SOLUTIONS TO PROBLEMS IN CHAPTER 2

Problem 2.1. The results of tests to determine the modulus of rupture (MOR) for a set of
timber beams are shown in Table P2.1.
A. Plot the relative frequency and cumulative frequency histograms.
B. Calculate the sample mean, standard deviation, and coefficient of variation.
C. Plot the data on normal probability paper.

Solution:
A. For the histogram plots, the interval size is chosen to be 250. There are 45 data points.

Interval Relative Cumulative

Frequency Frequency
3250-3500 0 0
3500-3750 0.06667 0.066667
3750-4000 0.11111 0.177778
4000-4250 0.02222 0.200000
4250-4500 0.06667 0.266667
4500-4750 0.11111 0.377778
4750-5000 0.08889 0.466667
5000-5250 0.15556 0.622222
5250-5500 0.11111 0.733333
5500-5750 0.04444 0.777778
5750-6000 0.11111 0.888889
6000-6250 0.02222 0.911111
6250-6500 0.04444 0.955556
6500-6750 0 0.955556
6750-7000 0 0.955556
7000-7250 0.04444 1
7250-7500 0 1

Relative Frequency

0.18
0.16
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
3250-3500

3750-4000

4250-4500

4750-5000

5250-5500

5750-6000

6250-6500

6750-7000

7250-7500

iv
 Cumulative Frequency

1.2
1
0.8
0.6
0.4
0.2
0

3250-3500

3750-4000

4250-4500

4750-5000

5250-5500

5750-6000

6250-6500

6750-7000

7250-7500
B. Using Eqns. 2.25 and 2.26, sample mean = x = 5031 and sample standard deviation = sX =
880.4. The coefficient of variation based on sample parameters is s X / x = 0.175.
C. The step-by-step procedure described in Section 2.5 is followed to construct the plot on
normal probability paper.

MOR data on normal probability paper

3
Standard Normal Variate

-1

-2

-3
0 2000 4000 6000 8000
MOR

-------------------------------------------

Problem 2.2. A set of test data for the load-carrying capacity of a member is shown in Table
P2.2.
A. Plot the test data on normal probability paper.
B. Plot a normal distribution on the same probability paper. Use the sample mean and
standard deviation as estimates of the true mean and standard deviation.
C. Plot a lognormal distribution on the same normal probability paper. Use the sample mean
and standard deviation as estimates of the true mean and standard deviation.

v
D. Plot the relative frequency and cumulative frequency histograms.

Solution:
A. Using Eqns. 2.25 and 2.26, the sample mean, sample standard deviation, and sample
coefficient of variation are
x = 4.127 s X = 0.1770 CoV = s X / x = 0.04289
To plot on normal probability paper, we follow the step-by-step procedure outlined in
Section 2.5.

Raw Data Sorted i i/(N+1)=pi Φ-1(pI)

3.95 3.74 1 0.05 -1.64485
4.07 3.90 2 0.10 -1.28155
4.14 3.95 3 0.15 -1.03643
3.99 3.97 4 0.20 -0.84162
4.21 3.99 5 0.25 -0.67449
4.39 4.04 6 0.30 -0.5244
4.21 4.05 7 0.35 -0.38532
3.90 4.07 8 0.40 -0.25335
3.74 4.07 9 0.45 -0.12566
4.28 4.14 10 0.50 0
4.15 4.15 11 0.55 0.125661
4.04 4.21 12 0.60 0.253347
4.26 4.21 13 0.65 0.385321
4.41 4.22 14 0.70 0.524401
4.22 4.26 15 0.75 0.67449
4.07 4.28 16 0.80 0.841621
3.97 4.36 17 0.85 1.036433
4.05 4.39 18 0.90 1.281551
4.36 4.41 19 0.95 1.644853

Normal Probability Paper

3
Standard Normal

2
1
Variate

0
-1 3.5 3.7 3.9 4.1 4.3 4.5
-2
-3
Capacity

Raw data Normal Distribution

Lognormal distribution

vi
B. See part A. To plot the normal distribution on normal probability paper: (1) Calculate
values of the standard normal variable z for some arbitrary values of x (in ascending
order) in the range of interest. The formula for z is
x − µ X x − 4.127
z= =
σX 0.1770
(2) Plot z versus x on standard linear graph paper. The plot is shown in the graph in Part
A. Note that the relationship between z and x is linear, so only two points are needed to
plot the graph.
C. To plot a lognormal distribution, we need the lognormal distribution parameters. We will
assume that µX= x and σX=sX.
σln2 X = ln(1 + VX2 ) = ln(1 + (0.04289) 2 ) ⇒ σln X = 0.04287
µ ln X = ln(µ X ) − 0.5σln2 X = ln(4.127) − 0.5(0.04287) 2 = 1.417
To plot the lognormal distribution on normal probability paper: (1) Calculate FX(x) for
some arbitrary values of x (in ascending order) in the range of interest. FX(x) is the
lognormal distribution, and it can be calculated as shown in Section 2.4.3. (2) Use the
values of FX(xi) as the values of pi for plotting points on normal probability paper. (3)
Calculate zi = Φ-1(pi). (4) Plot zi versus xi. The plot is shown in the graph in Part A.
D. There are 19 data points. The interval size was chosen to be 0.05.

Interval Number Cumulative %

3.70-3.75 1 5.26%
3.75-3.8 0 5.26%
3.8-3.85 0 5.26%
3.85-3.9 1 10.53%
3.9-3.95 1 15.79%
3.95-4.0 2 26.32%
4.0-4.05 2 36.84%
4.05-4.1 2 47.37%
4.1-4.15 2 57.89%
4.15-4.2 0 57.89%
4.2-4.25 3 73.68%
4.25-4.3 2 84.21%
4.3-4.35 0 84.21%
4.35-4.4 2 94.74%
4.4-4.45 1 100.00%

vii
 Relative Frequency Plot

0.18
0.16

Relative frequency
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
3.70-3.75

3.8-3.85

3.9-3.95

4.0-4.05

4.1-4.15

4.2-4.25

4.3-4.35

4.4-4.45
Intervals

Cumulative Frequency Plot

100.00%
80.00%
60.00%
40.00%
20.00%
.00%
3.70-3.75

3.8-3.85

3.9-3.95

4.0-4.05

4.1-4.15

4.2-4.25

4.3-4.35

4.4-4.45

Intervals

--------------------------------------

Problem 2.3. For the data in Table 2.5, calculate the statistical estimate of the correlation
coefficient using Equation (2.99).

Solution:
The formula is
n
 n 
∑ (x i − x )(yi − y )  ∑ x i yi  − nxy
1 1  i =1 
ρˆ XY = i =1
=
n −1 s Xs Y n −1 s Xs Y

Note that is doesn’t matter which variable is x and which is y. There are 100 data points.
After manipulating the data, you should find:

viii
 ′
for f c x = 2743.82 s X = 520.082
for E y = 2991380 s Y = 361245
ρˆ XY = 0.806

--------------------------------------------

Problem 2.4. A variable X is to be modelled using a uniform distribution. The lower bound
value is 5, and the upper bound value is 36.
A. Calculate the mean and standard deviation of X.
B. What is the probability that the value of X is between 10 and 20?
C. What is the probability that the value of X is greater than 31?
D. Plot the CDF on normal probability paper.

Solution:
A. The value of a is 5, and the value of b is 36. (Refer to Eqn. 2.31.) Using Eqns. 2.32
and 2.33,
a+b
µX = = 20.5
2
(b − a ) 2
σ2X = = 80.1 ⇒ σX = 8.95
12

B. Using Eqn. 2.31 and the definition of CDF (Eqn. 2.13),

x
1 1
FX ( x ) = ∫ dξ = ( x − 5)
5
b−a 31
Therefore, using Eqn. 2.15
15 5
P(10 ≤ x ≤ 20) = FX (20) − FX (10) = − = 0.3226
31 31
C.
26
P(X > 31) = 1 − P(X ≤ 31) = 1 − FX (31) = 1 − = 0.1613
31
D. To plot the uniform distribution on normal probability paper: (1) Calculate FX(x) for
some arbitrary values of x (in ascending order) in the range of interest. FX(x) is the
uniform distribution, and its range is limited. (2) Use the values of FX(xi) as the values
of pi for plotting points on normal probability paper. (3) Calculate zi = Φ-1(pi). (4)
Plot zi versus xi on standard linear graph paper as shown below.

ix
 Uniform distribution

Standard Normal Variable

2

0
0 10 20 30 40
-1

-2

-3
X

----------------------------------------

Problem 2.5. The dead load D on a structure is to be modelled as a normal random variable
with a mean value of 100 and a coefficient of variation of 8%.
A. Plot the PDF and CDF on standard graph paper.
B. Plot the CDF on normal probability paper.
C. Determine the probability that D is less than or equal to 95.
D. Determine the probability that D is between 95 and 105.

Solution:
A. The formulas for PDF and CDF are given by Eqns. 2.34 and 2.39. The value of σD is
VDµD = 0.08(100)=8. Appendix B can be used to determine values of CDF, or a
computer spreadsheet program can be used.

PDF Function

0.06
0.05
0.04
PDF

0.03
0.02
0.01
0
70 80 90 100 110 120 130
x

x
 CDF Function

1
0.8
0.6

CDF
0.4
0.2
0
70 80 90 100 110 120 130
x

B. To plot the normal distribution on normal probability paper: (1) Calculate values of
the standard normal variable z for some arbitrary values of d (in ascending order) in
the range of interest. The formula for z is
d − µ D d − 100
z= =
σD 8
(2) Plot z versus d on standard linear graph paper. Note that the relationship between
z and d is linear, so only two points are needed to plot the graph.

Normal Probability Paper

4
3
2
1
0
-1 70 90 110 130
-2
-3
-4
d

C.
 95 − 100 
P(D ≤ 95) = Φ  = Φ (−0.625) = 0.266
 8 

xi
D.
 105 − 100   95 − 100 
P(95 ≤ D ≤ 105) = Φ  − Φ  = Φ (0.625) − Φ (−0.625) = 0.468
 8   8 
----------------------------------------

Problem 2.6. The ground snow load q (in pounds per square foot) is to be modelled as a
lognormal random variable. The mean value of the ground snow load is 8.85 psf, and the
standard deviation is 5.83 psf.
A. Plot the PDF and CDF on standard graph paper.
B. Plot the CDF on normal probability paper.
C. Determine the probability that q is less than or equal to 7.39 psf.
D. Determine the probability that q is between 6 and 8 psf.

Solution:
A. Since q is lognormal, ln(q) is normally distributed with mean µln(q) and standard
deviation σln(q). These parameters can be found using Eqns. 2.48 and 2.49. You can’t
use the small V approximation here because Vq is 0.659 which is larger than 0.2.
2
 5.83 
σln(
2
q ) = ln(1 + Vq ) = ln(1 + 
2
 ) ⇒ σln( q ) = 0.6004
 8.85 
µ ln( q ) = ln(µ q ) − 0.5σln(
2
q ) = ln(8.85) − 0.5(0.6004) = 2.000
2

The formulas for PDF and CDF are given by Eqns. 2.52 and 2.47. Both of these can
be calculated using information from the standard normal distribution.

Lognormal PDF

0.2

0.15

0.1

0.05

0
0 10 20 30 40 50
q

xii
 Lognormal CDF

1
0.8
0.6
0.4
0.2
0
0 10 20 30 40 50
q

B. To plot the lognormal distribution on normal probability paper: (1) Calculate Fq(q) for
some arbitrary values of q (in ascending order) in the range of interest. Fq(q) is the
lognormal distribution, and it can be calculated as shown in Section 2.4.3. (2) Use the
values of Fq(qi) as the values of pi for plotting points on normal probability paper. (3)
Calculate zi = Φ-1(pi). (4) Plot zi versus qi. on standard linear graph paper. Since the
lognormal and normal distributions are related to each other, you can show that the
relationship between z and q is
ln(q ) − µ ln( q ) ln(q ) − 2.000
z= =
σln( q ) 0.6004

Normal Probability Paper

3
2
1
0
-1 0 10 20 30 40 50

-2
-3
q

C.
 ln(7.39) − 2.000 
P(q ≤ 7.39) = Φ  ≈ Φ (0) = 0.5
 0.6004 

xiii
D.
 ln(8) − 2.000   ln(6) − 2.000 
P(6 ≤ q ≤ 8) = Φ  − Φ 
 0.6004   0.6004 
= Φ (0.132) − Φ (−0.346)
= 0.188
-------------------------------------

Problem 2.7. The yield stress of A36 steel is to be modelled as a lognormal random variable
with a mean value of 36 ksi and a coefficient of variation of 10%.
A. Plot the PDF and CDF on standard graph paper.
B. Plot the CDF on normal probability paper.
C. Determine the probability that the yield stress is greater than 40 ksi.
D. Determine the probability that the yield stress is between 34 and 38 ksi.

Solution:
A. Since X is lognormal, ln(X) is normally distributed with mean µlnX and standard
deviation σlnX. These parameters can be found using Eqns. 2.48 and 2.49. We can use
the small V approximation here because VX is 0.10 which is smaller than 0.2.
σln2 X = ln(1 + VX2 ) ≈ VX2 ⇒ σln X = VX = 0.10
µln X = ln(µ X ) − 0.5σln2 X ≈ ln(µ X ) = 3.584
The formulas for PDF and CDF are given by Eqns. 2.52 and 2.47. Both of these can
be calculated using information from the standard normal distribution.

Lognormal PDF

1.2
1
0.8
PDF

0.6
0.4
0.2
0
0 20 40 60 80
x

xiv
 Lognormal CDF

1
0.8
0.6

CDF
0.4
0.2
0
0 20 40 60 80
x

B. To plot the lognormal distribution on normal probability paper: (1) Calculate FX(x) for
some arbitrary values of x (in ascending order) in the range of interest. FX(x) is the
lognormal distribution, and it can be calculated as shown in Section 2.4.3. (2) Use the
values of FX(xi) as the values of pi for plotting points on normal probability paper. (3)
Calculate zi = Φ-1(pi). (4) Plot zi versus xi. Since the lognormal and normal
distributions are related to each other, you can show that the relationship between z
and x is
ln(x ) − µ ln X ln(x ) − 3.584
z= =
σln X 0.1

Normal Probability Paper

4
3
2
1
0
-1 0 20 40 60 80
-2
-3
-4
x

C.
 ln(40) − 3.584 
P(X > 40) = 1 − P(X ≤ 40) = 1 − Φ  = 1 − Φ(1.049) = 0.147
 0.1 

xv
D.
 ln(38) − 3.584   ln(34) − 3.584 
P(34 ≤ X ≤ 38) = Φ  − Φ 
 0.1   0.1 
= Φ (0.536) − Φ (−0.576)
= 0.422
-------------------------------------

Problem 2.8. The annual extreme wind speed at a particular location is to be modelled as an
extreme Type I random variable. The mean value of the extreme wind is 50 miles per hour
(mph) and the coefficient of variation is 15%.
A. Plot the PDF and CDF on standard graph paper.
B. Plot the CDF on normal probability paper.
C. Determine the probability that the annual maximum wind speed is greater than 50
mph.
D. Determine the probability that the annual maximum wind speed is less than 50 mph.
E. Determine the probability that the wind speed will be between 40 and 60 mph.

Solution:
A. Let X denote the random variable representing wind speed.
µ X = 50 VX = 0.15 ⇒ σX = VXµ X = 7.5
To plot the PDF and CDF, we need the parameters for the Type I distribution. They can
be found using Eqns. 2.62 and 2.63.
1.282
α= = 0.1709 u = µ X − 0.45σX = 46.63
σX
The PDF and CDF functions are given by Eqns. 2.58 and 2.59.

Type I PDF

0.07
0.06
0.05
PDF

0.04
0.03
0.02
0.01
0
0 20 40 60 80 100
x

xvi
 Type I CDF

1
0.8
0.6

CDF
0.4
0.2
0
0 20 40 60 80 100
x

B. To plot the Type I CDF on normal probability paper: (1) Calculate FX(x) for some
arbitrary values of x (in ascending order) in the range of interest. FX(x) is the Type I
distribution. (2) Use the values of FX(xi) as the values of pi for plotting points on normal
probability paper. (3) Calculate zi = Φ-1(pi). (4) Plot zi versus xi on standard linear graph
paper.

Normal Probability Paper

0
0 20 40 60 80 100
-1

-2

-3
x

C.
P(X > 50) = 1 − P(X ≤ 50) = 1 − FX (50)
= 1 − exp(− exp(−α(50 − u )))
= 1 − 0.5700
= 0.43

xvii
D.
P(X ≤ 50) = FX (50)
= exp(− exp(−α(50 − u )))
= 0.5700
E.
P(40 ≤ X ≤ 60) = FX (60) − FX (40)
= exp(− exp(−α(60 − u ))) − exp(− exp(−α(40 − u )))
= 0.9032 − 0.04482 = 0.8584
---------------------------------

Problem 2.9. The peak ground acceleration A that is expected at a site in a 50 year time
period is modelled as an extreme Type II random variable with u = 0.2g (g is the acceleration
of gravity) and k = 3.
A. Plot the PDF and CDF on standard graph paper.
B. Plot the CDF on normal probability paper.
C. Determine the probability that the ground acceleration will be between 0.15g and 0.2g.
D. Determine the probability that the ground acceleration will be greater than 0.3g.

Solution:
A. The parameters u and k are already known. The formulas for PDF and CDF are given in
Eqns. 2.64 and 2.65.

Type II PDF

7
6
5
4
PDF

3
2
1
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
x

xviii
 Type II CDF

0.8

0.6

CDF 0.4

0.2

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
x

B. To plot the Type II CDF on normal probability paper: (1) Calculate FX(x) for some
arbitrary values of x (in ascending order) in the range of interest. FX(x) is the Type II
distribution. (2) Use the values of FX(xi) as the values of pi for plotting points on normal
probability paper. (3) Calculate zi = Φ-1(pi). (4) Plot zi versus xi on standard linear graph
paper.

Normal Probability Paper

3
Standard normal variable

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
-1

-2

-3
x

xix
C.
P(0.15 ≤ A ≤ 0.2) = FA (0.2) − FA (0.15)
  u k    u k 

= exp −   − exp −  
  0.2     0.15  
   
= 0.3679 − 0.09345
= 0.2745
D.
P(A > 0.3) = 1 − P(A ≤ 0.3)
  u k 
= 1 − exp −  
  0.3  
 
= 1 − 0.7436
= 0.2564
-----------------------------------

Problem 2.10. The occurrence of major snow storms (snowfall greater than 1 foot) for a
particular location is assumed to follow a Poisson distribution. Over 33 years of snowfall
records, such storms have occurred 94 times.
A. What is the probability of 2 or more major storms in the next year?
B. What is the probability of exactly 2 storms occurring in the next 3 years?
C. What is the return period of these storms?
D. What is the probability of more than 3 storms in the next 2 years?

Solution:
The average occurrence rate, ν, of storms is
94
ν= = 2.85 occurrences/year
33
A. The time interval, t, is 1 year.
P( N ≥ 2) = 1 − P( N < 2)
= 1 − [P( N = 0) + P( N = 1)]
 (νt )0 − νt (νt )1 − νt 
= 1−  e + e 
 0! 1! 
−ν −ν
= 1 − e − νe
= 0.777

xx
B. The time interval, t, is 3 years.
( ν t ) 2 − νt
P( N = 2) = e
2!

=
[(2.85)(3)]2 e−( 2.85)(3)
2!
= 7.07 x10− 3
C. The return period is defined as the inverse of the average occurrence rate.
1
τ= = 0.351 years/occurrence
ν
D. The time interval, t, is 2 years.
P( N > 3) = 1 − P( N ≤ 3)
= 1 − [P( N = 0) + P( N = 1) + P( N = 2) + P( N = 3)]
 (νt )0 − νt (νt )1 − νt (νt ) 2 − νt (νt )3 − νt 
= 1−  e + e + e + e 
 0! 1! 2! 3! 
= 0.820

---------------------------------

xxi
 SOLUTIONS TO PROBLEMS IN CHAPTER 3

Problem 3.1. Derive Equation (3.2).

Solution:
The simplest derivation uses the fact that the expectation operator E[ ] is a linear operator.
Therefore,
[
µ Y = E[Y ] = E a 0 + ∑ a i X i ]
[
= E[a 0 ] + E ∑ a i X i ]
= a 0 + ∑ E[a i X i ]
= a 0 + ∑ a i E[X i ]
= a 0 + ∑ a iµ X i

An alternate derivation uses the integral definition of expected value (Eqn. 2.28 extended to
apply to a random vector) and the concept of marginal density functions (Eqn. 2.85). By the
definition of expected value for a function of random variables,
+∞ +∞ +∞
µY = ∫
−∞−∞
∫ ∫ [a + ∑ a x ] f
−∞
0 i i X 1 X 2 X n ( x1 , x 2 ,  , x n ) dx1dx 2  dx n

where f X1X 2 X n ( x1 , x 2 ,, x n ) is the n-dimensional joint density function. By multiplying out
terms and carrying out n-1 of the integrals in each term to get to marginal density functions,
we get
+∞
µ Y = a 0 + ∑ ∫ a i x if X i ( x i ) dx i
−∞

= a 0 + ∑ a iµ X i
---------------------------------

Problem 3.2. Derive Equation (3.3).

Solution:
The simplest derivation uses the fact that the expectation operator E[ ] is a linear operator.
Therefore,

xxii
 Courtesy of CRC Press/Taylor & Francis Group

Stress Standard
cylinder

Stress

f ΄c

6 12 Failure

Strain

Figure 2.1  Concrete cylinder test considered in Example 2.1.

002x001.eps
 Courtesy of CRC Press/Taylor & Francis Group

Sample space Event Random variable

Real numbers
Interval
Figure 2.2  Schematic representation of a random variable as a function.

002x002.eps
 Courtesy of CRC Press/Taylor & Francis Group

pX(x)

1.0
0.65

0.5
0.20
0.05 0.10
0.0 x
1 2 3 4
Figure 2.3  A probability mass function.

002x003.eps
 Courtesy of CRC Press/Taylor & Francis Group

FX(x)

1.00
1.0
0.90

0.5
0.25
0.05
0.0 x
1 2 3 4
Figure 2.4  A CDF for a discrete random variable.

002x004.eps
 Courtesy of CRC Press/Taylor & Francis Group

4.0

3.0

2.0

1.0

0.6 0.8 1.0 1.2 1.4

Figure 2.5  Example of a PDF.

002x005.eps
 Courtesy of CRC Press/Taylor & Francis Group

FX(x)

0
0.6 0.8 1.0 1.2 1.4
Figure 2.6  Example of a CDF.

002x006.eps
 Courtesy of CRC Press/Taylor & Francis Group

fx(x)

a x
Figure 2.7  Relationship between CDF and PDF described by Equation 2.13.

002x007.eps
 Courtesy of CRC Press/Taylor & Francis Group

fX(x)

a b x
Figure 2.8  Graphical representation of F X (b) − F X (a) in Equation 2.15.

002x008.eps
 Courtesy of CRC Press/Taylor & Francis Group

fX(x)

PDF
1
b–a

x
0 a µX b

FX(x)

1
CDF

0.5

x
0 a b
Figure 2.9  PDF and CDF of a uniform random variable.

002x009.eps
 Courtesy of CRC Press/Taylor & Francis Group

fX(x)

μX x

1.0

0.5

0.0 x
Figure 2.10  PDF and CDF of a normal random variable.

002x010.eps
 Courtesy of CRC Press/Taylor & Francis Group

φ(z)

0 z

Figure 2.11  PDF ϕ(z) for a standard normal random variable.

002x011.eps
 Courtesy of CRC Press/Taylor & Francis Group

Φ(z)

0.5

0 z
Figure 2.12  CDF Ф(z) for a standard normal random variable.

002x012.eps
 Courtesy of CRC Press/Taylor & Francis Group

fX(x) fX(x)
0.8 0.8

0.7 0.7 σ = 1/2

0.6 0.6
μ=0
0.5 0.5
μ=1
0.4 0.4
μ = –1 σ=1
0.3 0.3
σ=2
0.2 0.2

0.1 0.1

0 0
–4 –3 –2 –1 0 1 2 3 4 –4 –3 –2 –1 0 1 2 3 4

FX(x) FX(x)
σ = 1/2
1 1
μ = –1

σ=1 σ=2
μ=1
μ=0
0 0
–4 –3 –2 –1 0 1 2 3 4 –4 –3 –2 –1 0 1 2 3 4
Figure 2.13  Examples of PDFs and CDFs for normal random variables.

002x013.eps
 Courtesy of CRC Press/Taylor & Francis Group

fX(x)

μX
x
x x
Figure 2.14  Normal random variable PDF is symmetrical about the mean.

002x014.eps
 Courtesy of CRC Press/Taylor & Francis Group

fX(x)

0
1000 1500 2000 x
Figure 2.15  PDF of normal random variable in Example 2.4.

002x015.eps
 Courtesy of CRC Press/Taylor & Francis Group

fX(x)

x
Figure 2.16  PDF of a lognormal random variable.

002x016.eps
 Courtesy of CRC Press/Taylor & Francis Group

fX(x)

k=1

k=2
k=3

x
Figure 2.17  PDFs of gamma random variables.

002x017.eps
 Courtesy of CRC Press/Taylor & Francis Group

fX(x)

x
Figure 2.18  PDF of an extreme Type I random variable.

002x018.eps
 Courtesy of CRC Press/Taylor & Francis Group

fX(x)

x
Figure 2.19  PDF for an extreme Type II random variable.

002x019.eps
 Courtesy of CRC Press/Taylor & Francis Group

Fx(x)

0 x
Figure 2.20  The S-shaped CDF for a normal random variable.

002x020.eps
 Courtesy of CRC Press/Taylor & Francis Group

0.9999

0.9990 3
0.9980

0.9900
0.9800 2

0.9500
0.9000
1

Standard normal variate

0.8000
0.7000
Probability

0.6000
0.5000 0
0.4000
0.3000
0.2000
–1
0.1000
0.0500

0.0200 –2
0.0100
0.0050
0.0020
0.0010 –3
0.0005

0.0001
Figure 2.21  Example of normal probability paper.

002x021.eps
 Courtesy of CRC Press/Taylor & Francis Group

0.995 Normal
probability
0.990 scale
2 0.980

0.950
0.900
0.841
1
0.800
Standard normal variable

0.700
σX
0.600
0 0.500
0 1 2 3 4 5 6 7
0.400
σX
0.300
0.200 0.159 µX
–1
0.100
0.050
–2 0.020
0.010
0.005

–3

Figure 2.22 Interpretation of a straight line plot on normal probability paper in terms of the
mean and standard deviation of the normal random variable.

002x022.eps
 Courtesy of CRC Press/Taylor & Francis Group

0.9999

0.9990 3
0.9980

0.9900 _
x = 6.2, s X= 0.62
0.9800 2

0.9500

0.9000
1
0.8000

Standard normal variate

Probability

0.7000
0.6000
0.5000 0
0.4000
0.3000
0.2000
–1
0.1000

0.0500
Start here,
0.0200 –2
minimum data point
0.0100
0.0050
0.0020
0.0010 –3
0.0005

0.0001
0 1 2 3 4 5 6 7 8 9
Figure 2.23  Data from Example 2.7 plotted on normal probability paper.

002x023.eps
 Courtesy of CRC Press/Taylor & Francis Group

0.9999

0.9990 3
0.9980

0.9900
0.9800 2

0.9500
0.9000
1

Standard normal variate

0.8000
0.7000
Probability

0.6000
0.5000 0
0.4000
0.3000
0.2000
–1
0.1000

0.0500

0.0200 –2
0.0100
0.0050
0.0020
0.0010 –3
0.0005

0.0001
0 50 100 150 200
Gross vehicle weight (kips)
Figure 2.24  Observed data on GVW plotted on normal probability paper (1 kip = 4.448 kN).

002x024.eps
 Courtesy of CRC Press/Taylor & Francis Group

0.9999

0.9990 3
0.9980

0.9900
0.9800 2

0.9500
0.9000
1

Standard normal variate

0.8000
0.7000
Probability

0.6000
0.5000 0
0.4000
0.3000
0.2000
–1
0.1000

0.0500

0.0200 –2
0.0100
0.0050
0.0020
0.0010 –3
0.0005

0.0001
44 48 52 56
Yield stress (ksi)
Figure 2.25  Test results of yield stress of steel on commercial normal probability paper.

002x025.eps
 Courtesy of CRC Press/Taylor & Francis Group

0.9999

0.9990 3
0.9980

0.9900
0.9800 2

0.9500
0.9000
1

Standard normal variate

0.8000
0.7000
Probability

0.6000
0.5000 0
0.4000
0.3000
0.2000
–1
0.1000

0.0500

0.0200 –2
0.0100
0.0050
0.0020
0.0010 –3
0.0005

0.0001
1 2 3 4 5 6 7 8
Shear strength (ksi)
Figure 2.26  Test results of shear strength of spot welds on normal probability paper.

002x026.eps
 Courtesy of CRC Press/Taylor & Francis Group

14

12

Relative frequency (%) 10

0
1500 2000 2500 3000 3500 4000 4500
Concrete strength, f΄c (psi), 150 psi interval
Figure 2.27  Relative frequency histogram for concrete strength (Example 2.9).

002x027.eps
 Courtesy of CRC Press/Taylor & Francis Group

100

80

Cumulative frequency (%)

60

40

20

0
1500 2000 2500 3000 3500 4000 4500
Concrete strength, f΄c (psi), 150 psi interval
Figure 2.28  Cumulative frequency histogram for concrete strength (Example 2.9).

002x028.eps
 Courtesy of CRC Press/Taylor & Francis Group

(a) 8 100

Cumulative frequency (%)

7

Relative frequency (%)

80
6
5 60
4
3 40

2
20
1
0 0
1500 2000 2500 3000 3500 4000 4500 1500 2000 2500 3000 3500 4000 4500

(b) 60 100

Cumulative frequency (%)

Relative frequency (%)

50 80
40
60
30
40
20

10 20

0 0
1500 2000 2500 3000 3500 4000 4500 1500 2000 2500 3000 3500 4000 4500

Figure 2.29 Influence of interval size on appearance of histogram (Example 2.9) (a) for interval
of 50 psi and (b) for interval of 1000 psi.

002x029.eps
 Courtesy of CRC Press/Taylor & Francis Group

35

30

25
Relative frequency (%)
20

15

10

0
50 100 150
Gross vehicle weight (kips)
Figure 2.30  Relative frequency histogram for data in Table 2.4 (Example 2.10).

002x030.eps
 Courtesy of CRC Press/Taylor & Francis Group

100

80

Cumulative frequency (%)

60

40

20

0
50 100 150
Gross vehicle weight (kips)
Figure 2.31  Cumulative frequency histogram for data in Table 2.4 (Example 2.10).

002x031.eps
 Courtesy of CRC Press/Taylor & Francis Group

= Intersection of sets

= Union of sets
Figure 2.32 A Venn diagram showing the difference between the intersection and union of
events.

002x032.eps
 Courtesy of CRC Press/Taylor & Francis Group

0.35

0.30

Relative frequency
0.25

0.20

0.15

0.10

0.05

2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 4.00

Random variable X1 = modulus of elasticity (106 psi)
0.40
0.35
0.30
Relative frequency

0.25
0.20
0.15
0.10
0.05

1.5 2.0 2.5 3.0 3.5 4.0 4.5

Random variable X2 = compressive strength (103 psi)
Figure 2.33  Relative frequency histograms for X1 and X2 considered independently.

002x033.eps
 Courtesy of CRC Press/Taylor & Francis Group

4.5

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.02

psi)
4.0
rength (10 3
0.00 0.00 0.00 0.00 0.02 0.02 0.01 0.02
3.5
compressive st

0.00 0.00 0.00 0.04 0.06 0.09 0.01 0.00

3.0

0.00 0.00 0.02 0.15 0.15 0.04 0.00 0.00

2.5
X2, maximum

0.02 0.04 0.09 0.12 0.03 0.00 0.00 0.00

2.0

0.00 0.04 0.01 0.00 0.00 0.00 0.00 0.00

1.5
2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 4.00
X1, modulus of elasticity (106 psi)
Figure 2.34  Relative frequency histogram for both X1 and X2.

002x034.eps
 Courtesy of CRC Press/Taylor & Francis Group

1.00 1.00

0.98 0.90
0.96
0.94
0.91 0.80
0.79 0.90
0.70

Cumulative frequency
0.77
0.60
0.71
0.67 0.50
0.53
0.40

0.49 0.30

0.20
0.22
0.35 0.10
0.32
0.12
0.00
h
0.20 gt
en

5
4.
tr
es

0
0.10

4.
iv

5
ss

3.
0.02 e
3.
0 pr
m si)
0.04
5
0.00 co 3 p
2.

0
um (1
0
2.

2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 4.00

im
X1, modulus of elasticity (106 psi) ax
,m
X2
Figure 2.35  Cumulative frequency histogram for both X1 and X2.

002x035.eps
 Courtesy of CRC Press/Taylor & Francis Group

X2 X2

X1 X1

Linearly correlated random variables,

ρ<0
Uncorrelated random variables

X2 X2

X1 X1
Linearly correlated random Linearly perfectly correlated random
variables, ρ > 0 variables, ρ = 1
X2 X2

X1 X1
Linearly correlated random Nonlinearly correlated random
variables, ρ = –1 variables
Figure 2.36  Examples of correlated and uncorrelated random variables.

002x036.eps