similar “grab bag” sample to Earth much sooner than the hypothetical, cache-grabbing sample return mission.
It is convenient that the person in charge of restoring “realism” and “responsibility” to the mission development process is doing so after his budget-busting mission is off the ground. AA Stern’s New Horizons mission did not stay under it cost cap (http://www.space.com/spacenews/businessmonday_051212.html) yet it is off with its full complement of functional instruments to Pluto and beyond. NASA HQ also does not acknowledge its role in stressing the MSL development process by levying operational requirements on the mission that were not desired by the scientific community. HQ insisted upon maintaining functionality at latitudes between 60N and 60S despite the fact that 90% of the landing sites proposed in the first Mars Science Laboratory landing site workshop were well within these latitudes. Striving to meet the imposed goal of low temperature operations at high latitudes cost the project at least a year of effort and resources.
If you choose to you use this comment, I ask you keep me anonymous.
