5. Process Improvement
5.6. Case Studies
5.6.3. Catapult Case Study

5.6.3.8.

Conclusions and Next Step

Conclusions

The output and goals of this case study were:

1. Determine the most important factors.
2. Determine the best settings for the factors.
3. Determine a good predicition equation for the data.

The various dex plots and Yates analysis showed that the Arm Length (X4), the Number of Bands (X3), the Band Height (X1), and the interaction between X1, X2, X3, and X4 were the most important factors. The Yates analysis gave us the prediction equation for the data

The dex contour gave us the best settings for the two most important factors (X4 and X3). Using the block plot, we can determine the best setting for the remaining factors. These can be summarized as follows.

• Band Height (X1) = +1 = 4.75 inches
• Start Angle (X2) = -1 = 0 degrees
• Number of Bands (X3) = +1 = 2 bands used
• Arm Length (X4) = +1 = 4 inches
• Stop Angle (X5) = +1 = 80 degrees

Next Step

The next step in the analysis would be to maximize (or minimize in some cases) the value of the response variable. This normally involves the use of response surface designs. Instead of maximizing distance, we might instead be more interested in hitting a specific target value.

Full and fractional designs are typically used to identify the most important factors. These designs are then often followed by a response surface design using the identified important factors to optimize the response variable. This is a common sequence in designed experiments in engineering and scientific applications. Note the iterative nature of this approach. That is, you typically do not design one large experiment to answer all your questions. Rather you run a series of smaller experiments, each of which answers

For this particular case study, we will not show a subsequent response surface design and analyis. The JMP analysis of this data shows an appproach to hitting three specific target values.