11/02/2015

PROBABILITY METHODS IN ENGINEERING

Chapter

Scatter diagram, linear regression

and correlation
Goodness of fit, Fitting a straight
line, parabola, and other types

PROBABILITY & STATISTICS

BY

Weekly Fuel Consumption y

(MMcf)

13
11

ARSHAD ALI

The Simple Linear Regression

The dependent (or response) variable is the
variable we wish to understand or predict

The independent (or predictor) variable is the

variable we will use to understand or predict
the dependent variable

Regression analysis is a statistical technique

that uses observed data to relate the
dependent variable to one or more
independent variables

9
7
5
20

30

40

50

60

70

CIVIL ENGINEERING BUITEMS, QUETTA

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

The objective of regression analysis is to build

a regression model (or predictive equation)
that can be used to describe, predict and
control the dependent variable on the basis of
the independent variable

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

Objective of Regression Analysis

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

Scatter Plot and Linear Regression Model

ARSHAD ALI ([email protected])

ARSHAD ALI ([email protected])

Regression
One of the main purposes of curve fitting is to estimate one of the variables
(the dependent variable) from the other (the independent variable). The
process of estimation is often referred to as regression. If y is to be
estimated from x by means of some equation, we call the equation a
regression equation of y on x and the corresponding curve a regression

curve of y on x.

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

11/02/2015

Generally, more than one curve of a given type will appear to fit a set of data. To
avoid individual judgment in constructing lines, parabolas, or other approximating
curves, it is necessary to agree on a definition of a best fitting line, best-fitting
parabola, etc.
To motivate a possible definition, consider following figure in which the data points
are (x1, y1), . . . , (xn, yn).

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

For a given value of x, say, x1, there will be a difference between the value y1and the
corresponding value as determined from the curve C. We denote this difference by d1,
which is sometimes referred to as a deviation, error, or residual and may be positive,
negative, or zero. Similarly, corresponding to the values x2, . . . , xn, we obtain the
deviations d2, . . . , dn

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

The Method of Least Squares

ARSHAD ALI ([email protected])

10

y na b x
xy a x b x
CIVIL ENGINEERING BUITEMS, QUETTA

11

ARSHAD ALI ([email protected])

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

Comparison of Correlation Coefficient

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

12

ARSHAD ALI ([email protected])

11/02/2015

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

13

14

The value of r is such that -1 < r < +1. The + and

signs are used for positive linear correlations and
negative linear correlations, respectively

Negative correlation: If x and y have a strong negative linear correlation, r is

close
to -1. An r value of exactly -1 indicates a perfect negative
fit. Negative values
indicate a relationship between x and y such
that as values for x increase, values for y decrease.

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

If x and y have a strong positive linear correlation, r is

close

to +1. An r value of exactly +1 indicates a

perfect positive fit. Positive values

indicate a

relationship between x and y variables

such that as values for x increases,
values for y also increase.

CIVIL ENGINEERING BUITEMS, QUETTA

17

ARSHAD ALI ([email protected])

ARSHAD ALI ([email protected])

16

positive correlation: If x and y have a strong positive linear correlation, r is

close to +1. An r value of exactly +1 indicates a perfect positive fit. Positive
values
indicate a relationship between x and y variables such that as
values for x increases, values for y also increase.

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

15

Positive Correlation

CIVIL ENGINEERING BUITEMS, QUETTA

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

18

ARSHAD ALI ([email protected])

11/02/2015

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

19

20

Regression ( A dictionary Meaning)

the act of going back to a previous place or

state; return or reversion.
2.retrogradation; retrogression.
3.Biology. reversion to an earlier or less advanced
state or form or to a common or general type.
4.Psychoanalysis. the reversion to a chronologically
earlier or less adapted pattern of behavior and
feeling.
5.a subsidence of a disease or its manifestations:

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

One of the main purposes of curve fitting is to estimate one of

the variables (the dependent variable) from the other (the
independent variable).

ARSHAD ALI ([email protected])

22

Example Problem

Average Hourly
Week Temperature x
(deg F)

Weekly Fuel
Consumption y
(MMcf)

28

12.4

28

11.7

32.5

12.4

39

10.8

45.9

9.4

57.8

9.5

58.1

Weekly Fuel Consumption y (MMcf)

13
12
11

FUELCON

The process of estimation is often referred to as regression. If

y is to be estimated from x by means of some equation, we
call the equation a regression equation of y on x and the
corresponding curve a regression curve of y on x.

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

21

Regression

CIVIL ENGINEERING BUITEMS, QUETTA

10
9
8
7
6
5
20

30

40

50

60

70

TEMP

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

Least Square Model for Straight Line

y a bx

xy ax bx 2

y a bx
y a bx

xy ax bx
xy ax bx

y na b x

xy a x b x

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

7.5

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

23

Example Problme

62.5

ARSHAD ALI ([email protected])

24

y na b x
xy a x b x

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

11/02/2015

y na b x
y

x2

12.4

28.0

784.00

347.20

11.7

28.0

784.00

327.60

12.4

32.5

1056.25

403.00

10.8

39.0

1521.00

421.20

9.4

45.9

2106.81

431.46

9.5

57.8

3340.84

549.10

8.0

58.1

3375.61

464.80

7.5

62.5

3906.25

468.75

81.7

351.8

16874.8

3413.1

xy

81.7 8 a 351.8 b

xy a x b x

3413.1 351.8 a 16874.8 b

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

25

Example Problem

a 15.84
b 0.1289

y x x xy
2

26

Example Problem

Least Square Model for

2nd

degree polynomial
xy ax bx 2 cx 3

xy ax bx 2 cx 3

y a bx cx 2

xy ax bx

y a bx cx
2

y a bx cx
y na b x c x

xy ax bx

cx 3

xy ax bx cx
2

xy a x b x

c x3

cx 3

xy ax bx cx
2

y a x 2 b x 3 c x 4

y 15.84 0.1289 x
CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

x2

x3

x4

xy

x2y

28.0

12.4

784.00

21952.00

614656.00

347.20

9721.6

28.0

11.7

784.00

21952.00

614656.00

327.60

9172.8

32.5

12.4

1056.25

34328.13

1115664.06

403.00

13097.5

39.0

10.8

1521.00

59319.00

2313441.00

421.20

16426.8

45.9

9.4

2106.81

96702.58

4438648.38

431.46

19804.01

57.8

9.5

3340.84

193100.55

11161211.91

549.10

31737.98

58.1

8.0

3375.61

196122.94

11394742.87

464.80

27004.88

62.5

7.5

3906.25

244140.63

15258789.06

468.75

29296.88

351.8

81.7

16874.8

867617.8

46911809.3

3413.1

156262.4

x y

xy x

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

27

CIVIL ENGINEERING BUITEMS, QUETTA

28

x2

x3

x4

xy

x2y

81.7

351.8

16874.8

867617.8

46911809.3

3413.1

156262.4

y na b x c x
xy a x b x
x

c x3

y a x 2 b x 3 c x 4

a 15.63
b 0.118
c 1.145

ARSHAD ALI ([email protected])

81.7 8a b 351.8 c 16874.8

3413.1 a 351.8 b 16874.8 c 867617.8
156262.4 a 16874.8 b 867617.8 c 46911809.3

351.8
16874.8 a
81.7 8
34131.1 351.8
16874.8
867617.8 b

156262.4 16874.8 867617.8 46911809.3 c

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

29

ARSHAD ALI ([email protected])

81.7 8a b 351.8 c 16874.8

3413.1 a 351.8 b 16874.8 c 867617.8
156262.4 a 16874.8 b 867617.8 c 46911809.3

y a bx cx 2
y 15.63 0.118 x 1.145 x 2

ARSHAD ALI ([email protected])

30

a 15.63
b 0.118
c 1.145

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

11/02/2015

ARSHAD ALI ([email protected])

35

ARSHAD ALI ([email protected])

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

CIVIL ENGINEERING BUITEMS, QUETTA

33

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

CIVIL ENGINEERING BUITEMS, QUETTA

31

CIVIL ENGINEERING BUITEMS, QUETTA

32

ARSHAD ALI ([email protected])

34

ARSHAD ALI ([email protected])

36

ARSHAD ALI ([email protected])

11/02/2015

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

12x16y=5

38

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

39

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

37

40

16x+106y=47

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

PROBLEM

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

41

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

42

Problem Extension of Steel Wires

Ten steel wires of diameter 0.5 mm and length 2.5 m were extended in a laboratory
by applying vertical forces of varying magnitudes. Results are as follows:

F kg

15

19

25

35

42

48

53

56

62

65

mm

1.7

2.1

2.5

3.4

3.9

4.9

5.4

5.7

6.6

7.2

(a)

Estimate the parameters of a simple linear regression model with force as the
explanatory variable.

(b)

Find 95% condence limits for the two parameters.

(c)

Make Microsoft Excel calcs of the same and compare with the manual Clacs

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

11/02/2015

USING MICROSOFT
EXCEL

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

Step 2
Go to insert

44

Step 1
Enter Data in Microsoft Excel

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

45

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

43

46

Select Data

Select Scatter Chart

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

47

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

48

7.2

5.7

6.6

5
5.4

3.4

4.9
3.9

2.1

3
2

2.5

Step 3

1.7

0
Left Click on scatter data, right click & select
Add Data Labels
CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

10

20

CIVIL ENGINEERING BUITEMS, QUETTA

30

40

50

60

70

ARSHAD ALI ([email protected])

11/02/2015

Step 4
Right Click on data points and
Select Add Trend line

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

7.2

5.7

6.6

5
5.4

3.4

4.9
3.9

2.1

50

Step 5
Select Linear Model

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

51

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

49

52

2
2.5

1.7

0
0

10

20

30

40

50

60

Step 6

70

Right Click on Regression Line

Select Format Trend line

ARSHAD ALI ([email protected])

Step 7

Check Display Eq
Check R squared

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

53

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

54

7.2

y = 0.1062x - 0.1212
R = 0.9833

6
mm

CIVIL ENGINEERING BUITEMS, QUETTA

5.7
6.6

5.4

3.4

4.9
3.9

2.1

3
2

2.5

1.7

0
0

10

20

30

40

50

60

70

Force (kg)
CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

11/02/2015

Linear Regression
Equation

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

55

y a bx

y a bx
y

na b x i

x y

a xi b xi2

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

15

yi

xi yi

xi2

1.7

25.5

225

19

2.1

39.9

361

25

2.5

62.5

625

35

3.4

119

1225

42

3.9

163.8

1764

48

4.9

235.2

2304

53

5.4

286.2

2809

56

5.7

319.2

3136

62

6.6

409.2

3844

Regression Equations

y
x y
i

2128.5 420a 20518b

7.2

468

4225

43.4

2128.5

20518

x y

a xi b xi2

43.4 10a 420b

65

na b x i

Substituting Values

420

ARSHAD ALI ([email protected])

58

10a 420b 43.4

420a 20518b 2128.5

Arrange equation variables for elimination

217
10a 420b

5
4257
420a 20518b

2
i

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

Multiply10a 420b

217
by 42
5

9114
420a 17640b
5
therefore the set of equations become

9114
420a 17640b
5
4257
420a 20518b
2

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

59

CIVIL ENGINEERING BUITEMS, QUETTA

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

57

xi

56

Step 1: Make Equations needed to find a and b

ARSHAD ALI ([email protected])

60

Subtructing as follows
4257
2
9114
420a 17640b
5
-

420a 20518b

2878b

ARSHAD ALI ([email protected])

CIVIL ENGINEERING BUITEMS, QUETTA

3057
10
3057
0.10622
28780

ARSHAD ALI ([email protected])

10

11/02/2015

420a 17640

Substitute the value of b as follows

3057 9114

28780
5
420a

9114
3057
17640
5
28780

420a

9114 2696274

5
1439

420a

ARSHAD ALI ([email protected])

y= a + bx
y 0.1212 0.10622 x

1.7

19

2.1

Y 0.1212 0.10622 19 1.90

25

2.5

Y 0.1212 0.10622 25 2.53

35

3.4

Y 0.1212 0.10622 35 3.60

42

3.9

Y 0.1212 0.10622 42 3.60

48

4.9

Y 0.1212 0.10622 35 3.60

53

5.4

Y 0.1212 0.10622 35 3.60

56

5.7

Y 0.1212 0.10622 35 3.60

62

6.6

Y 0.1212 0.10622 35 3.60

65

7.2

Y 0.1212 0.10622 35 3.60

Y 0.1212 0.10622 15 1.47

CIVIL ENGINEERING BUITEMS, QUETTA

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

63

Y 0.1212 0.10622 x

15

ARSHAD ALI ([email protected])

Evaluating the values of all determiniants

50 2100
50 41036 2100 840 287800
840 41036

217 2100
217 41036 2100 4257 34888
4257 41036

64

50a 2100b
217
840a 41036b 4257

Writing the matrix Form

50 2100 a 217
840 41036 b 4257

Find coefficient matrix (D), A matrix (Da) and B matrix (Db)

217 2100
Da

4257 41036

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

65

ARSHAD ALI ([email protected])

Alternate Method for solving

50 2100
D

840 41036

Alternate Method for solving

det ( Dy )

Hence the Regression Equation is as follows

366324 1
4361

0.1212
7195 420 35975

Finding Y at given value of x by using Regression Equation

det ( Dx )

62

366324
7195

CIVIL ENGINEERING BUITEMS, QUETTA

det ( D)

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

61

9114
420a 17640 b
5

50 217
Db

840 4257

ARSHAD ALI ([email protected])

66

Alternate Method for solving

Using Cramer Rule

det ( Dx ) 34888
4361

0.1212
det ( D) 287800
35975

det ( Dy )
det ( D)

30570
3057

0.10622
287800 28780

50 217
50 4257 217 840 30570
840 4257

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

11

11/02/2015

Correlation

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

67

Defined as

covariance x, y
Var x Var y

ARSHAD ALI ([email protected])

XY

X2

X2Y

X3

X4

15

1.7

25.5

225

382.5

3375

50625

19

2.1

39.9

361

758.1

6859

130321

25

2.5

62.5

625

1562.5

15625

390625

35

3.4

119

1225

4165

42875

1500625

42

3.9

163.8

1764

6879.6

74088

3111696

48

4.9

235.2

2304

11289.6

110592

5308416

53

5.4

286.2

2809

15168.6

148877

7890481

56

5.7

319.2

3136

17875.2

175616

9834496

62

6.6

409.2

3844

25370.4

238328

14776336

65

7.2

468

4225

30420

274625

17850625

420

43.4

2128.5

20518

113871.5

1090860

60844246

xy

CIVIL ENGINEERING BUITEMS, QUETTA

xy2
x2 y2

x 2 x 2 y 2 y 2
i i i i
n
n n
n
n XY X Y
n X 2 X 2 n Y 2 Y 2

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

69

Table for Calculations

x y x
i

Simplifies to

CIVIL ENGINEERING BUITEMS, QUETTA

68

Coefficient of Correlation

ARSHAD ALI ([email protected])

70

Coefficient of Correlation

n XY X Y
n X 2 X 2 n Y 2 Y 2

Substituting

X 420 , Y 43.4 , X Y 2128.5, X

r

20518 , Y 2 221.38

10 2128.5 420 43.4

10 20518 4202 10 221.38 43.4 2

0.9916 or 99.16%

ARSHAD ALI ([email protected])

PROBLEM

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

71

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

72

Problem

Consider the lab-measured water content in randomly selected field soil specimens
from a particular site, and the corresponding water content. estimated by a fast,
inexpensive method which measures the gas pressure created when the soil is
mixed with a chemical which reacts with water. If it is sufficiently accurate, the
second method will provide an inexpensive way to obtain more frequent watercontent samples during the quality control of soil compaction on a highway project.
The purpose is to predict true water content given an observed value of the fast
test. Find the estimated regression line using the following data

n 67, x 13.8 , y 13.8 , xi yi 2128.5 , x i2 13260

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

12

11/02/2015

na b x i

x y

a xi b xi2

x 13.8

x
x

Substituting Values

We get

13.8

13.8

67

n 67, x 13.8 , y 13.8

x y

13145 , x i 13260

13.8 67

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

73

74

67 a 924.6 b 924.6
924.6 a 13260 b 13145

STEP 1: Get rid of the fractions.by multiplying first equation by 5 and second
equation by 10. After multiplying we have the following system:

924.6

335 x 4623 y
4623
9246 x 132600 y 131455

Similarly

924.6 67 a b 924.6

924.6

13145 a 924.6 b 13260

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

STEP 2: Find coefficient matrix (D), X matrix (Dx) and Y matrix (Dy). In this
example we have.

4623
335
D

9246 132600
4623
4623
Dx

131455 132600

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

75

CIVIL ENGINEERING BUITEMS, QUETTA

4623
335
Dy

9246 131455

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

STEP 4: Solve for the a and b using Cramer Rule.

det ( Dx ) 5293335

3.1569
det ( D ) 1676742

det ( D y )
det ( D )

1293167
0.771
1676742

y a bx
CIVIL ENGINEERING BUITEMS, QUETTA

The estimated regression model

76

STEP 3: Find determinants for D, Dx and Dy.

det ( D )

335
4623
335 132600 4623 9246 1676742
9246 132600

det ( Dx )

4623
4623
4623 132600 4623 131455 5293335
131455 132600

det ( D y )

335
4623
335 131455 4623 9246 1293167
9246 131455

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

77

ARSHAD ALI ([email protected])

ARSHAD ALI ([email protected])

78

PROBLEM

y 3.1569 0.771x
ARSHAD ALI ([email protected])

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

13

11/02/2015

health effects research has prompted the Environmental Protection Agency

to reduce allowable arsenic levels in drinking water so that many water
systems are no longer compliant with standards. This has spurred interest
in the development of methods to remove arsenic. The accompanying data
on and arsenic removed (%) by a particular process was read from a
scatter plot in the article Optimizing Arsenic Removal During Iron
Removal: Theoretical and Practical Considerations

80

Scatter Diagram

80
70

Arsenic remove d %

Arsenic is found in many ground-waters and some surface waters. Recent

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

79

Problem: Ground water Analyis

60
50
40
30
20
10

(J. of Water Supply Res. and Tech., 2005: 545560).

6.5

7.5

X 7.01 7.11 7.12 7.24 7.94 7.94 8.04 8.05 8.07 8.9 8.94 8.95 8.97 8.98 9.85 9.86 9.86 9.87
Y

60 67 66 52 50 45 52 48 40 23 20 40 31 26

CIVIL ENGINEERING BUITEMS, QUETTA

22 13

ARSHAD ALI ([email protected])

ARSHAD ALI ([email protected])

X 152.7
Y 671
X Y 5380.84
, X 2 1312.6764

9.5

10

10.5

ARSHAD ALI ([email protected])

Step 1: Find XY and X2 as it was done in the table below.

X
7.01
7.11
7.12
7.24
7.94
7.94
8.04
8.05
8.07
8.9
8.94
8.95
8.97
8.98
9.85
9.86
9.86
9.87

Y
60
67
66
52
50
45
52
48
40
23
20
40
31
26
9
22
13
7

XY
420.6
476.37
469.92
376.48
397
357.3
418.08
386.4
322.8
204.7
178.8
358
278.07
233.48
88.65
216.92
128.18
69.09

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

83

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

Step 2: Find the sum of every column:

81

CIVIL ENGINEERING BUITEMS, QUETTA

8.5

x p-H

82

XX
49.1401
50.5521
50.6944
52.4176
63.0436
63.0436
64.6416
64.8025
65.1249
79.21
79.9236
80.1025
80.4609
80.6404
97.0225
97.2196
97.2196
97.4169

YY
3600
4489
4356
2704
2500
2025
2704
2304
1600
529
400
1600
961
676
81
484
169
49

ARSHAD ALI ([email protected])

84

y
x y
i

na b x i

a xi b x

n 18
2
i

X 152.7
Y 671

Substituting the values of

671 18 a 152.7 b
5380.84 152.7 a 1312.6764 b

X Y 5380.84
, X 2 1312.6764

31231

31231

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

14

11/02/2015

18 a 152.7 b 671
152.7 a 1312.6764 b 5380.84

18a

1527
b 671
10

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

85

Re organizing Equations

86

STEP 1: Get rid of the fractions.by multiplying first equation by 10 and second
equation by 2500.

180 x 1527 y 6710

381750 x 3281691y 13452100

1527
3281691
134521
a
b
10
2500
25

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

STEP 2: Find coefficient matrix (D), X matrix (Dx) and Y matrix (Dy). In this
example we have.

1527
180
D

381750 3281691
1527
6710
Dx

13452100 3281691

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

87

CIVIL ENGINEERING BUITEMS, QUETTA

6710
180
Dy

381750 13452100

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

STEP 4: Solve for the x and y.

det ( Dx ) 1478789910
a

190.27
det ( D)
7772130

det ( Dy )
det ( D)

140164500
18.03
7772130

88

STEP 3: Find determinants for D, Dx and Dy.

det ( D)

180
1527
180 3281691 1527 381750 7772130
381750 3281691

det ( Dx )

6710
1527
6710 3281691 1527 13452100 1478789910
13452100 3281691

det ( Dy )

180
6710
180 13452100 6710 381750 140164500
381750 13452100

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

89

ARSHAD ALI ([email protected])

ARSHAD ALI ([email protected])

Another Direct Method to find a and b

90

Y X X XY
n X X
2

6711312.6764 152.7 5380.84

18 1312.6764 152.7 2
190.268

n XY X Y
n X 2 X

18 5380.84 152.7 671

18 1312.6764 152.7

18.034
CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

15

11/02/2015

y
y

91

a b x
190.268 18.034 x

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

Step 4: Substitute a and b in regression equation formula

92

Correlation Coefficient

n XY X Y
n X 2 X 2 n Y 2 Y 2

18 5380.84 152.7 671

18 1312.6764 152.7 2 18 31231 6712

0.9505
CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

PROBLEM

CIVIL ENGINEERING BUITEMS, QUETTA

CIVIL ENGINEERING BUITEMS, QUETTA

95

ARSHAD ALI ([email protected])

ARSHAD ALI ([email protected])

94

Japans high population density has resulted in a multitude of resource-usage

problems. One especially serious difficulty concerns waste removal. The article
Innovative Sludge Handling Through Pelletization Thickening (Water Research,
1999: 32453252) reported the development of a new compression machine for
processing sewage sludge. An important part of the investigation involved
relating the moisture content of compressed pellets (y, in %) to the machines
filtration rate (x, in kg-DS/m/hr). The following data was read from a graph in the
article:
x 125.3 98.2 201.4 147.3 145.9 124.7 112.2 120.2 161.2 178.9 159.5 145.8 75.1 151.4 144.2 125 198.8 132.5 159.6 110.7
y

77.9 76.8 81.5 79.8 78.2 78.3 77.5 77 80.1 80.2 79.9 79 76.7 78.2 79.5 78.1 81.5 77

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

Scatter Plot

ARSHAD ALI ([email protected])

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

93

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

Step 1: Find XY , X2 and Y2 as it was done in the table below.

96

XY

XX

YY

125.3

77.9

9760.87

15700.09

6068.41

98.2

76.8

7541.76

9643.24

5898.24

201.4

81.5

16414.1

40561.96

6642.25

147.3

79.8

11754.54

21697.29

6368.04

145.9

78.2

11409.38

21286.81

6115.24

124.7

78.3

9764.01

15550.09

6130.89

112.2

77.5

8695.5

12588.84

6006.25

120.2

77

9255.4

14448.04

5929

161.2

80.1

12912.12

25985.44

6416.01

178.9

80.2

14347.78

32005.21

6432.04

159.5

79.9

12744.05

25440.25

6384.01

145.8

79

11518.2

21257.64

6241

75.1

76.7

5760.17

5640.01

5882.89

151.4

78.2

11839.48

22921.96

6115.24

144.2

79.5

11463.9

20793.64

6320.25

125

78.1

9762.5

15625

6099.61

198.8

81.5

16202.2

39521.44

6642.25

132.5

77

10202.5

17556.25

5929

159.6

79

12608.4

25472.16

6241

110.7

78.6

8701.02

12254.49

6177.96

CIVIL ENGINEERING BUITEMS, QUETTA

79 78.6

ARSHAD ALI ([email protected])

16

11/02/2015

X 2817.9
Y 1574.8
X Y 222657.88
X

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

97

Step 2: Find the sum of every column to get:

y
x y
i

na b x i

a xi b x

n 20

X 2817.9

2
i

Y 1574.8
Substituting the values of

X Y 222657.88

1574.8 20 a 2817.9 b

415949.85

98

Substituting the values in following equations

X
Y

222657.88 2817.9 a 415949.85 b

ARSHAD ALI ([email protected])

28179
7874
20 x
y
10
5

20 a 2817.9 b 1574.8
2817.9 a 415949.85 b 222657.88

28179
8318997
5566447
x
y
10
20
25

STEP 1: Get rid of the fractions.by multiplying first equation by 10 and

second equation by 100.

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

99

ARSHAD ALI ([email protected])

STEP 2: Multiply each equation by a number that will create opposite coefficients

for x. In this case we will multiply first equation by -28179 and the second equation
by 20.

5635800a 794056041b 443762892

5635800a 831899700b 445315760
STEP 3: add the two equations together to eliminate a from the system

37843659b 1552868

281790a 41594985b 22265788

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

STEP 5: substitute the value for b into the original equation to solve for a.

28179 1552868 7874

10 37843659
5

1533899096
21024255

a 72.96

ARSHAD ALI ([email protected])

1552868
0.041
37843659

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

101

CIVIL ENGINEERING BUITEMS, QUETTA

124039.58

100

200a 28179b 15748

Solving for a

415949.85

124039.58

CIVIL ENGINEERING BUITEMS, QUETTA

20a

ARSHAD ALI ([email protected])

102

The estimated Regression Equation is

y a bx
y 72.96 0.041x

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

17

11/02/2015

PROBLEM

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

Values of modulus of elasticity (MOE, the ratio of stress, i.e., force per unit area, to
strain, i.e., deformation per unit length, in GPa) and flexural strength (a measure of
the ability to resist failure in bending, in MPa) were determined for a sample of

concrete beams of a certain type, resulting in the following data (read from a graph
in the article Effects of Aggregates and Microfillers on the Flexural Properties of
Concrete, Magazine of Concrete Research, 1997: 8198):

MOE

29.8 33.2 33.7 35.3 35.5 36.1 36.2 36.3 37.5 37.7 38.7 38.8 39.6 41 42.8 42.8 43.5 45.6 46 46.9 48 49.3 51.7 62.6 69.8 79.5

Strength 5.9

7.2

7.3

6.3

8.1

6.8

7.6

6.8

6.5

6.3

7.9

8.2

8.7

7.8

9.7 7.4 7.7 9.7 7.8

80

7.7 11.6 11.3 11.8 10.7

104

Problem Statement

The article Promising Quantitative Nondestructive Evaluation Techniques for

Composite Materials (Materials Evaluation, 1985: 561565) reports on a study to
investigate how the propagation of an ultrasonic stress wave through a substance
depends on the properties of the substance. The accompanying data on fracture
strength (x, as a percentage of ultimate tensile strength) and attenuation ( y, in
neper/cm, the decrease in amplitude of the stress wave) in fiberglass-reinforced
polyester composites was read from a graph that appeared in the article. The simple
linear regression model is suggested by the substantial linear pattern in the scatter
plot

12

30

36

40

45

57

62

67

71

78

93

94

100

105

3.3

3.2

3.4

2.8

2.9

2.7

2.6

2.5

2.6

2.2

2.3

2.1

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

105

Problem Statement

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

103

Use the accompanying Minitab output to obtain the equation of the least squares
line for predicting strength from modulus of elasticity, and then predict strength for a

ARSHAD ALI ([email protected])

106

Problem Statement
No-fines concrete, made from a uniformly graded coarse aggregate and a
cement-water paste, is beneficial in areas prone to excessive rainfall because of
its excellent drainage properties. The article Pavement Thickness Design for NoFines Concrete Parking Lots, J. of Trans. Engr., 1995: 476484) employed a
least squares analysis in studying how y = porosity (%) is related to x = unit
weight (pcf) in concrete specimens. Consider the following representative data:
x

99

101.1

102.7

103

105.4

107

108.7

110.8

112.1

112.4

113.6

113.8

115.1

115.4

120

28.8

27.9

27

25.2

22.8

21.5

20.9

19.6

17.1

18.9

16

16.7

13

13.6

10.8

Obtain the equation of the estimated regression line. Then create a scatter plot of
the data and graph the estimated line. Does it appear that the model relationship
will explain a great deal of the observed variation in y? What happens if the
estimated line is used to predict porosity when unit weight is 135? Why is this not
a good idea?

beam whose modulus of elasticity is 40. Would you feel comfortable using the least
squares line to predict strength when modulus of elasticity is 100? Explain.
CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

For the past decade, rubber powder has been used in asphalt cement to improve
performance. The article Experimental Study of Recycled Rubber-Filled HighStrength Concrete (Magazine of Concrete Res., 2009: 549556) includes a
regression of y = axial Strength Mpa, x = cube strength in Mpa based on the
following sample data: Obtain the Regression line and Coefficient of Correlation
from the given data
x

112.3

97

92.7

86

102

99.2

95.8

103.5

89

86.7

75

71

57.7

48.7

74.3

73.3

68

59.3

57.8

48.5

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

107

Problem Statement

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

108

Problem Statement

Wrinkle recovery angle and tensile strength are the two most important
characteristics for evaluating the performance of cross-linked cotton fabric. An
increase in the degree of crosslinking, as determined by ester carboxyl band
absorbance, improves the wrinkle resistance of the fabric (at the expense of
reducing mechanical strength). The accompanying data on x = absorbance
and y= wrinkle resistance angle was read from a graph in the paper Predicting the
Performance of Durable Press Finished Cotton Fabric with Infrared Spectroscopy
(Textile Res. J., 1999: 145151).
x
y

0.115 0.126 0.183 0.246 0.282 0.344 0.355 0.452 0.491 0.554 0.651
334

342

355

363

365

372

381

392

400

412

420

a. Does the simple linear regression model appear to be appropriate? Explain.

b. What wrinkle resistance angle would you predict for a fabric specimen having an
absorbance of .300?
c. What would be the estimate of expected wrinkle resistance angle when
absorbance is .300?
CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

18

11/02/2015

Calcium phosphate cement is gaining increasing attention for use in bone repair
applications. The article Short-Fibre Reinforcement of Calcium Phosphate Bone
Cement (J. of Engr. in Med., 2007: 203211) reported on a study in which
polypropylene fibers were used in an attempt to improve fracture behavior. The
following data x = fiber weight %age on and y = compressive strength (MPa) was
provided by the articles authors.
x

1.25

1.25

1.25

9.94

11.67

11

13.44

9.2

9.92

9.79

10.99 11.32 12.29

1.25

2.5

2.5

2.5

2.5

8.69

9.91

10.45

2.5

7.5

7.5

7.5

7.5

10

10

10

10

10.25

7.89

7.61

8.07

9.04

6.63

6.43

7.03

7.63

7.35

6.94

7.02

7.67

a. Fit the simple linear regression model to this data. Then determine the
proportion of observed variation in strength that can be attributed to the model
relationship between strength and fiber weight.
b. The average strength values for the six different levels of fiber weight are 11.05,
10.51, 10.32, 8.15, 6.93, and 7.24, respectively. The cited paper included a
figure in which the average strength was regressed against fiber weight.
c. Obtain the equation of this regression line and calculate the corresponding
coefficient of determination. Explain the difference between the r2 value for this
regression and the r2 value obtained in (a).

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

Variations in clay brick masonry weight have implications not only for structural and
acoustical design but also for design of heating, ventilating, and air conditioning
systems. The article Clay Brick Masonry Weight Variation (J. of Architectural Engr.,
1996: 135137) gave a scatter plot of y = mortar dry density (lb/ft3 ) versus x = mortar
air content (%) or a sample of mortar specimens, from which the following
representative data was read:
x

5.7

6.8

9.6

10

10.7

12.6

14.4

15

119

121.3

118.2

124

112.3

114.1

112.2

115.1

15

16.2

17.8

19

19.7

20.6

25

111

107.2

108.9

108

111

106.2

105

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

Corrosion of steel reinforcing bars is the most important durability problem for
reinforced concrete structures. Carbonation of concrete results from a chemical
reaction that lowers the pH value by enough to initiate corrosion of the rebar.
Representative data on x = carbonation depth (mm) and y = strength (MPa) for a
sample of core specimens taken from a particular building follows (read from a plot in
the article The Carbonation of Concrete Structures in the Tropical Environment of
Singapore, Magazine of Concrete Res., 1996: 293300).
x
y
x
y

8
22.8
38
19.5

15
27.2
40
12.4

16.5
23.7
45
13.2

CIVIL ENGINEERING BUITEMS, QUETTA

20
17.1
50
11.4

20
21.5
50
10.3

27.5
18.6
55
14.1

30
16.1
55
9.7

30
23.4
59
12

35
13.4
65
6.8

ARSHAD ALI ([email protected])

The accompanying data was read from a graph that appeared in the article
Reactions on Painted Steel Under the Influence of Sodium Chloride, and
Combinations Thereof (Ind. Engr. Chem. Prod. Res. Dev., 1985: 375378). The
independent variable is SO2 deposition rate (mg/m2 /d), and the dependent
variable is steel weight loss (g/m2 ).

14

18

40

43

45

112

280

350

470

500

560

1200

a) Construct a scatter plot. Does the simple linear regression model appear to be
reasonable in this situation?
b) Calculate the equation of the estimated regression line.
c) What percentage of observed variation in steel weight loss can be attributed to the
model relationship in combination with variation in deposition rate?
d) Because the largest x value in the sample greatly exceeds the others, this observation
may have been very influential in determining the equation of the estimated line. Delete
this observation and recalculate the equation.
e) Does the new equation appear to differ substantially from the original one (you might
consider predicted values)?
ARSHAD ALI ([email protected])

112

Problem Statement
During oil drilling operations, components of the drilling assembly may suffer from sulfide
stress cracking. The article Composition Optimization of High-Strength Steels for
Sulfide Cracking Resistance Improvement (Corrosion Science, 2009: 28782884) reported
on a study in which the composition of a standard grade of steel was analyzed. The
following data on y = threshold stress (% SMYS) and x = yield strength (MPa) was read
from a graph in the article (which also included the equation of the least squares line).

635

644

711

708

836

820

810

870

856

923

878

937

948

100

93

88

84

77

75

74

63

57

55

47

43

38

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

113

Problem Statement

110

Problem Statement

CIVIL ENGINEERING BUITEMS, QUETTA

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

111

Problem Statement

PROBABILITY METHODS IN ENGINEERING

PROBABILITY METHODS IN ENGINEERING

109

Problem Statement

ARSHAD ALI ([email protected])

114

Problem Statement

The catch basin in a storm-sewer system is the interface between surface runoff and
the sewer. The catch-basin insert is a device for retrofitting catch basins to improve
pollutant removal properties. The article An Evaluation of the Urban Storm water
Pollutant Removal Efficiency of Catch Basin Inserts (Water Envir. Res., 2005: 500
510) reported on tests of various inserts under controlled conditions for which inflow
is close to what can be expected in the field. Consider the following data, read from a
graph in the article, for one particular type of insert on x = amount filtered (1000s of
liters) and y = % total suspended solids removed.

23

45

68

91

114

136

159

182

205

228

53.3

26.9

54.8

33.8

29.9

8.2

17.2

12.2

3.2

11.1

CIVIL ENGINEERING BUITEMS, QUETTA

ARSHAD ALI ([email protected])

19

11/02/2015

PROBABILITY METHODS IN ENGINEERING

115

Thanks
Note:
If you feel any difficulty in understanding the lectures or if you could not get
the soft copy of the lecture you can email me. I will try my best to answer.

Your suggestions are always welcome to improve the lecture presentations

Personal:
Official:
CIVIL ENGINEERING BUITEMS, QUETTA

[email protected]
[email protected]
ARSHAD ALI ([email protected])

20