Sep 27 2008
What’s Wrong with Our Manned Space Program, Part 3
“Crash programs fail because they are based on the theory that, with nine women pregnant, you can get a baby a month.” — Wernher von Braun
There is another reason NASA is less effective than they could be (and this applies to unmanned NASA as well): poor financial thinking.
Let me remind you that these are my personal opinions and not those of NASA or those of my company.
Now, I mentioned budget in my post on politics, but it bears repeating. You can’t run programs that last 20 years if you don’t have consistent funding, and that means funding that provides the money for the program when it’s needed, not “next fiscal year” or not “use it this year even if you aren’t ready.” If you commit to a program, and the program is doing as promised, you need to support the program through fruition, even with administration changes, even as times change. [Note that I agree programs that have demonstrated their inability to do as planned deserve to be canned.] I feel this inability to follow through is one reason why the Shuttle doesn’t have a viable alternative today when so many alternatives have been in the works over the past decade or so.
But it’s really more than that. There’s a great deal of penny-wise, pound-foolish going on. Take for instance, government oversight on contracts. It was explained to me that contractors are dinged for having too many spare parts for contracted work, that someone counted things like bolts and dinged the contract for having half a dozen more than necessary. As a result of this policy, even for simple, easily obtained parts, they buy minimum lots, which are much more expensive per unit than buying in bulk. Now, you might say, so what? In some cases, buying the minimum doubles and triples the price for buying five times more of an item. Still a bargain you say?
Let’s do some math. I need a dozen bolts. I can buy them in sets of 12 for $4/each: 48. I can buy 120 for $1 each. $48:$120 - a bargain. Except, one didn’t pass inspection and one broke during testing, so I need more. $48 dollars. However, I also couldn’t get any more work done while I waited for purchasing to process the order, so people, who still get paid, waited while administrators/purchasers ordered the part and the vendor pulled it together and sent it. Say, it’s a small organization, with administration spending three hours on the purchase, making it and getting it approved: minimum $45, but probably more. Meanwhile, say two mechanics were sitting on their hands or doing make work waiting for the part they needed for the say week they waited: $1600. Now, to save $72 dollars, you have spent $1743 and you only have ten spares. Recap: $165:1743 and you’d have 108 spares with no schedule hit.
And that’s with an easy-to-get part. On some unusual items, like pins for hermetically sealed connectors (which are easy to lose/bend/break) or bolts made of an unusual alloy like TZM, the wait could be months, even more than a year. Pound foolish.
Budget vagaries often require buying parts or committing to vendors early in the design process because of long wait times or money available this fiscal year rather than next. Learning important things out that should call out design changes mean either your money is wasted if the parts no longer work or you have a less than optimal design using the parts you’ve bought.
And that leads me to the most cost ineffective practice of all: going cheap. My budget is tight now, so I get what’s “cost effective” even if it doesn’t work as well. Then it breaks during testing. So I build another one. It survives during testing (but it’s weakened though no one knows it). And then it breaks in flight. So, I have to build another one, test it, fly it up whereever it goes (and, believe me, the cost to fly any item is astronomical) and the crew has to learn how to replace or repair it and then take the time out to do it. Meanwhile, the precious time in orbit is also wasted (as well as untold man-hours on the ground figuring out what to do to fix it and making it so). No equipment, if it were made of diamond encrusted platinum, is more expensive than having to correct it on orbit. Pound foolish.
If we are going to make as much out of NASA as we can, we are going to have to realize that doing something twice (or more) is never cost effective over doing it right the first time. And don’t get me started about the needless risk.
I remember once, talking to a hardware provider about a repair we were doing that we had done before and were having to do again, only now with a hurry-up, extra mission to repair. I asked, since we were sending up replacements, if we had corrected the problems that were causing them to fail so frequently. They said, “No, it would be too expensive to redesign them.” *Head bop* More expensive than adding a $300 million mission? I don’t think so. Pound foolish.
