HW 4

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

CSCI 688

Homework 4

Megan Rose Bryant


Department of Mathematics
William and Mary

October 6, 2014
4.20 The effect of five different ingredients (A,B,C,D,E) on the reaction time of a chemical process is being
studied. Each batch of new material is only large enough to permit five runs to be made. Furthermore, each
run requires approximately 1 12 hours, so only five runs can be made in one day. The experimenter decides
to run the experiment as a Latin square so that day and batch effects may be systematically controlled. She
obtains the data that follow. Analyze the data from this experiment (use α = 0.05) and draw conclusions.

Day
Batch 1 2 3 4 5
1 A=8 B=7 D=1 C=7 E=3
2 C = 11 E=2 A=7 D=3 B=8
3 B=4 A=9 C = 10 E =1 D=5
4 D=6 C=8 E=6 B=6 A = 10
5 E=4 D=2 B=3 A=8 C=8

General Linear Model: Reaction Time versus Batch, Day, Ingredient

Method

Factor coding (-1, 0, +1)

Factor Information

Factor Type Levels Values


Batch Random 5 1, 2, 3, 4, 5
Day Random 5 1, 2, 3, 4, 5
Ingredient Fixed 5 1, 2, 3, 4, 5

Analysis of Variance

Source DF Adj SS Adj MS F-Value P-Value


Batch 4 15.44 3.860 1.23 0.348
Day 4 12.24 3.060 0.98 0.455
Ingredient 4 141.44 35.360 11.31 0.000
Error 12 37.52 3.127
Total 24 206.64

Model Summary

S R-sq R-sq(adj) R-sq(pred)


1.76824 81.84% 63.69% 21.19%

Based on the obtained minitab results, we can conclude that the chemical ingredients have a significant effect
the reaction time because it has a p-value of 0. Also, we can conclude that the effects of batch and day are
negligible on reaction time (once they are controlled for) because they both have p-values greater than our
α of .05. The normal probability plot and the residual graphs below do not give us any reason to question
our normality or identical independent assumptions.
4.33 An engineer is studying the mileage performance characteristics of five types of gasoline additives.
In the road test he wishes to use cars as block; however, because of a time constraint, he must use an
incomplete block design. He runs the balanced design with the five blocks that follow. Analyze the data from
this experiment (use α = 0.05) and draw conclusions.

Day
Additive 1 2 3 4 5
1 17 14 13 12
2 14 14 13 10
3 12 13 12 9
4 13 11 11 12
5 11 12 10 8
General Linear Model: Mileage versus Car, Additive

Method

Factor coding (-1, 0, +1)


Rows unused 5

Factor Information

Factor Type Levels Values


Car Random 5 1, 2, 3, 4, 5
Additive Fixed 5 1, 2, 3, 4, 5

Analysis of Variance

Source DF Adj SS Adj MS F-Value P-Value


Car 4 35.23 8.8083 9.67 0.001
Additive 4 35.73 8.9333 9.81 0.001
Error 11 10.02 0.9106
Total 19 76.95

2
Model Summary

S R-sq R-sq(adj) R-sq(pred)


0.954257 86.98% 77.52% 56.97%

Based on the obtained Minitab output, we can conclude that both the car and the gasoline additive have
a statistically significant affect on the mileage as both have p-values of 0.01. The normal probability plot
and the residual graphs included below do not give us any reason to question the normality or identically
independently distributed assumptions. Though it is woth noting that there seems to be some mild grouping
in the NPP.

4.42 Verify that a BIBD with the parameters a = 8, r = 8, k = 4, and b = 16 does not exist.
We know that for a BIBD with those parameters to exist, there must exist an integer λ such that

r(k − 1) 8(4 − 1) 24
λ= = = 6∈ Z
a−1 8−1 7
Therefore, a BIBD does not exist with those parameters.

5.3 The yield of a chemical process is being studied. The two most important variables are thought to be
the pressure and the temperature. Three levels of each factor are selected and a factorial experiment with
two replicates is performed. The yield data are as follows.

Pressure (psig)
Temperature ◦ C 200 215 230
150 90.4 90.7 90.2
90.2 90.6 90.4
160 90.1 90.5 89.9
90.3 90.6 90.1
170 90.5 90.8 90.4
90.7 90.9 90.1

3
a.) Analyze the data and draw conclusions. Use α− = 0.05.
General Linear Model: Yield versus Temp, Pressure

Method

Factor coding (-1, 0, +1)

Factor Information

Factor Type Levels Values


Temp Fixed 3 1, 2, 3
Pressure Fixed 3 1, 2, 3

Analysis of Variance

Source DF Adj SS Adj MS F-Value P-Value


Temp 2 0.30111 0.15056 8.47 0.009
Pressure 2 0.76778 0.38389 21.59 0.000
Temp*Pressure 4 0.06889 0.01722 0.97 0.470
Error 9 0.16000 0.01778
Total 17 1.29778

Model Summary

S R-sq R-sq(adj) R-sq(pred)


0.133333 87.67% 76.71% 50.68%

Based on the obtained Minitab output, we would conclude that both pressure and temperature are significant,
since they have p-values that are less than our α of 0.05. However, we would also conclude that the interaction
between pressure and temperature is not significant since the p-value is 0.47, which is larger than our α.

b.) Prepare appropriate residual plots and comment on the model’s adequacy.
The normal probability and residual plots do not give us significant cause to question the normality or
identically distributed assumptions, though it is important to note that there seems to be more of a variation
in the third batches of both temperature and pressure.

4
c.) Under what conditions would you operate this process?
Comparisons for Yield

Tukey Pairwise Comparisons: Response = Yield, Term = Temp

Grouping Information Using the Tukey Method and 95% Confidence

Temp N Mean Grouping


3 6 90.5667 A
1 6 90.4167 A B
2 6 90.2500 B

Means that do not share a letter are significantly different.

Tukey Pairwise Comparisons: Response = Yield, Term = Pressure

Grouping Information Using the Tukey Method and 95% Confidence

Pressure N Mean Grouping


2 6 90.6833 A
1 6 90.3667 B
3 6 90.1833 B

Means that do not share a letter are significantly different.

We see from our Tukey Pairwise Comparisons for Yield that the process should be operated using temperature
3: 170◦ and pressure 2: 215 psig for the highest yield. It is important to note that Tukey did not find a
significant difference between temperature 1 and temperature 3 or temperature 1 and temperature 2, however
we will select temperature 3 both because it has the highest mean yield and because it is statistically
significantly different from temperature 2. This selection is also supported by the contour plot included
below.

5
5.4 An engineer suspects that the surface finish of a meat part is influenced by the feed rate and the depth
of cut. He selects three feed rates and four depths of cut. He then conducts a factorial experiment and obtains
the folowing data:

Feed Rate Depth of Cut (in)


(in/min) 0.15 0.18 0.20 0.25
74 79 82 99
0.20 64 68 88 104
60 73 92 96
92 98 99 104
0.25 86 104 108 110
88 88 95 99
99 104 108 114
0.30 98 99 110 111
102 95 99 107

a.) Analyze the data and draw concusions. Use α = 0.05.


General Linear Model: Surface Finish versus Feed Rate, Depth of Cut

Method

Factor coding (-1, 0, +1)

Factor Information

Factor Type Levels Values


Feed Rate Fixed 3 1, 2, 3
Depth of Cut Fixed 4 1, 2, 3, 4

Analysis of Variance

6
Source DF Adj SS Adj MS F-Value P-Value
Feed Rate 2 3160.5 1580.25 55.02 0.000
Depth of Cut 3 2125.1 708.37 24.66 0.000
Feed Rate*Depth of Cut 6 557.1 92.84 3.23 0.018
Error 24 689.3 28.72
Total 35 6532.0

Model Summary

S R-sq R-sq(adj) R-sq(pred)


5.35931 89.45% 84.61% 76.26%

Based on the obtained Minitab output, we can conclude that both Feed Rate and Depth of Cut are significant
since they have p-values of less than our α of 0.05. We can also declare the interaction between Feed Rate
and Depth of Cut to be significant as it has a p-value of 0.018, which is also less than α.

b.) Prepare appropriate residuals plots and comment on the model’s adequacy.
The normal probability and residual plots do not give us any reason to doubt our normality and independent
identical distribution assumptions.

c.) Obtain point estimates of the mean surface finish at each feed rate.
Fitted
Term Mean SE Mean
Feed Rate
1 81.58 1.55
2 97.58 1.55
3 103.83 1.55
Depth of Cut
1 84.78 1.79
2 89.78 1.79
3 97.89 1.79
4 104.89 1.79

7
We are only interested in the point estimates of the mean surface finish at each feed rate, so we will use a
one-factor plot.

d.) Find the P-values for the tests in part (a).


As obtained in part a.), the p-values are as follows:

Factor P-Value
Feed Rate 0.000
Depth of Cut 0.000
Feed Rate * Depth of Cut 0.018

5.5 For the data in problem 5.4 compute a 95 percent confidence interval estimate of the mean difference
in response for feed rates of 0.20 and 0.25 in/min.

Tukey Pairwise Comparisons: Response = Surface Finish, Term = Feed Rate

Grouping Information Using the Tukey Method and 95% Confidence

Feed
Rate N Mean Grouping
3 12 103.833 A
2 12 97.583 B
1 12 81.583 C

Means that do not share a letter are significantly different.

Tukey Simultaneous Tests for Differences of Means

Difference
of Feed Difference SE of Simultaneous Adjusted
Rate Levels of Means Difference 95% CI T-Value P-Value
2 - 1 16.00 2.19 (10.54, 21.46) 7.31 0.000
3 - 1 22.25 2.19 (16.79, 27.71) 10.17 0.000

8
3 - 2 6.25 2.19 ( 0.79, 11.71) 2.86 0.023

Individual confidence level = 98.02%

We are only interested in a 95% confidence interval for the mean difference between levels 1 and 2 of Feed
Rate. Therefore, our desired confidence interval is (10.54, 21.46) e.g. 16 + / − 5.46.

5.20 In Problem 5.3, suppose that we wish to reject the null hypothesis with a high probability if the
difference in the true mean yield at any two pressures is as great as 0.5. If a reasonable prior estimate of
the standard deviation of yield is 0.1, how many should be run?
We know that

a = 3b = 3D = 0.5
Therefore, we can calculate the desired sample size by finding the minimum value of φ2 .

naD2 n ∗ 3 ∗ (0.5)2
φ2 = 2
= = 12.5n
2bσ 2 ∗ (3) ∗ (0.1)2
For a sample size of n = 2, we see that we have a φ = 5. Using the given formulas and the charts available
in Appendix V (the Operating Charecteristic Curves. . . ) we can obtain the following:

n φ2 φ ν1 ν2 β
2 25 5 2 9 ¡ .05
For ν1 = 2 and ν2 = 9, we have a β < 0.05 on the OCC graph. Therefore, a sample size of 2 is sufficient to
detect a difference of 0.5 between pressures.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy