MSL Revisited

The Washington Post has an article on the problems of the Mars Science Laboratory. The most interesting quote is reproduced below, otherwise, I don't there's a lot new here for regular readers of this blog.

What I found more interesting were the comments on NASA Watch about the MSL problems. Most are thoughtful, and several points of view are present.

The best quote from the WP article: "Richard Cook, the project manager, said that in calculating the cost and the amount of time necessary for designing the mission, 'we didn't extrapolate how much more complex it was' than the Spirit and Opportunity mission."

Editorial comments: I have been involved with multi-hundreds of millions of dollars private industry technology development programs. Based on that experience, what I see with MSL is nothing new -- it seems to be a rule of human nature that we can't extrapolate effort required for programs that are significantly more complicated than their predecessors. (And it's not fun being the one to explain to customers who've already invested their R&D dollars based on your schedule that, hmmm, we're having a big slip.)

In the Washington Post article, Alan Stern is quoted as saying, "We need to go to a strategy where we can access Mars frequently and take advantage of what we've already invented." After having been burned myself in cost-overuns and schedule slips of technology development I find that I have developed a liking for incrementalism. I think we could have advanced our knowledge of Mars significantly without taking on the risk of MSL.

However, smart people disagree with me (and some of them actually got to make this decision). There is also an argument that along with science one of NASA's missions is to develop new technology. If big technology missions are going to be undertaken, what is the best way to do so so they don't eat the budget? Based on my experience (and please comment if your experience leads you to a different conclusion), I throw out two ideas:

1) Could the design phase have been significantly lengthened so that the full scope of the design effort could have been better known. MSL received its formal go-ahead in September 2006 (although I'm sure a lot of work had already been done) for a 2009 launch. An alternative would have been to keep the same start date but plan from the beginning to launch in 2011 and use the extra time to better identify problems. In the Planetary Sciences Subcommittee reviewing the MSL slip, someone asked how the Juno Jupiter mission development was doing. The answer was that it was rock solid because, thanks to a schedule replan, the upfront design period had been much longer than normal.

2) Perhaps we could recognize that big development efforts are likely to slip and create a budet item for them. If the planetary program can afford (to pick an arbitrary figure) $300M a year to put towards Flagship missions, then spend that much. If the mission requires more than intially planned, delay the launch and use the money from this line item.

I am a fan of MSL and excited to think of what we will learn. I think that the knowledge and exploration will be worth the investment. I just wonder if we can change the rules of the game for these big missions so that the impact on the rest of the program is less.