 |
  File size : 5.6MB Format : MPEG |
 |
Q. What is your opinion or advice on the recent failure of JAXA's H-IIA or Nozomi project? President Kennedy said, "Victory has a hundred fathers, defeat is an orphan." First, I would say that you should all remember that those of us in other space programs who've also suffered failures share your anguish. It's a very painful process. And there's no easy way to get through it except to do what needs to be done. So the question is, what needs to be done? I think in all countries, the most important thing is to look at the causes of the failures objectively. And to do it in a way that individuals are not necessarily harmed but the facts really come out. To some extent that's easier to do in the United States, I think, than in Japan, because the United States cultural attitude is adversarial. We argue in public. That's okay. It's not considered bad manners. People don't feel hurt if somebody challenges their idea publicly. Japan's culture has evolved in a different way to reduce public conflict. And the consequence of that is the Japanese need to find culturally acceptable ways to have this discourse about what went wrong and why, and do it in a way that doesn't injure the individuals involved or the institutions, but really does get out the facts, and most important, the underlying circumstances. |
|
|||
 File size : 14.5MB Format : MPEG |
|
Q. You also have several experiences of accidents in past NASA Mars missions, which you were involved with. How did you analyze the information from those accidents? Well, I had the mixed pleasure of being a scientist involved with the Mars Polar Lander and also with the Deep Space II microprobe, a penetrator for Mars, both of which were supposed to land on December 3, 1999. And I was also going to play a role on the Mars climate orbiter. It was to be a wonderful period of time. Unfortunately the Mars climate orbiter failed in October, due to bad internal communications within both JPL and Lockheed Martin, the main contractor. Mars Polar Lander failed on December 3rd due to a software/hardware problem, a rather subtle one, that even so should have been caught in the reviews and testing. And Deep Space II failed for reasons we're not sure of. It may have been due to Mars' surface environment being more difficult than we had expected. It may have actually crashed into very cold and hard ice right at the surface that we hadn't anticipated and it wasn't strong enough to survive such a high shock. So December 3, 1999 was a bad day indeed. But it got worse for me, because I was asked by the then head of NASA to serve for three months on the investigation board that looked into those failures in great detail. So I ended up knowing a lot about those missions and their failures. In English we use the term "proximate cause" for the immediate cause of the failure. And then there is the big cause behind it, which may be the most important conclusion from a failure report. In the case of the proximate cause for Mars Polar Lander, it was a subtle problem between the hardware and software and the way the landing system deployed at the end. That should have been caught, in this case at the contractor's facility in Boulder, Colorado, and it wasn't. The key person on this subject for both of those reviews couldn't be in two places at once. Thus the one person who had enough knowledge to see that there might be a problem wasn't available to do it. Why? Because the amount of money made available was so low that both JPL and the contractor had to cut their manpower low. Too low. Well, you should ask, why was that? And the answer is, you never really know how much a new development in space is going to cost, and thus there should be adequate cost reserves, which wasn't the case. The tragedy in this case, the real cause behind the proximate cause, was that JPL management did not go back to NASA and say forcefully, "look, we are sorry, we told you it was going to cost this much; it's actually going to cost a bit more for us to do it safely." They didn't do that. They are the responsible parties in my view. And I've told them this and so did the commission that investigated it. A similar problem prevailed in the Mars climate orbiter failure, and may also have affected the DS II. So the answer from that failure review board was that part of the problems were institutional, getting people to think about things differently. But that finding makes us realize that the real culprit, the underlying cause of this was a mindset, an attitude that there was something unchallengeable about the project. The cost cap was absolutely non-negotiable, as far as JPL top management were concerned. They had committed to an unprecedentedly low total cost while feeling great pressure from NASA to lower costs significantly below what even the successful Pathfinder and Mars Global Surveyor cost, which in turn were much, much less expensive than the Viking landers of 1976. Even when serious concerns about reliability were expressed by both JPL project personnel and the contractor leadership, the JPL leadership resisted going back to NASA. That was the primary mistake. I believe a similar mindset has beset the Shuttle program. The Shuttle program has had a doctrine, a mindset that underlay both the Columbia and the Challenger disasters. There was the belief - the doctrine - that Shuttle could be made to be a safe operational vehicle. In fact, as the Columbia Accident Investigation Board concluded, Shuttle can never be made safe. It was a great R&D success, but it could never have been a safe, reliable operational vehicle. There were many indications of that intrinsic risk, but the doctrine was, "We can fly this. We should do it." If somebody stood up and said, "Hey, I've got concerns," the managers said, "Sorry, we have a schedule to meet, we're going to do it." And that's why those Shuttle failures took place. I think this underlying pattern of doctrinal attitudes in the U.S. failures is significant to Japan because one has to look both at the proximate cause - what really caused H-IIA to fail, what really caused Nozomi or Adeos to fail? - and then, "Well, why wasn't that fixed? Why wasn't that dealt with?" And that gets you back into the ideas that were in people's heads. Thus, probably the most important thing for JAXA is to try to understand that and then change it.
|
