Design of Experiment
This workshop utilizes TURN9DOE, an MS EXCEL-based Visual Basic program, to simulate a
turning operation that is being used to manufacture a bronze bushing (Fig. 1). The goal of the
workshop is to apply the general class of 2k-p fractional factorial designs to improve the centering
of two lathes that perform the turning operations in order to improve the process capability.
Learning Objective
The objective of this computer workshop is to give you practice in the planning, conducting, and
interpreting designed experiments. In addition, it is expected that you will need to combine the
results of fractional factorial designs to unconfound variable interactions. Finally, based on the
experiment analysis the student is asked to determine and validate variable levels that will
improve the process capability.
Figure 1: Bronze bushing.
Background
The turning process producing a bronze bushing was studied using statistical process control
with the finish of the outside surface serving as the quality characteristic of interest. After
corrective actions were taken to eliminate special causes, the X and R charts for each machine
showed good statistical control. Investigation of process capability for the individual machines
revealed that two problems still exist:
1. Only one of the three machines is operating at the nominal surface roughness (Ra); the
specification is 70 ± 15 microinches.
2. All three of the machines exhibit excessive variability in surface roughness values.
The process engineer working on this problem had recently become acquainted with the
techniques of statistical design of experiments. He decided to try to apply these techniques in an
attempt to learn more clearly what factors might influence surface roughness, either
independently, or in concert with each other, or both. He met with several individuals associated
with the process and they developed a list of potentially important factors together with the
feasible ranges of variation for these factors. The table on the next page provides this list of
factors. The table also provides the current state of affairs with respect to each of these factors.
Description of TURN9DOE
The TURN9DOE program has been developed to simulate the act of running experiments with
the nine process variables shown in the table below. The program produces a surface roughness
value as a test result for any given combination of process settings you specify.
Factor Feasible Range for Experiments Current Operating Condition
Machine Used Only the Nacirema and LeLathe
lathes can be used for
experimentation
Nacirema, LeLathe, Rex
Use of Cutting Fluid Either present or absent No cutting fluid is being used
Cutting Speed 900 – 1150 sfpm 1000 – 1100 sfpm
Feed Rate 0.0070 – 0.0095 ipr 0.008 – 0.0089 ipr
Tool Nose Radius 0.0156” to 0.0625” in 0.0156”
increments
Fixed at 0.0312”
Tool Material Two grades (A & B) of cemented
carbide tools are applicable
Grade A is being used
Tool Back Rake Angle 0 degrees to 30 degrees Varies between 5 and 10
degrees positive
Tool Side Cutting Edge
Angle
5 degrees to 30 degrees Fixed at 10 degrees
Depth of Cut for Final
Pass
0.005” to 0.030” Varies between 0.005” and
0.015”
Assignment
The process engineer needs guidance in designing the tests and analyzing the data. He calls on
you for help. Use the general classes of 2k and/or 2k-p factorial and fractional factorial designs
to discover what changes might be initiated to improve the two off-target machines.
The following guidance may be useful:
1. You should be concerned only with process centering at this point, i.e., we will work only
with location effects.
2. You may experiment with all (9) factors or with any subset you choose – it’s all up to you.
3. You should run no more than about 30-50 experiments in the entire study.
4. You should think in terms of running several smaller and perhaps related experiments as
opposed to one big experiment.
5. You are free to choose the high and low levels for the experiments anywhere within the
feasible range – again, it’s all up to you.
Ultimately, you will determine the processing conditions that will center the machines about the
nominal surface roughness value (70 µ in.).
To complete the assignment, the following tasks must be undertaken:
1. Design your experiment. The levels of the variables are up to you, but within the guidelines
established above. Multiple designs may need to be performed to unconfound the variable
interactions associated with a highly fractionated design.
2. Run TURN9DOE to collect sample values for each set of levels.
3. Establish a model for the process based on the analysis of your results.
4. Use plotting software to characterize the response surface(s) and levels for the processing
variables that center each machine.
5. Perform validation tests to confirm that the identified processing conditions do indeed result
in centered machines.
6. Prepare a report that details the findings of your study.
The report, describing what you have learned about the process, must include at least the
following:
1. An executive summary, which summarizes the contents and conclusions of your report in
approximately one page.
2. A body of the report, which includes the following components:
a. An introductory section, which briefly describes the problem at hand and indicates the
methodology to be employed in determining the recommended operating conditions.
b. A procedures section, which describes in detail the results of your investigation. Be sure
to list all experiments you conducted, the data obtained, and the associated analysis.
Don’t forget to indicate the logic used in your evolving design of experiments strategy.
Include graphs of the relevant response surface(s). Carefully explain your
recommendations and the logic associated with them. Show the results of confirmatory
tests. You should state clearly how you arrived at any conclusions.
3. A conclusions section, which summarizes your study and its findings.
4. An appendix section, which includes supporting material (e.g., data and calculations).
ORDER THIS ESSAY HERE NOW AND GET A DISCOUNT !!!
Place an order today and get 13% Discount (Code GAC13)
