 |
Chapter 5The Space Shuttle Challenger |
 |
Introduction | Temperature and O-Ring Failures | The Probability of Failure | Refereces and Resources
Introduction
The 25th flight of the National Aeronautics and Space
Administration (NASA)
space shuttle program took off on January 20, 1986. Just after liftoff a puff of
gray smoke could be seen coming from the right solid rocket booster.
Seventy-three seconds into the flight, the space shuttle Challenger had climbed
10 miles into the air and then exploded into a fireball. All seven astronauts
died.
The cause of the explosion was determined to be an O-ring failure in the right solid rocket booster. Cold weather was a contributing factor. In hindsight, the possible dangers of launching the shuttle in cold weather are clear. Indeed, the day before the launch, some of the engineers who had designed the solid rockets argued strongly against the launch. After a series of meetings with NASA officials, these engineers were overruled. Thus, the Challenger disaster has become a case study in the possible catastrophic consequences of poor data analysis.
The best way to understand the background of the space shuttle Challenger is to visit the U.S. National Aeronautics Space Administration itself.
1. Read The History of Flight 51-L at Kennedy Space Center to learn the history and mission.
A great deal of research has gone into the background of the explosion of the space shuttle Challenger. The most exhaustive study is Report of the Presidential Commission on the Space Shuttle Challenger Accident, known as the Rogers' Commission Report.
2. Read the Excerpts from the Rogers' Report from Johnson Space Center.
 |
Temperature and O-Ring Failures
The Shuttle solid rocket booster is assembled in three sections. Each joint between sections has a pair of rubber O-rings (a primary O-ring and a secondary O-ring) that are designed to seal the joint and prevent the escape of hot gasses.
The following table gives the temperature and the number of O-ring failures for each of the previous 24 shuttle flights. The term failure is used here in a very broad sense, and occurs whenever there is significant erosion of the O-rings at a joint or blow-by of the hot gasses at the joint. Since there are two rockets, each with three joints, the number of O-ring failures for a launch is between 0 and 6. Flight number 4 has a missing data point because the rockets were lost at sea.
| Flight Number | Date | Temperature (F) | O-Ring Failures |
|---|---|---|---|
| 1 | 04/12/81 | 66 | 0 |
| 2 | 11/12/81 | 70 | 1 |
| 3 | 03/22/82 | 69 | 0 |
| 4 | 06/27/82 | 80 | * |
| 5 | 11/11/82 | 68 | 0 |
| 6 | 04/04/83 | 67 | 0 |
| 7 | 06/18/83 | 72 | 0 |
| 8 | 08/30/83 | 73 | 0 |
| 9 | 11/28/83 | 70 | 0 |
| 10 | 02/03/84 | 57 | 1 |
| 11 | 04/06/84 | 63 | 1 |
| 12 | 08/30/84 | 70 | 1 |
| 13 | 10/05/84 | 78 | 0 |
| 14 | 11/08/84 | 67 | 0 |
| 15 | 01/24/85 | 53 | 3 |
| 16 | 04/12/85 | 67 | 0 |
| 17 | 04/29/85 | 75 | 0 |
| 18 | 06/17/85 | 70 | 0 |
| 19 | 07/29/85 | 81 | 0 |
| 20 | 08/27/85 | 76 | 0 |
| 21 | 10/03/85 | 79 | 0 |
| 22 | 10/30/85 | 75 | 2 |
| 23 | 11/26/85 | 76 | 0 |
| 24 | 01/12/86 | 58 | 1 |
As always, we face the problem of accurately measuring variables.
3. Do you see any problems with the measurement of the Failure variable?
4. Construct the frequency distribution and sketch the histogram for the number of O-ring failures.
5. Sketch the bar graph that shows the number of O-ring failures as a function of flight number. Does this graph given any useful information on the possible relationship between temperature and O-ring failure?
6. Sketch the scatterplot for the two variables temperature and O-ring failures. Does this graph suggest a possible link between temperature and O-ring failures?
7. Do you think that the relationship suggested in the scatterplot can be extrapolated to a temperature of 31º (the approximate temperature on the day that Challenger was launched)? Discuss the potential problems with such extrapolation.
|
The Probability of Failure
In the conclusion of the Rogers' Report, recommendations are given in order to avert a future similar incident. The importance of knowledge and control is emphasized, and thus, it is essential that the probability of failure be estimated.
8. View the graph NASA Space Shuttle O-Ring Failures at York University Statistical Consulting Service.
The analysis used for the graph in Exercise 8 is called logistic regression. Even though you have not covered this technique in class, you can perhaps understand the information in the graph in a general way.. The black curve shows the predicted probability of failure at the low 31º temperature at the time of the launch. The red curves show 95% confidence bands for the probability values. When there is little data, the confidence regions are wide.
9. In the graph cited in the last exercise, note how the estimated probability and the width of the confidence bounds vary with temperature.
One of the members of the team investigating the explosion was the Nobel Prize winning physicist Richard Feynman. In Feynman's personal observations he points out that
It appears that there are enormous differences of opinion as to the probability of a failure with loss of vehicle and of human life. The estimates range from roughly 1 in 100 to 1 in 100,000. The higher figures come from the working engineers, and the very low figures from management. What are the causes and consequences of this lack of agreement? Since 1 part in 100,000 would imply that one could put a Shuttle up each day for 300 years expecting to lose only one, we could properly ask "What is the cause of management's fantastic faith in the machinery?"
10. Verify Feynman's computation: a failure probability of 1/100,000 corresponds to an average of one failure every 300 years, launching a Shuttle each day.
11. Comment on Feynman's last question: "What is the cause of management's fantastic faith in the machinery?".
|
References and Resources
1. Kennedy Space Center > History of Mission 51-L
2. BBC Online > The Challenger Shuttle
3. Johnson Space Center > Report of the Presidential Commission on the Space Shuttle Challenger Accident
4. York University Statistical Consulting Service > Complex Analysis on the O-Rings
5. The Virtual School > Feynman's Personal Observations
6. Edward Tufte, Visual Explanations: Images and Quantities, Evidence and Narrrative, Graphics Press, 1997.
|