ch03 PDF

Chapter 3
Modeling Process Quality
LEARNING OBJECTIVES
After completing this chapter you should be able to:
1. Construct and interpret visual data displays, including the stem-and-leaf plot, the histogram, and
the box plot
2. Compute and interpret the sample mean, the sample variance, the sample standard deviation,
and the sample range
3. Explain the concepts of a random variable and a probability distribution
4. Understand and interpret the mean, variance, and standard deviation of a probability
distribution
5. Determine probabilities from probability distributions
6. Understand the assumptions for each of the discrete probability distributions presented
7. Understand the assumptions for each of the continuous probability distributions presented
8. Select an appropriate probability distribution
9. Use probability plots
10. Use approximations for some hypergeometric and binomial distributions
IMPORTANT TERMS AND CONCEPTS

Approximations to probability distributions Percentile
Binomial distribution Poisson distribution
Box plot Population
Central limit theorem Probability distribution
Continuous distribution Probability plotting
Descriptive statistics Quartile
Discrete distribution Random variable
Exponential distribution Run chart
Gamma distribution Sample
Geometric distribution Sample average
Histogram Sample standard deviation
Hypergeometric probability distribution Sample variance
Interquartile range Standard deviation
3-2 CHAPTER 3 MODELING PROCESS QUALITY
Lognormal distribution Standard normal distribution

Mean of a distribution Statistics
Median Stem-and-leaf display
Negative binomial distribution Time series plot
Normal distribution Uniform distribution
Normal probability plot Variance of a distribution
Pascal distribution Weibull distribution
EXERCISES
New exercises are marked with .
3.1.
The content of liquid detergent bottles is being analyzed. Twelve bottles, randomly selected from the
process, are measured, and the results are as follows (in fluid ounces): 16.05, 16.03, 16.02, 16.04, 16.05,
16.01, 16.02, 16.02, 16.03, 16.01, 16.00, 16.07.
(a) Calculate the sample average.

x   xi n  16.05  16.03   16.07 12  16.029 oz
n
i 1
(b) Calculate the sample standard deviation.
 
n n 2
 xi   xi
2
n
i 1 i 1 (16.052   16.072 )  (16.05   16.07)2 12
s   0.0202 oz
n 1 12  1
MTB > Stat > Basic Statistics > Display Descriptive Statistics
Descriptive Statistics: Ex3-1
Variable N N* Mean SE Mean StDev Minimum Q1 Median Q3

Ex3-1 12 0 16.029 0.00583 0.0202 16.000 16.012 16.025 16.047
Variable Maximum
Ex3-1 16.070
CHAPTER 3 MODELING PROCESS QUALITY 3-3
3.2.
The bore diameters of eight randomly selected bearings are shown here (in mm): 50.001, 50.002,
49.998, 49.997, 50.000, 49.996, 50.003, 50.004

x   xi n   50.001  49.998   50.004  8  50.000 mm
n
i 1
 
n n 2
 xi   xi
2
n
i 1 i 1 (50.0012   50.0042 )  (50.001   50.004)2 8
s   0.0029 mm
n 1 8 1

Ex3-2 8 0 50.000 0.00103 0.00290 49.996 49.997 50.001 50.003
Variable Maximum
Ex3-2 50.004
3.3. 
The service time in minutes from admit to discharge for ten patients seeking care in a hospital
emergency department are 21, 136, 185, 156, 3, 16, 48, 28, 100, and 12. Calculate the mean and
standard deviation of the service time.
Variable N N* Mean SE Mean StDev Minimum Q1 Median Q3 Maximum

Ex3-3 10 0 70.5 21.4 67.7 3.0 15.0 38.0 141.0 185.0
3.4. 
The Really Cool Clothing Company sells its products through a telephone ordering process. Since
business is good, the company is interested in studying the way that sales agents interact with their
customers. Calls are randomly selected and recorded, then reviewed with the sales agent to identify
ways that better service could possibly be provided or that the customer could be directed to other
items similar to those they plan to purchase that they might also find attractive. Call handling time
(length) in minutes for 20 randomly selected customer calls handled by the same sales agent are as
follows: 5, 26, 8, 2, 5, 3, 11, 14, 4, 5, 2, 17, 9, 8, 9, 5, 3, 28, 22, and 4. Calculate the mean and standard
deviation of call handling time.
Variable N N* Mean SE Mean StDev Minimum Q1 Median Q3 Maximum

Ex3-4 20 0 9.50 1.77 7.92 2.00 4.00 6.50 13.25 28.00
3.5.
The nine measurements that follow are furnace temperatures recorded on successive batches in a
semiconductor manufacturing process (units are °F): 953, 955, 948, 951, 957, 949, 954, 950, 959

x   xi n   953  955   959  9  952.9 °F
n
i 1
 
n n 2
 xi   xi
2
n
i 1 i 1 (9532   9592 )  (953   959)2 9
s   3.7 °F
n 1 9 1

Ex3-5 9 0 952.89 1.24 3.72 948.00 949.50 953.00 956.00
Variable Maximum
Ex3-5 959.00
3.6.
Consider the furnace temperature data in Exercise 3.5.
(a) Find the sample median of these data.
In ranked order, the data are {948, 949, 950, 951, 953, 954, 955, 957, 959}. The sample median is the
middle value.
(b) How much could the largest temperature measurement increase without changing the sample
median?
Since the median is the value dividing the ranked sample observations in half, it remains the same
regardless of the size of the largest measurement.
3.7.
Yield strengths of circular tubes with end caps are measured. The first yields (in kN) are as follows: 96,
102, 104, 108, 126, 128, 150, 156
x   xi n   96  102   156  8  121.25 kN

n
i 1
 
n n 2
 xi   xi
2
n
i 1 i 1 (962   1562 )  (96   156)2 8
s   22.63 kN
n 1 8 1

Ex3-7 8 0 121.25 8.00 22.63 96.00 102.50 117.00 144.50
Variable Maximum
Ex3-7 156.00
3.8.
The time to failure in hours of an electronic component subjected to an accelerated life test is shown in
Table 3E.1. To accelerate the failure test, the units were tested at an elevated temperature (read down,
then across).
(a) Calculate the sample average and standard deviation.

(d) Find the sample median and the lower and upper quartiles.
Descriptive Statistics: Ex3-8_Table 3E.1
Variable N N* Mean SE Mean StDev Minimum Q1 Median

Ex3-8_Table 3E.1 40 0 130.05 1.43 9.06 117.00 124.00 128.00
Variable Q3 Maximum
Ex3-8_Table 3E.1 135.25 161.00
3.8. continued
(b) Construct a histogram.
Use n  40  7 bins
MTB > Graph > Histogram > Simple
(Edit X-axis scale to change binning to 7 intervals)
Histogram of Ex3-8 / Table 3E.1
20
15
Frequency
10
0
112 120 128 136 144 152 160
Ex3-8 / Table 3E.1
Note that the histogram has 6 bins. The number of bins can be changed by editing the X scale.
However, when 7 bins are specified, Minitab generates a 6-bin histogram. Constructing a 7-bin
histogram requires manual specification of the bin cut points. Recall that this formula is an
approximation, and therefore either 6 or 8 bins should suffice for assessing the distribution of the data.
3.8. continued
(c) Construct a stem-and-leaf plot.
MTB > Graph > Stem-and-Leaf
Stem-and-Leaf Display: Ex3-8 / Table 3E.1
Stem-and-leaf of Ex3-8 / Table 3E.1 N = 40

Leaf Unit = 1.0
2 11 89
12 12 0112334444
(12) 12 555556788999
16 13 011133
10 13 677
7 14 00122
2 14
2 15 1
1 15
1 16 0
3.9.
The data shown in Table 3E.2 are chemical process yield readings on successive days (read down, then
across). Construct a histogram for these data. Comment on the shape of the histogram. Does it
resemble any of the distributions that we have discussed in this chapter?
(Edit X-axis scale to change binning to 9 intervals)

20
15
Frequency
10
0
84 88 92 96 100
Ex3-9 / Table 3E.2
The distribution appears to have two “peaks” or central tendencies, at about 88 and 94. This is also
referred to as a bimodal distribution. It does not resemble any of the continuous distributions discussed
in this chapter (normal, lognormal, exponential, gamma, or Weibull).
3.10.
An article in Quality Engineering (Vol. 4, 1992, pp. 487–495) presents viscosity data from a batch
chemical process. A sample of these data is presented in Table 3E.3 (read down, then across).
(a) Construct a stem-and-leaf display for the viscosity data.
Stem-and-Leaf Plot (Unordered)

2 12o|68
6 13*|3134
12 13o|776978
28 14*|3133101332423404
(15) 14o|585669589889695
37 15*|3324223422112232
21 15o|568987666
12 16*|144011
6 16o|85996
1 17*|0
Stem Freq|Leaf
3.10. continued
(b) Construct a frequency distribution and histogram.

20
15
Frequency
10
0
13 14 15 16 17
Ex3-10 / Table 3E.3
(c) Convert the stem-and-leaf plot in part (a) into an ordered stem-and-leaf plot. Use this graph to assist
in locating the median and the upper and lower quartiles of the viscosity data.

Leaf Unit = 0.10
2 12 68
6 13 1334
12 13 677789
28 14 0011122333333444
(15) 14 555566688889999
37 15 1122222222333344
21 15 566667889
12 16 011144
6 16 56899
1 17 0
3.10.(c) continued
median observation rank is (0.5)(80) + 0.5 = 40.5

x0.50 = (14.9 + 14.9)/2 = 14.9
Q1 observation rank is (0.25)(80) + 0.5 = 20.5

Q1 = (14.3 + 14.3)/2 = 14.3
Q3 observation rank is (0.75)(80) + 0.5 = 60.5

Q3 = (15.6 + 15.5)/2 = 15.55
(d) What are the tenth and ninetieth percentiles of viscosity?
10th percentile observation rank = (0.10)(80) + 0.5 = 8.5

x0.10 = (13.7 + 13.7)/2 = 13.7
90th percentile observation rank is (0.90)(80) + 0.5 = 72.5

x0.90 = (16.4 + 16.1)/2 = 16.25
3.11.
Construct and interpret a normal probability plot of the volumes of the liquid detergent bottles in
Exercise 3.1.
MTB > Graph > Probability Plot > Simple and select Normal Distribution
Probability Plot of Ex3-1

Normal - 95% CI
99
Mean 16.03
StDev 0.02021
95 N 12
AD 0.297
90
P-Value 0.532
80
70
Percent
60
50
40
30
20
10
1
15.950 15.975 16.000 16.025 16.050 16.075 16.100
Ex3-1
When plotted on a normal probability plot, the data points tend to fall along a straight line, indicating
that a normal distribution adequately describes the volume of detergent.
3.12.
Construct and interpret a normal probability plot of the nine furnace temperature measurements in
Exercise 3.5.

Normal - 95% CI
99
Mean 70.5
StDev 67.74
95 N 10
AD 0.644
90
P-Value 0.065
80
70
Percent
60
50
40
30
20
10
1
-200 -100 0 100 200 300
Ex3-3
When plotted on a normal probability plot, the data points tend to fall along a straight line, indicating a
normal distribution adequately describes the furnace temperatures.
3.13.
Construct a normal probability plot of the failure time data in Exercise 3.8. Does the assumption that
failure time for this component is well modeled by a normal distribution seem reasonable?
Probability Plot of Ex3-8 / Table 3E.1

Normal - 95% CI
99
Mean 130.0
StDev 8.914
95 N 40
AD 1.259
90
P-Value <0.005
80
70
Percent
60
50
40
30
20
10
1
100 110 120 130 140 150 160
Ex3-8 / Table 3E.1
When plotted on a normal probability plot, the data points do not fall along a straight line, indicating
that the normal distribution does not reasonably describe the failure times.
3.14.
Construct a normal probability plot of the chemical process yield data in Exercise 3.9. Does the
assumption that process yield is well modeled by a normal distribution seem reasonable?

Normal - 95% CI
99.9
Mean 89.48
StDev 4.158
99
N 90
AD 0.956
95 P-Value 0.015
90
80
70
Percent
60
50
40
30
20
10
5
0.1
75 80 85 90 95 100 105
Ex3-9 / Table 3E.2
When plotted on a normal probability plot, the data points do not fall along a straight line, indicating
that the normal distribution does not reasonably describe process yield.
3.15.
Consider the viscosity data in Exercise 3.10. Construct a normal probability plot, a lognormal probability
plot, and a Weibull probability plot for these data. Based on the plots, which distribution seems to be
the best model for the viscosity data?

Normal - 95% CI
99.9
Mean 14.90
StDev 0.9804
99
N 80
AD 0.249
95 P-Value 0.740
90
80
70
Percent
60
50
40
30
20
10
5
0.1
11 12 13 14 15 16 17 18 19
Ex3-10 / Table 3E.3
MTB > Graph > Probability Plot > Simple and select Lognormal Distribution

Lognormal - 95% CI
99.9
Loc 2.699
Scale 0.06595
99
N 80
AD 0.216
95 P-Value 0.841
90
80
70
Percent
60
50
40
30
20
10
5
0.1
11 12 13 14 15 16 17 18 19
Ex3-10 / Table 3E.3
3.15. continued
MTB > Graph > Probability Plot > Simple and select Weibull Distribution

Weibull - 95% CI
99.9
99 Shape 16.10
Scale 15.36
90
N 80
80
70 AD 1.032
60
50 P-Value <0.010
40
30
20
Percent
10
5
3
2
0.1
9 10 11 12 13 14 15 16 17 18
Ex3-10 / Table 3E.3
Both the normal and lognormal distributions appear to be reasonable models for the data; the plot
points tend to fall along a straight line, with no bends or curves. However, the plot points on the
Weibull probability plot are not straight—particularly in the tails—indicating it is not a reasonable
model.
3.16.
Table 3E.4 contains 20 observations on cycles to failure of aluminum test coupons subjected to repeated
alternating stress of 15,000 psi at 20 cycles per second. Construct a normal probability plot, a lognormal
probability plot, and a Weibull probability plot for these data. Based on the plots, which distribution
seems to be the best model for the cycles to failure for this material?

Normal - 95% CI
99
Mean 8700
StDev 6157
95 N 20
AD 0.549
90
P-Value 0.137
80
70
Percent
60
50
40
30
20
10
1
-10000 0 10000 20000 30000
Ex3-16 / Table 3E.4
3.16. continued

Lognormal - 95% CI
99
Loc 8.776
Scale 0.8537
95 N 20
AD 0.521
90
P-Value 0.163
80
70
Percent
60
50
40
30
20
10
1
1000 10000 100000
Ex3-16 / Table 3E.4

Weibull - 95% CI
99
Shape 1.464
Scale 9624
90
N 20
80
AD 0.336
70
60 P-Value >0.250
50
40
30
Percent
20
10
3
2
1
100 1000 10000 100000
Ex3-16 / Table 3E.4
Plotted points do not tend to fall on a straight line on any of the probability plots, though the Weibull
distribution appears to best fit the data in the tails.
3.17.
An important quality characteristic of water is the concentration of suspended solid material (in ppm).
Table 3E.5 contains 40 measurements on suspended solids for a certain lake. Construct a normal
probability plot, a lognormal probability plot, and a Weibull probability plot for these data. Based on the
plots, which distribution seems to be the best model for the concentration of suspended solids?

Normal - 95% CI
99
Mean 9.470
StDev 22.56
95 N 40
AD 8.426
90
P-Value <0.005
80
70
Percent
60
50
40
30
20
10
1
-50 0 50 100 150
Ex3-17 / Table 3E.5
3.17. continued

Lognormal - 95% CI
99
Loc 0.9347
Scale 1.651
95 N 40
AD 0.201
90
P-Value 0.873
80
70
Percent
60
50
40
30
20
10
1
0.01 0.1 1 10 100 1000
Ex3-17 / Table 3E.5

Weibull - 95% CI
99
Shape 0.6132
Scale 5.782
90
N 40
80
AD 0.637
70
60 P-Value 0.091
50
40
30
Percent
20
10
3
2
1
0.0001 0.001 0.01 0.1 1 10 100
Ex3-17 / Table 3E.5
The lognormal distribution appears to be a reasonable model for the concentration data. Plotted points
on the normal and Weibull probability plots tend to fall off a straight line.
3.18. 
Consider the outpatient service times in Exercise 3.3. Construct a normal probability plot, an exponential
probability plot, and a Weibull probability plot for these data. Do any of these distributions seem to be a
reasonable probability model for the data? Based on the plots, which distribution is the best one to
describe the outpatient service time?

Normal - 95% CI
99
Mean 70.5
StDev 67.74
95 N 10
AD 0.644
90
P-Value 0.065
80
70
Percent
60
50
40
30
20
10
1
-200 -100 0 100 200 300
Ex3-3
3.18. continued
MTB > Graph > Probability Plot > Simple and select Exponential Distribution

Exponential - 95% CI
99
Mean 70.50
N 10
90
AD 0.368
80
P-Value 0.663
70
60
50
40
30
Percent
20
10
3
2
1
1 10 100 1000
Ex3-3

Weibull - 95% CI
99
Shape 0.9711
Scale 69.62
90
N 10
80
AD 0.337
70
60 P-Value >0.250
50
40
30
Percent
20
10
3
2
1
0.01 0.1 1 10 100 1000
Ex3-3
Plotted points do not tend to fall on a straight line on any of the probability plots, though the
exponential and Weibull distributions appear to best fit the data in the tails.
3.19. 
Consider the call handling times in Exercise 3.4. Construct a normal probability plot, an exponential
probability plot, and a gamma probability plot for these data. Do any of these distributions seem to be a
reasonable probability model for the data? Based on the plots, which distribution is the best one to
describe the call handling time?

Normal - 95% CI
99
Mean 9.55
StDev 7.736
95 N 20
AD 1.383
90
P-Value <0.005
80
70
Percent
60
50
40
30
20
10
1
-20 -10 0 10 20 30 40
Ex3-4
3.19. continued
MTB > Graph > Probability Plot > Simple and select Exponential Distribution

Exponential - 95% CI
99
Mean 9.550
N 20
90
AD 0.924
80
P-Value 0.136
70
60
50
40
30
Percent
20
10
3
2
1
0.1 1 10 100
Ex3-4
MTB > Graph > Probability Plot > Simple and select Gamma Distribution

Gamma - 95% CI
99
Shape 1.935
95 Scale 4.937
90 N 20
80 AD 0.520
70 P-Value 0.212
60
50
40
Percent
30
20
10
5
3
2
1
0.1 1 10 100
Ex3-4
Plotted points do not tend to fall on a straight line on any of the probability plots, though the gamma
distribution appears to best fit the data in the tails.
3.20.
Consider the viscosity data in Exercise 3.10. Assume that reading down, then across, gives the data in
time order. Construct and interpret a time-series plot.
MTB > Graph > Time Series Plot > Simple
Time Series Plot of Ex3-10 / Table 3E.3

17
16
Ex3-10 / Table 3E.3
15
14
13
12
1 8 16 24 32 40 48 56 64 72 80
Index
From visual examination, there are no trends, shifts or obvious patterns in the data, indicating that time
order is not an important source of variability.
3.21.
Reconsider the yield data in Exercise 3.9. Construct a time-series plot for these data. Interpret the plot.

100
95
Ex3-9 / Table 3E.2
90
85
1 9 18 27 36 45 54 63 72 81 90
Index
Time may be an important source of variability, as evidenced by potentially cyclic behavior.

3.22.
Consider the concentration of suspended solids from Exercise 3.17. Assume that reading across, then
down, gives the data in time order. Construct and interpret a time-series plot.

140
120
100
Ex3-17 / Table 3E.5
80
60
40
20
4 8 12 16 20 24 28 32 36 40
Index
Although most of the readings are between 0 and 20, there are two unusually large readings (9, 35), as
well as occasional “spikes” around 20. The order in which the data were collected may be an important
source of variability.
3.23.
Consider the chemical process yield data in Exercise 3.9. Calculate the sample average and standard
deviation.
n
94.1  93.2   84.5
x   xi n   89.476
i 1 90
2
n
 n  (94.1   84.5)2
 x 2
i    xi 
 i 1 
n (94.12   84.52 ) 
S i 1
 90  4.158
n 1 90  1
Descriptive Statistics: Ex3-9 / Table 3E.2

Variable N N* Mean SE Mean StDev Minimum Q1 Median
Ex3-9 / Table 3E.2 90 0 89.476 0.438 4.158 82.600 86.100 89.250
Variable Q3 Maximum
Ex3-9 / Table 3E.2 93.125 98.000
3.24.
Consider the chemical process yield data in Exercise 3.9. Construct a stem-and-leaf plot for the data and
compare it with the histogram from Exercise 3.9. Which display provides more information about the
process?

Leaf Unit = 0.10
2 82 69
6 83 0167
14 84 01112569
20 85 011144
30 86 1114444667
38 87 33335667
43 88 22368
(6) 89 114667
41 90 0011345666
31 91 1247
27 92 144
24 93 11227
19 94 11133467
11 95 1236
7 96 1348
3 97 38
1 98 0
Neither the stem-and-leaf plot nor the frequency histogram reveals much about an underlying
distribution or a central tendency in the data. The data appear to be fairly well scattered. The stem-
and-leaf plot suggests that certain values may occur more frequently than others; for example, those
ending in 1, 4, 6, and 7.
3.25.
Construct a box plot for the data in Exercise 3.1.
MTB > Graph > Boxplot > Simple
Boxplot of Ex3-1
Boxplot of Ex3-1
16.07
16.06
16.05
16.04
Ex3-1
16.03
16.02
16.01
16.00
In a box plot of normally distributed data, the median line is in the middle of the box, and the two
whiskers are the same length. This data is approximately normally distributed.
3.26.
Construct a box plot for the data in Exercise 3.2.
MTB > Graph > Boxplot > Simple
Boxplot of Ex3-2
Boxplot of Ex3-2
50.0075
50.0050
50.0025
Ex3-2
50.0000
49.9975
49.9950
In a box plot of normally distributed data, the median line is in the middle of the box, and the two
whiskers are the same length. This data is approximately normally distributed.
3.27.
Suppose that two fair dice are tossed and the random variable observed—say, x—is the sum of the two
up faces. Describe the sample space of this experiment, and determine the probability distribution of x.
x: {the sum of two up faces}

sample space: {2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12}
Calculate the probability of rolling a 2: obtained by rolling a 1 on each die:

Pr{ x  2}  Pr{1,1}  1  1  1
6 6 36
Calculate the probability of rolling a 3: 1 and 2 or 2 and 1:

Pr{ x  3}  Pr{1,2}  Pr{2,1}  1  1  1  1  2
6 6 6 6 36   
Calculate the probability of rolling a 4:
  
Pr{ x  4}  Pr{1,3}  Pr{2,2}  Pr{3,1}  1  1  1  1  1  1  3
6 6 6 6 6 6 36   
...
1 / 36; x  2 2 / 36; x  3 3 / 36; x  4 4 / 36; x  5 5 / 36; x  6 6 / 36; x  7

p(x)  
5 / 36; x  8 4 / 36; x  9 3 / 36; x  10 2 / 36; x  11 1 / 36; x  12 0; otherwise
3.28.
Find the mean and variance of the random variable in Exercise 3.27.
x   xi p(xi )  2 1 36   3 2 36    12 1 36   7
11
i 1
 xi p(xi )    xi p(xi )

n n
n
 i 1  5.92  72 11
S  i 1   0.38
n 1 10
3.29.
A mechatronic assembly is subjected to a final functional test. Suppose that defects occur at random in
these assemblies, and that defects occur according to a Poisson distribution with parameter  = 0.02.
(a) What is the probability that an assembly will have exactly one defect?
This is a Poisson distribution with parameter  = 0.02, x ~ POI(0.02).

e 0.02 (0.02)1
Pr{ x  1}  p(1)   0.0196
1!
(b) What is the probability that an assembly will have one or more defects?
e 0.02 (0.02)0
Pr{ x  1}  1  Pr{ x  0}  1  p(0)  1   1  0.9802  0.0198
0!
(c) Suppose that you improve the process so that the occurrence rate of defects is cut in half to  0.01.
What effect does this have on the probability that an assembly will have one or more defects?
This is a Poisson distribution with parameter  = 0.01, x ~ POI(0.01).
e 0.01 (0.01)0
Pr{ x  1}  1  Pr{ x  0}  1  p(0)  1   1  0.9900  0.0100
0!
Cutting the rate at which defects occur reduces the probability of one or more defects by approximately
one-half, from 0.0198 to 0.0100.
3.30.
The probability distribution of x is f (x)  ke x ,0  x   . Find the appropriate value of k. Find the mean
and variance of x.

For f(x) to be a probability distribution,  f (x)dx must equal unity.


x x 
 ke dx  [ke ]0  k[0  1]  k  1
0
This is an exponential distribution with parameter  = 1.
 = 1/ = 1 (Eqn. 3.32)
2 = 1/2 = 1 (Eqn. 3.33)
3.31.
The random variable x takes on the values 1, 2, or 3 with probabilities (1 + 3k)/3, (1 + 2k)/3, and
(0.5 + 5k)/3, respectively.
(1  3k) / 3; x  1 (1  2k) / 3; x  2

p(x)  
(0.5  5k) / 3; x  3 0; otherwise
(a) Find the appropriate value of k.

To solve for k, use F (x)   p(x )  1
i
i 1
(1  3k)  (1  2k)  (0.5  5k)
1
3
10k  0.5
k  0.05
(b) Find the mean and variance of x.
3  1  3(0.05)   1  2(0.05)   0.5  5(0.05) 

   x p(x )  1     2    3    1.867
i i  3   3   3 
i 1
3
 2   x2 p(x )  2  12 (0.383)  22 (0.367)  32 (0.250)  1.8672  0.615
i i
i 1
(c) Find the cumulative distribution function.
 1.15 1.15  1.1 1.15  1.1  0.75

F (x)    0.383; x  1  0.750; x  2  1.000; x  3
 3 3 3
3.32.
The probability distribution of the discrete random variable x is p(x) = krx, 0 < r < 1. Find the appropriate
value for k if x = 0, 1, ...
p(x)  kr x ; 0  r  1; x  0,1,2,

F (x)   kr x  1 by definition
i 0
k 1 1  r    1
k 1r
3.33.
A manufacturer of electronic calculators offers a one-year warranty. If the calculator fails for any reason
during this period, it is replaced. The time to failure is well modeled by the following probability
distribution:
f (x)  0.125e0.125 x x  0
(a) What percentage of the calculators will fail within the warranty period?
This is an exponential distribution with parameter  = 0.125: Pr{ x  1}  F (1)  1  e0.125(1)  0.118 .
Approximately 11.8% will fail during the first year.
(b) The manufacturing cost of a calculator is $50, and the profit per sale is $25. What is the effect of
warranty replacement on profit?
Mfg. cost = $50/calculator and Sale profit = $25/calculator

Net profit = $[-50(1 + 0.118) + 75]/calculator = $19.10/calculator.
The effect of warranty replacements is to decrease profit by $5.90/calculator.
3.34.
The net contents in ounces of canned soup is a random variable with probability distribution
4(x  11.75) 11.75  x  12.25
f ( x)  
4(12.75  x) 12.25  x  12.75
Find the probability that a can contains less than 12 ounces of product.
12 12 12
4 x2
 
12
Pr{ x  12}  F (12)  f (x)dx  4(x  11.75)dx   47 x 11.75  11.875  11.75  0.125
 11.75
2 11.75
3.35.
A production process operates with 1% nonconforming output. Every hour a sample of 25 units of
product is taken, and the number of nonconforming units counted. If one or more nonconforming units
are found, the process is stopped and the quality control technician must search for the cause of
nonconforming production. Evaluate the performance of this decision rule.
This is a binomial distribution with parameter p = 0.01 and n = 25. The process is stopped if x  1.
 25 
Pr{ x  1}  1  Pr{ x  1}  1  Pr{ x  0}  1    (0.01)0 (1  0.01)25  1  0.78  0.22
0
This decision rule means that 22% of the samples will have one or more nonconforming units, and the
process will be stopped to look for a cause. This is a somewhat difficult operating situation.
This exercise may also be solved using Excel or Minitab. In Excel, use the BINOM.DIST function:
=BINOM.DIST(0,25,0.01,TRUE) = 0.7778
MTB > Calc > Probability Distribution > Binomial
Cumulative Distribution Function
Binomial with n = 25 and p = 0.01
x P( X <= x )
0 0.777821
3.36.
Continuation of Exercise 3.35. Consider the decision rule described in Exercise 3.35. Suppose that the
process suddenly deteriorates to 3% nonconforming output. How many samples, on average, will be
required to detect this?
x ~ BIN(25, 0.03) Stop process if x  1.

 25 
Pr{ x  1}  1  Pr{ x  1}  1  Pr{ x  0}  1    (0.03)0 (1  0.03)25  1  0.47  0.53
0
This probability may also be found using Excel or Minitab. In Excel, use the BINOM.DIST function:
=BINOM.DIST(0,25,0.03,TRUE) = 0.4670
x P( X <= x )
0 0.466975
After deteriorating to 3% nonconforming, the probability of a sample containing one or more

nonconforming units, and subsequently stopping the process, is 0.53. The number of samples required
to obtain a sample with at least one nonconforming unit has a geometric distribution, with p = 0.53. On
average, 1 / p = 1 / 0.53 = 1.9 samples are required to stop the process after the deterioration.
3.37.
A random sample of 50 units is drawn from a production process every half hour. The fraction of
nonconforming product manufactured is 0.02. What is the probability that pˆ  0.04 if the fraction
nonconforming really is 0.02?
This is a binomial distribution with parameter p = 0.02 and n = 50.

4
 50 
Pr{ pˆ  0.04}  Pr{ x  2}    (0.02)x (1  0.02)(50  x )
x 0  x 
 50   50   50 
   (0.02)0 (1  0.02)50    (0.02)1 (1  0.02)49     (0.02)4 (1  0.02)46  0.921
0 1 4
Therefore, the probability is 0.921 that the sample fraction nonconforming can be pˆ  0.04 if the
population fraction nonconforming really is p = 0.02.
3.38.
A sample of 100 units is selected from a production process that is 2% nonconforming. What is the
probability that p̂ will exceed the true fraction nonconforming by k standard deviations, where k = 1, 2,
and 3?

  0.02(1  0.02) 100  0.014
Pr{pˆ  k  p}  1  Pr{ pˆ  k  p}  1  Pr{ x  n(k  p)}
k=1
1  Pr{ x  n(k  p)}  1  Pr{ x  100(1(0.014)  0.03)}  1  Pr{ x  3.4}  1  Pr{ x  3}
3  100 
100  x
1   (0.02) (1  0.02)
x
x 0  x 
 100   100   100   100  

 1    (0.02) (0.98)  
0 100
 (0.02) (0.98)  
1 99
 (0.02) (0.98)  
2 98 3 97
 (0.02) (0.98) 
 0   1   2   3  
 1   0.133  0.271  0.273  0.182   1  [0.859]  0.141
k=2
4  100 
100  x
  100  
1   x
(0.02) (0.98)  1  0.859   4 96
 (0.01) (0.99) 
x 0  x   4 
 
 1   0.859  0.090  1  [0.949]  0.051
k=3
6  100 
100  x
  100   100  
1   x
(0.02) (0.98)  1  0.949    (0.02) (0.98)  
5 95 6 94
 (0.02) (0.98) 
x 0  x   5   6 
 
 1   0.949  0.035  0.011  1  [0.996]  0.004
3.39. 
Suppose that 10% of the adult population has blood chemistry parameters consistent with a diagnosis of
a pre-diabetic condition. Of four volunteer participants in a health screening study, what is the
probability that one of them is pre-diabetic?
Probability Density Function
x P( X = x )
1 0.2916
3.40. 
Patients arriving at an outpatient clinic are routinely screened for high blood pressure. Assume that this
condition occurs in 12% of the population.
These are Bernoulli trials, with constant probability of success p = 0.12 on each trial, where success is
finding a patient with high blood pressure.
(a) What is the probability that the third patient of the day has high blood pressure?
This is a geometric distribution, looking for the probability that the first patient with high blood pressure
is the third patient.
Pr  X  3  1  0.12  0.12  0.093
2
MTB > Calc > Probability Distribution > Geometric
Geometric with p = 0.12
x P( X = x )
3 0.092928
* NOTE * X = total number of trials.

3.40. continued
(b) What is the average number of patients that must be seen to find the first patient with high blood
pressure?
This is a geometric distribution, looking for the average number of patients that must be seen before
finding the first patient with high blood pressure.
  E(X)  1 / p  1 / 0.12  8.3 patients
(c) If the clinic typically sees 40 patients each day, what is the probability of finding 6 patients with high
blood pressure?
This is a binomial distribution, with p = 0.12 and n = 40, looking for exactly 6 patients with high blood
pressure.
 40 
Pr{ x  6}    (0.12)6 (1  0.12)34  0.nnn
6
x P( X = x )
6 0.148473
3.41. 
A stock brokerage has four computers that are used for making trades on the New York Stock Exchange.
The probability that a computer fails on any single day is 0.005. Failures occur independently. Any failed
computers are repaired after the exchange closes, so each day can be considered an independent trial.
(a) What is the probability that all four computers fail on one day?
Pr{all four computers fail on one day} = 0.005^4 = 0.000000000625
(b) What is the probability that at least one computer fails on a day?
Pr{at least one fails} = 1 – Pr{none fails} = 1 – (1 – 0.005)^4 = 0.020
(c) What is the mean number of days until a specific computer fails?
This is a geometric distribution, with p = 0.005

  E(X)  1 / p  1 / 0.005  200
3.42. 
A computer system uses passwords consisting of the lowercase letters (a–z) and the integers (0–9).
There are 10,000 users with unique passwords. A hacker randomly selects (with replacement)
passwords in an attempt to break into the system.
(a) Suppose that 8000 of the users have six-character passwords. What is the mean and standard
deviation of the number of attempts required before the hacker selects a legitimate password?
These are Bernoulli trials, with x = 8000 unique passwords, or successes; n = (26+10)^6 = 36^6 =
2,176,782,336 possible passwords, or trials; p = 8000/2,176,782,336 = 0.00000368 constant probability
of success on each trial, where success is randomly selecting a legitimate six-character password.
Selecting the first legitimate password is a geometric distribution.
  E (X )  1 / p  1 / 0.00000368  271,739
  (1  p) p2  (1  0.00000368) 0.000003682  271.739
(b) Suppose that 2000 of the users have three-character passwords. What is the mean and standard
deviation of the number of attempts required before the hacker selects a legitimate password?
These are Bernoulli trials, with x = 2000 unique passwords, or successes; n = (26+10)^3 = 36^3 =46.656
possible passwords, or trials; p = 2000/46.656 = 0.0429 constant probability of success on each trial,
where success is randomly selecting a legitimate three-character password. Selecting the first legitimate
password is a geometric distribution.
  E ( X )  1 / p  1 / 0.0429  23
  (1  p) p2  (1  0.0429) 0.04292  23
3.43.
An electronic component for a medical X-ray unit is produced in lots of size N = 25. An acceptance
testing procedure is used by the purchaser to protect against lots that contain too many nonconforming
components. The procedure consists of selecting five components at random from the lot (without
replacement) and testing them. If none of the components is nonconforming, the lot is accepted.
(a) If the lot contains two nonconforming components, what is the probability of lot acceptance?
This is a hypergeometric distribution with N = 25 and n = 5, without replacement.

Given D = 2 and x = 0:
 2  25  2 
  
0 5  0  (1)(33,649)
Pr{Acceptance}  p(0)      0.633
 25  (53,130)
 
5
This exercise may also be solved using Excel (Excel Function HYPGEOMDIST(x, n, D, N)) or Minitab.
MTB > Calc > Probability Distribution > Hypergeometric
Hypergeometric with N = 25, M = 2, and n = 5
x P( X <= x )
0 0.633333
(b) Calculate the desired probability in (a) using the binomial approximation. Is this approximation
satisfactory? Why or why not?
For the binomial approximation to the hypergeometric, p = D/N = 2/25 = 0.08 and n = 5.
 5
Pr{acceptance}  p(0)    (0.08)0 (1  0.08)5  0.659
0
This approximation, though close to the exact solution for x = 0, violates the rule-of-thumb that n/N =
5/25 = 0.20 be less than the suggested 0.1. The binomial approximation is not satisfactory in this case.
(c) Suppose the lot size was N = 150. Would the binomial approximation be satisfactory in this case?
For N = 150, n/N = 5/150 = 0.033  0.1, so the binomial approximation would be a satisfactory
approximation the hypergeometric in this case.
3.43. continued
(d) Suppose that the purchaser will reject the lot with the decision rule of finding one or more
nonconforming components in a sample of size n, and wants the lot to be rejected with probability at
least 0.95 if the lot contains five or more nonconforming components. How large should the sample size
n be?
Find n to satisfy Pr{x  1 | D  5}  0.95, or equivalently Pr{x = 0 | D = 5} < 0.05.

 5  25  5   5  20 
     
 0  n  0   0  n 
p(0)  
 25   25 
   
n n
try n  10
 5  20 
  
0 10
p(0)     
(1)(184,756)
 0.057
 25  (3,268,760)
 
 10 
try n  11
 5  20 
  
0 11
p(0)     
(1)(167,960)
 0.038
 25  (4,457,400)
 
 11 
Let sample size n = 11.
3.44.
A lot of size N = 50 contains four nonconforming units. What is the probability that a sample of five units
selected at random contains exactly one nonconforming unit? What is the probability that it contains
one or more nonconformances?
This is a hypergeometric distribution with N = 50, n = 5, and D = 4.
 4  50  4 
  
 1  5  1  (4)(163,185)
Pr{ x  1}  p(1)    0.308
 50  (2,118,760)
 
5
x P( X = x )
1 0.308076
 4  46 
  
0 5
Pr{ x  1}  1  Pr{ x  0}  1  p(0)  1      1  0.647  0.353
 50 
 
5
x P( X = x )
0 0.646960
3.45.
A textbook has 500 pages on which typographical errors could occur. Suppose that there are exactly 10
such errors randomly located on those pages. Find the probability that a random selection of 50 pages
will contain no errors. Find the probability that 50 randomly selected pages will contain at least two
errors.
This is a hypergeometric distribution with N = 500 pages, n = 50 pages, and D = 10 errors. Checking n/N
= 50/500 = 0.1  0.1, the binomial distribution can be used to approximate the hypergeometric, with p =
D/N = 10/500 = 0.020.
 50 
Pr{ x  0}  p(0)    (0.020)0 (1  0.020)500  (1)(1)(0.364)  0.364
0
Pr{ x  2}  1  Pr{ x  1}  1  [Pr{ x  0}  Pr{ x  1}]  1  p(0)  p(1)
 50 
 1  0.364    (0.020)1 (1  0.020)50 1  1  0.364  0.372  0.264
1
3.46.
Surface-finish defects in a small electric appliance occur at random with a mean rate of 0.05 defects per
unit. Find the probability that a randomly selected unit will contain at least one surface-finish defect.
This is a Poisson distribution with  = 0.05 defects/unit.

e0.05 (0.05)0
Pr{ x  1}  1  Pr{ x  0}  1  p(0)  1   1  0.951  0.049
0!
This exercise may also be solved using Excel or Minitab. In Excel, use the POISSON function:
=POISSON(0,0.05,TRUE) = 0.9512
MTB > Calc > Probability Distribution > Poisson
Poisson with mean = 0.05
x P( X <= x )
0 0.951229
3.47.
Glass bottles are formed by pouring molten glass into a mold. The molten glass is prepared in a furnace
lined with firebrick. As the firebrick wears, small pieces of brick are mixed into the molten glass and
finally appear as defects (called “stones”) in the bottle. If we can assume that stones occur randomly at
the rate of 0.00001 per bottle, what is the probability that a bottle selected at random will contain at
least one such defect?
This is a Poisson distribution with  = 0.00001 stones/bottle.

e 0.00001 (0.00001)0
Pr{ x  1}  1  Pr{ x  0}  1   1  0.99999  0.00001
0!
3.48.
The billing department of a major credit card company attempts to control errors (clerical, data
transmission, etc.) on customers’ bills. Suppose that errors occur according to a Poisson distribution
with parameter  = 0.01. What is the probability that a customer’s bill selected at random will contain
one error?
This is a Poisson distribution with  = 0.01 errors/bill.

e 0.01 (0.01)1
Pr{ x  1}  p(1)   0.0099
1
3.49.
A production process operates in one of two states: the in-control state, in which most of the units
produced conform to specifications, and an out-of-control state, in which most of the units produced
are defective. The process will shift from the in-control to the out-of-control state at random. Every
hour, a quality control technician checks the process, and if it is in the out-of-control state, the
technician detects this with probability p. Assume that when the process shifts out of control it does so
immediately following a check by the inspector, and once a shift has occurred, the process cannot
automatically correct itself. If t denotes the number of periods the process remains out of control
following a shift before detection, find the probability distribution of t. Find the mean number of periods
the process will remain in the out-of-control state.
The distribution for t is based on the number of periods for which the process shifted from in control to
out of control, 1, with probability (p)1, and the periods from which the process remains in the out of
control state without a shift, t  1, with probability (1  p)t-1. There is only one permutation for this
situation. Hence, the distribution is the combination of the two states as follows:
Pr(t)  p(1  p)t 1 ; t  1,2,3, The mean is calculated using (3.5b):
 
d   t 1
   xi p(xi )   t  p(1  p)t 1   p q 
i 1 t 1 dq  t 1  p
3.50.
An inspector is looking for nonconforming welds in the gasoline pipeline between Phoenix and Tucson.
The probability that any particular weld will be defective is 0.01. The inspector is determined to keep
working until finding three defective welds. If the welds are located 100 feet apart, what is the
probability that the inspector will have to walk 5000 feet? What is the probability that the inspector will
have to walk more than 5000 feet?
This is a Pascal distribution with Pr{defective weld} = p = 0.01, r = 3 welds, and x = 1 + (5000/100) = 51.
 51  1  513
Pr{ x  51}  p(51)    (0.01) (1  0.01)  (1225)(0.000001)(0.617290)  0.0008
3
 3 1 
Pr{ x  51}  Pr{r  0}  Pr{r  1}  Pr{r  2}
 50   50   50 
   0.010 0.9950    0.0110.9949   0.012 0.9948  0.9862
0 1 2
3.51.
The tensile strength of a metal part is normally distributed with mean 40 pounds and standard deviation
5 pounds. If 50,000 parts are produced, how many would you expect to fail to meet a minimum
specification limit of 35-pounds tensile strength? How many would have a tensile strength in excess of
48 pounds?
x ~ N (40, 52); n = 50,000
How many fail the minimum specification, LSL = 35 lb.? Using the standard normal distribution,
Appendix II:
 35  40 
Pr{ x  35}  Pr z    Pr{ z  1}  (1)  0.159
 5 
The number that fail the minimum specification are (50,000)  (0.159) = 7950.
This exercise may also be solved using Excel (Excel Function NORMDIST(X, , , TRUE)) or Minitab.
MTB > Calc > Probability Distribution > Normal
Normal with mean = 40 and standard deviation = 5
x P( X <= x )
35 0.158655
3.51. continued
How many exceed 48 lb.?

 48  40 
Pr{ x  48}  1  Pr{ x  48}  1  Pr z    1  Pr{ z  1.6}  1  (1.6)  1  0.945  0.055
 5 
The number that exceed 48 lb are (50,000)  (0.055) = 2750.
3.52.
The output voltage of a power supply is normally distributed with mean 5 V and standard deviation 0.02
V. If the lower and upper specifications for voltage are 4.95 V and 5.05 V, respectively, what is the
probability that a power supply selected at random will conform to the specifications on voltage?
x ~ N(5, 0.022); LSL = 4.95 V; USL = 5.05 V
Pr{Conformance}  Pr{LSL  x  USL}  Pr{ x  USL}  Pr{ x  LSL}

 5.05  5   4.95  5 
      (2.5)  (2.5)  0.99379  0.00621  0.98758
 0.02   0.02 
3.53.
Continuation of Exercise 3.52. Reconsider the power supply manufacturing process in Exercise 3.52.
Suppose we wanted to improve the process. Can shifting the mean reduce the number of
nonconforming units produced? How much would the process variability need to be reduced in order to
have all but one out of 1000 units conform to the specifications?
The process, with mean 5 V, is currently centered between the specification limits (target = 5 V).
Shifting the process mean in either direction would increase the number of nonconformities produced.
Desire Pr{Conformance} = 1 / 1000 = 0.001. Assume that the process remains centered between the
specification limits at 5 V. Need Pr{x  LSL} = 0.001 / 2 = 0.0005.
(z)  0.0005
z   1 (0.0005)  3.29
LSL   LSL   4.95  5
z , so     0.015
 z 3.29
Process variance must be reduced to 0.0152 to have at least 999 of 1000 conform to specification.
3.54.
If x is normally distributed with mean  and standard deviation , and given that the probability that x is
less than 32 is 0.0228, find the value of .
x ~ N( ,42 ). Find  such that Pr{ x  32}  0.0228.

 1 (0.0228)  1.9991
32  
 1.9991
4
  4(1.9991)  32  40.0
3.55.
The life of an automotive battery is normally distributed with mean 900 days and standard deviation 35
days. What fraction of these batteries would be expected to survive beyond 1000 days?
x ~ N(900, 352)
Pr{ x  1000}  1  Pr{ x  1000}
 1000  900 
 1  Pr  x  
 35 
 1  (2.8571)
 1  0.9979
 0.0021
3.56.
A light bulb has a normally distributed light output with mean 7500 end foot-candles and standard
deviation of 100 end foot-candles. Find a lower specification limit such that only 0.5% of the bulbs will
not exceed this limit.
x ~ N(7500, 1002). Find LSL such that Pr{x < LSL} = 0.005
 1 (0.005)  2.5758
LSL  7500
 2.5758
100
LSL  100(2.5758)  7500  7242
3.57.
The specifications on an electronic component in a target-acquisition system are that its life must be
between 5000 and 10,000 h. The life is normally distributed with mean 7500 h. The manufacturer
realizes a price of $10 per unit produced; however, defective units must be replaced at a cost of $5 to
the manufacturer. Two different manufacturing processes can be used, both of which have the same
mean life. However, the standard deviation of life for process 1 is 1000 h, whereas for process 2 it is only
500 h. Production costs for process 2 are twice those for process 1. What value of production costs will
determine the selection between processes 1 and 2?
x1 ~ N(7500, 12 = 10002); x2 ~ N(7500, 22 = 5002); LSL = 5,000 h; USL = 10,000 h
sales = $10/unit, defect = $5/unit, profit = $10  Pr{good} + $5  Pr{bad} – c
For Process 1
proportion defective  p1  1  Pr{LSL  x1  USL}  1  Pr{ x1  USL}  Pr{ x1  LSL}
 10,000  7,500   5,000  7,500 
 1  Pr  z1    Pr  z1  
 1,000   1,000 
 1  (2.5)  (2.5)  1  0.9938  0.0062  0.0124
profit for process 1 = 10 (1 – 0.0124) + 5 (0.0124) – c1 = 9.9380 – c1
For Process 2
proportion defective  p2  1  Pr{LSL  x2  USL}  1  Pr{ x2  USL}  Pr{ x2  LSL}
 10,000  7,500   5,000  7,500 
 1  Pr  z2    Pr  z2  
 500   500 
 1  (5)  (5)  1  1.0000  0.0000  0.0000
profit for process 2 = 10 (1 – 0.0000) + 5 (0.0000) – c2 = 10 – c2
If c2 > c1 + 0.0620, then choose process 1

3.58.
A quality characteristic of a product is normally distributed with mean  and standard deviation .
Specifications on the characteristic are 6  x  8. A unit that falls within specifications on this quality
characteristic results in a profit of C0. However, if x < 6 the profit is –C1, whereas if x > 8, the profit is –C2.
Find the value of that maximizes the expected profit.
Proportion less than lower specification:

 6 
pl  Pr{ x  6}  Pr z    (6   )
 1 
Proportion greater than upper specification:
 8 
pu  Pr{ x  8}  1  Pr{ x  8}  1  Pr z    1  (8  )
 1 
Profit  C 0 pwithin  C1 pl  C2 pu
 C 0 [(8   )  (6   )]  C1 [(6   )]  C2 [1  (8   )]
 (C 0  C2 )[(8   )]  (C 0  C1 )[(6   )]  C2
d d 8  1 
[(8  )]    exp(t 2 / 2)dt 
d d    2 
Set s = 8 –  and use chain rule
d s 1  ds
d
d
[(8   )]    exp(t 2 / 2)dt  
1

exp 1 / 2  (8  )2 
ds   2  d 2
 1   1 
d(Profit)
d
 (C0  C2 )   
exp 1 / 2  (8  )2   (C0  C1 )   
exp 1 / 2  (6  )2 
 2   2 
Setting equal to zero


C0  C1 exp 1 / 2  (8   )
2

 exp(2  14)

C 0  C2 exp 1 / 2  (8   )2 
1   C C  
So   ln  0 1   14  maximizes the expected profit.
2   C 0  C2  
3.59.
Derive the mean and variance of the binomial distribution.
n
For the binomial distribution, p(x)    p x (1  p)n x ; x  0,1,..., n
x
  n   n 1
  E(x)   xi p(xi )   x   p x (1  p)n x   n  p  1  p 
n
p  np
i 1 x 0
 x  
 2  E[(x   )2 ]  E (x 2 )  [E (x)]2
  n  n x 
E (x 2 )   xi2 p(xi )   x 2   p x 1  p   np  (np)2  np2
n
i 1 x 0
 x  
 2  np  (np)2  np2   np  np(1  p)
2
3.60.
Derive the mean and variance of the Poisson distribution.
e  x
For the Poisson distribution, p(x)  ; x  0,1,
x!
 e    x      ( x 1)
 
 
  E[ x]   xi p(xi )   x  e   e    e  
x  0 (x  1)!
i 1 x 0
 x! 
 2  E[(x   )2 ]  E (x 2 )  [E (x)]2
   e  x 
E (x 2 )   xi2 p(xi )   x 2   
2
i 1 x 0
 x! 
 2  ( 2   )      
2
3.61.
Derive the mean and variance of the exponential distribution.
For the exponential distribution, f (x)  e x ; x  0
For the mean:
   xf (x)dx   x  e   x dx
 
0 0
Integrate by parts, setting u  x and dv   exp( x)

  1 1
uv   vdu   x exp   x  0   exp   x  dx  0  
0  
For the variance:

2
1
  E[(x   ) ]  E (x )  [E (x) ]  E (x )   
2 2 2 2 2

 
E (x 2 )   x 2 f (x)dx   x 2  exp( x)dx
 0
Integrate by parts, setting u  x 2 and dv   exp( x)


 2
uv   vdu   x exp( x)
2
 2  x exp( x)dx  (0  0) 
0
0
2
2 1 1
2   
2 2 2
3.62. 
Derive the mean and variance of the geometric distribution.
For the geometric distribution, p(x)  (1  p)x 1 p, x  1,2, and 0  p  1
Use the sum and the first and second derivatives of a geometric series:
  
a a 2a
g(r )   ar k  ; g '(r )   akr k 1  ; g "(r )   ak(k  1)r k 2  for  1  r  1
k 0 1r k 1 (1  r )2
k 2 (1  r )3
  E[ x]   xi p(xi )   x (1  p)x 1 p  

  p 1

i 1 x 1 (1  (1  p))2 p
To find the variance, also need the second factorial moment:

 
2p(1  p) 2(1  p)
E  X (X  1)   x(x  1)(1  p)x 1 p   p(1  p)x(x  1)(1  p)x 2  
i 1 i 2 (1  (1  p))3
p2
 2  E[(x   )2 ]  E (x 2 )  [E(x)]2  E  X ( X  1)  E( X )  E( X )

2
2(1  p) 1 1 2(1  p)  p  1 (1  p)
   2  2
p2 p p p2 p

ch03 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ch03 PDF

Uploaded by

Copyright:

Available Formats

Chapter 3

Modeling Process Quality

IMPORTANT TERMS AND CONCEPTS

Lognormal distribution Standard normal distribution

(a) Calculate the sample average.

(b) Calculate the sample standard deviation.

Descriptive Statistics: Ex3-1

Variable N N* Mean SE Mean StDev Minimum Q1 Median Q3

(a) Calculate the sample average.

(b) Calculate the sample standard deviation.

Descriptive Statistics: Ex3-2

Variable N N* Mean SE Mean StDev Minimum Q1 Median Q3

Descriptive Statistics: Ex3-3

Variable N N* Mean SE Mean StDev Minimum Q1 Median Q3 Maximum

Descriptive Statistics: Ex3-4

Variable N N* Mean SE Mean StDev Minimum Q1 Median Q3 Maximum

(a) Calculate the sample average.

(b) Calculate the sample standard deviation.

Descriptive Statistics: Ex3-5

Variable N N* Mean SE Mean StDev Minimum Q1 Median Q3

(a) Find the sample median of these data.

(a) Calculate the sample average.

x   xi n   96  102   156  8  121.25 kN

(b) Calculate the sample standard deviation.

Descriptive Statistics: Ex3-7

Variable N N* Mean SE Mean StDev Minimum Q1 Median Q3

(a) Calculate the sample average and standard deviation.

Descriptive Statistics: Ex3-8_Table 3E.1

Variable N N* Mean SE Mean StDev Minimum Q1 Median

(b) Construct a histogram.

MTB > Graph > Histogram > Simple

(Edit X-axis scale to change binning to 7 intervals)

Histogram of Ex3-8 / Table 3E.1

Histogram of Ex3-8 / Table 3E.1

(c) Construct a stem-and-leaf plot.

MTB > Graph > Stem-and-Leaf

Stem-and-Leaf Display: Ex3-8 / Table 3E.1

Stem-and-leaf of Ex3-8 / Table 3E.1 N = 40

(Edit X-axis scale to change binning to 9 intervals)

Histogram of Ex3-9 / Table 3E.2

Histogram of Ex3-9 / Table 3E.2

(a) Construct a stem-and-leaf display for the viscosity data.

Stem-and-Leaf Plot (Unordered)

(b) Construct a frequency distribution and histogram.

Histogram of Ex3-10 / Table 3E.3

Histogram of Ex3-10 / Table 3E.3

MTB > Graph > Stem-and-Leaf

Stem-and-Leaf Display: Ex3-10 / Table 3E.3

Stem-and-leaf of Ex3-10 / Table 3E.3 N = 80

median observation rank is (0.5)(80) + 0.5 = 40.5

Q1 observation rank is (0.25)(80) + 0.5 = 20.5

Q3 observation rank is (0.75)(80) + 0.5 = 60.5

(d) What are the tenth and ninetieth percentiles of viscosity?

10th percentile observation rank = (0.10)(80) + 0.5 = 8.5

90th percentile observation rank is (0.90)(80) + 0.5 = 72.5

Probability Plot of Ex3-1

Probability Plot of Ex3-3

Probability Plot of Ex3-8 / Table 3E.1

Probability Plot of Ex3-9 / Table 3E.2

Probability Plot of Ex3-10 / Table 3E.3

Probability Plot of Ex3-10 / Table 3E.3

Probability Plot of Ex3-10 / Table 3E.3

Probability Plot of Ex3-16 / Table 3E.4