Tuesday, March 23, 2004

The Space Shuttle and Risk

Saw an interesting article in the New York Times today about a problem they found with the Space Shuttle. Twenty years ago a contractor installed the gears wrong on the actuator for the rudder/speed-brake. Skip the following paragraph if you know what a rudder/speed-brake is.

The rudder is the control surface on the vertical tail of airplane. It controls the yaw motion; to visualize this, if your head was an airplane the yaw motion would be like shaking it no; it's a side-to-side motion. The rudder is primarily used to keep the airplane flying straight if there is a cross-wind. On the Shuttle, however, this surface performs double-duty and is also a speed-brake. In this instance, the rudder would split into two parts like the opening of a clam-shell, increasing the drag and slowing the vehicle.

Fortunately the improper installation did not cause any accidents because the part was never stressed to its limits. This is an example of a good design. It is important in designing an aerospace vehicle in particular to prepare for the worst-case situation that could be encountered. In structural design there are safety factors that are put on everything to account for the unexpected as well. In this instance it is likely that the additional margin provided by the safety factor kept the part from failing and killing everyone in the Shuttle. It's also a failure of design because it was possible to install the gears in reverse in the first place. If it had been designed such that the gears would only fit one-way the problem would have been avoided entirely.

I also noticed one of the NASA guys said that the misinstallation "could have resulted in a catastrophe if there had been unusual stress, like a launching problem that called for an emergency landing" which made me snort a little bit. If something goes wrong in the launch, the crew is essentially dead. Even if they are not blown up the volatile fuel surrounding them, the NASA plan for an aborted launch is to flip the Shuttle such that it is flying backwards and then fire the engines to head back to Florida for a landing. The chance of this maneuver being conducted successfully is virtually nil. Astronauts know this but fly anyway because they view the risk as acceptable.

People have unrealistic expectations of safety and generally misunderstand risk. All activities in life have a certain level of risk associated with them. One of the riskiest things you do almost everyday is drive a car. But there is a risk in taking a shower in the morning because you may slip and fall and die. There is risk involved in doing, or not doing, anything. You cannot completely eliminate the chance that something bad will happen, but it is important to consider risk. Everyone has a risk management strategy, even if it is just to ignore risk.

Everyone has their own risk threshold and it tends to get lower as we age. Consider the teenager who likes to race his friends, compared to Grandpa in his tank of a Cadillac. The perception of risk is also tainted by what we see in the news. This is why people think (or at least feel) that driving a car is safer than riding in an airplane. It may be that flying in the Shuttle is too risky for the vast majority of people, but if there still remains a cadre of people willing to fly despite the risk it is foolish to burden the Shuttle with expectations of a perfect flying record.

The level of risk must also be balanced against the cost it takes to reduced that risk. It might be possible to build a vehicle with a risk of essentially zero, but it would be so heavy and expensive as to be useless. Part of deciding where to place the line in deciding acceptable risk is in placing a value, however callous that sounds, on a human life. Incidentally this value is usually calculated to be about $1.5 million.

Does this mean that I support the Space Shuttle? No, I think it's an ancient piece of crap that is long over-due for replacement but hasn't yet been replaced because of a lack of political will. I also think it's foolish for us to expect that our space exploration will be without risk, or at least that nothing will actually happen. Perhaps someday the risk of going to space will be no more than flying a plane, but that day is far off into the future.

No comments: