ME470

Design Processes and Methodology

DOE Laboratory Crossbow Launcher
Objective:
The objective of this laboratory is to demonstrate your ability to determine if a process is in control,
conduct a DOE, and then analyze and use the results of the DOE experiment.

Background:
Each group will have one launcher. These launchers have several factors that can be varied:
1. Projectile type
2. Type of rubber band or spring
3. Pull back distance
4. Launch angle
5. Rubber band attachment

In order to complete all requirements for this laboratory within the allotted time, we have made some
decisions. The type of rubber band will remain constant.
Control Chart
Before we can use the launchers, we want to make sure that they are in control as defined in
statistical process control. We know there will be common cause variation that will result in some
variability of performance. We are looking for special cause problems in the operation of our device.
Thus, we will want to run a number of tests at one setting and use Minitabs capability to look for special
cause issues.

Using the settings supplied by your instructor, launch the projectile 30 times and enter the distances
obtained into a spreadsheet for analysis using Statistical Process Control. Use Minitab to answer Is
your process in control? (use all I-MR Options, Tests from Mini-Tab to check for special causes)
Show your results to your instructor before continuing.

Screening DOE
Use your Minitab file that with a randomized, two replicate, full factorial test with response of distance,
and four factors at two levels. A screening DOE will be performed to determine which of the factors
below are significant at a 95% confidence level. The four factors and their high/low values are listed
1. Ball type, (wiffle (holes) ball or practice (no holes) golf ball)
2. Rubber band attachment point (Pin 2 or Pin 4)
3. Pull back distance (Ruler hole 4 or 7)
4. Launcher angle (20 and 30 degrees)

Be careful to minimize variation in your set up. Try to use the same launch release each time. Use the
supplied aluminum foil to pinpoint the landing location of your projectile.

After you have taken the required data, analyze your results to determine which factors are significant.
Predictive Model (Equation)
Discuss your results with your instructor before constructing your predictive model (Equation). After
discussion with your instructor, construct the best model possible for your data (retain all terms with p
less than or equal to 0.2). Once you have constructed that model, you must use it to perform two tasks:
1. Determine the best settings to throw your selected projectile 36 inches.
2. Determine the distance that your selected projectile will be thrown if you use the practice ball
rubber band attachment Pin 4, Ruler hole 7, and launcher angle 20 degrees.

When you have you answered those questions, call your instructor over to watch as you do your
verification runs.

Required Submission:
Control Chart: State the goal of this experiment and report results including:
1. Data table with your raw data for the control chart.
2. Control chart with text discussing and justifying if the process is in control or not.
DOE: State the goal of this experiment and report results including:
3. Data table with your raw data for each treatment combination of the DOE.
4. Pareto chart from your screening run.
5. Main effects plot from your screening run.
6. Predictive equation obtained from best analysis of your data (indicate if coded or uncoded).
7. Settings that you used to throw your projectile the instructor specified distance. Actual distance
obtained for 3 throws.
8. Distance predicted for settings from step 2 of the Predictive Model. Actual distance obtained
for 3 throws.
9. Discussion of your results including
a. What factors, if any, were significant?
b. What interactions, if any, were significant?
c. What is the approximate uncertainty or your predictive model (equation), that is, the
prediction intervals? (Ask instructor if you dont know about prediction intervals)