LINEAR ALGEBRA F.Y. B. TECH.

(COMP)

JSPM’s
RAJARSHI SHAHU COLLEGE OF ENGINEERING
TATHAWADE, PUNE-33
An Autonomous Institute Affiliated to SavitribaiPhule Pune University

UNIT 1: System of Linear Equations

Contents: Rank of a Matrix, System of Linear Equations, Linear Dependence and Independence,
Linear and Orthogonal Transformations, Application to problems in Engineering.

1.1Introduction:Linear Algebra is a branch of mathematics which is widely used in many

scientific fields as well as in economic field. Many important management decisions today are made
on the basis of linear programming models that utilize hundreds of variables. The essential
information of a linear system can be recorded compactly in a rectangular array called as Matrix.
Therefore study of Matrices and system of linear equations has got lot of importance in linear
algebra.

Definition (Matrix):An ordered set of elements arranged in a rectangular array of rows and n
columns and enclosed by a pair of brackets [ ] is called a matrix of order

 a11 a12 ––
––
––
––
a1n

 ––
a21 a22 a2n
= –– –– –– ––
 –– –– –– –– ––
 am1 am2 –– –– amn 
or where and .

1.2Types of Matrices :
1. Row matrix :A matrix is said to be row matrix in which only one row and any number of
columns.
e.g. A = [1 – 2 7 – 5 1]
2. Column matrix :A matrix is said to be column matrix in which only one column and any
 – 26 
number of rows. e.g. A=
 0
 –4
 7
3. Null matrix :A matrix is said to be null matrix if all elements of matrix are zero.

 00 00 00 
It is denoted as O or Z. e.g. Z= 0 0 0
 0 0 0
4. Transpose of a matrix :A matrix obtained from a given matrix A by interchanging rows and
columns is called transpose of A. It is denoted as A or A e.g. :
T

 17 – 53 2
4
 1

7 –1 2
=  –1 –1
0  then A = 
T
If A 5 –3 0
0 
 2 –1 –9  4 2 0 –9

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 1

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
5. Square matrix : A matrix is said to be a square matrix if number of rows and columns are
equals
 1 –2 3

e.g. A= 4 –5 8

 9 7 0
6. Diagonal matrix :A square matrix is said to be diagonal matrix if all non-diagonal elements
are zeros.

e.g. A=[ ]

7. Scalar matrix:A diagonal matrix is said to be scalar matrix if all diagonal elements are equal.
 a11 a12 a13 
8. Determinant of the matrix : Determinant of a square matrix A= a21  a22 a23 is

 a31 a32 a33 
 a11 a12 a13 

| |  a21 a22 a23 =

 a31 a32 a33 
9. Non-singular matrix :A square matrix is said to be non singular matrix if |A|  0. Otherwise
it is said to be singular matrix.
T
10. Symmetric matrix : A square matrix is said to be symmetric matrix if A = A
 a h g  a h g
a. e.g. : A =   h b f  T  h b f
and A =  
 g f c  g f c
11. Skew Symmetric matrix :A square matrix is said to be Skew symmetric matrix if

Here for all .e.g. :[ ]

Note :Diagonal elements of skew symmetric matrix are zero since, .

12. Unit (Identity) matrix :A square matrix is said to be unit matrix if all diagonal (leading)
elements are unity and allnon diagonal elements are zeros. If unit matrix of order n, it is
denoted as In
 1 0 0
0  ; I2 = 
0
e.g. :I3= 0
1
1
  0 1
 0 0 1
13. Upper triangular matrix :A square matrix is said to be upper triangular matrix if all
elements below leading (principal) diagonal are zeros.

e.g.[ ] [ ]

14. Lower triangular matrix :A square matrix is said to be lower triangular matrix if all
elements above leading (principal) diagonal are zeros.

e.g.:[ ] [ ]

15. Minor of an element:Minor of an element aijof matrix A is denoted by Mijand is defined as

the determinant of a sub-matrix obtained by omitting the row and column of A.

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 2

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
a11 a12 a13
e.g. if A=a21 a22 a23

a31 a32 a33

a22 a23
Minor of an element a11=M11 = a a33
32

a21 a23
Minor of an element a12=M12 = a a33 etc.
31

16. Cofactor of an element:Cofactor of an element aijof matrix A is denoted by c and is

defined as –
 a11 a12 a13 
e.g. A =  a a22 a23 
 21 
 a31 a32 a33 
a22
1+1 a23
Cofactor of a11=(– 1) M11 = a a33
32

1+2 a21 a23

Cofactor of a12=(– 1) M12 = – a a33 etc.
31

17. Adjoint of a matrix:Adjoint of a square matrix is denoted by and is defined as the

transpose of the matrix formed by the cofactors of the elements of the given matrix .

e.g. [ ] [ ]

C.F. of C.F. of C.F. of

C.F. of C.F. of C.F. of
C.F. of C.F. of C.F. of

Adj [ ].

18. Inverse of a matrix:If is a square matrix and | | i.e. is a non-singular matrix then

| |
Adj .

1.3Rank of a matrix:
Definition (Rank of Matrix): The rank of matrix is defined as the highest order of its
nonvanishing minor and it is denoted as .

In other words, a matrix is said to be of rank if,

i. At least one minor of order is non-vanishing and
ii. Every minors of order vanishes.

Note: 1. If the order of matrix is , then

2. If there exist a non-vanishing minor oforder , then
3. If all minors of order are zero, then
4. Elementary transformation does not alter rank of matrix.
5. Rank of null matrix is zero i.e.
6.If is non singular matrix of order then
7.If is singular matrix of order n then

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 3

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
8. Rank of the identity (unit) matrix is equal to order of matrix.

1.4Rank of a matrix using Echelon form:

Definition (Echelon or Canonical form of a matrix) :Let be a matrix of order Then is said to be
in Echelon form if :

i. The first row of has the first non zero element equal to one (Leading one)
ii. Leading one of every row lies to the right of leading one of the previous row
iii. Below every leading one all elements are zero
iv. Rows of zero element is/are at the bottom of matrix
7 –9
1
 5 2

e.g.: A =
0 0 1 –2 5
0 0 0 0 1
0 0 0 0 0

Note :1. Every matrix can be reduce to Echelon form

2.To solve the system of linear equations we can use canonical form ofmatrix. It is
not necessary to obtain first nonzero element in the first r rows as ‘1.’
3. Rank of a matrix is the number of non zero rows in its echelon form.
OR
Rank of matrix = Total number of rows – Number of rows containing all zeros

Example 1.Find the rank of the following matrix by reducing it to Echelon form

–1 –1
 21 3
–1 –2 –4
A= 3 1 3 –2
6 3 0 –7

–1 –1
 21 3
–1 –2 –4
Solution.: Given matrix is, A= 3 1 3 –2
6 3 0 –7
Applying R12

 12 – 13 –– 21 –4
–1
 3 1 3 –2
6 3 0 –7
Applying R2 – 2R1 ; R3 – 3R1 ; R4 – 6R1
–1 –2 –4
 10 5 3 7
 0 4 9 10 
0 9 12 17 
Applying R2– R3

 10 – 11 –– 26 –4
–3
  0 4 9 10 
0 9 12 17 
Applying R3 – 4R2 ; R4 – 9R2

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 4

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
–1 –2 –4
 1
0 1 –6 –3
  0 0 33 22 
 0 0 66 44 
Applying R4 – 2R3
 10 – 11 –– 26 –4
–3
  0 0 33 22 
0 0 0 0
R3
Applying 33

 10 – 11 –– 26 –4
–3
  0 0 1 2/3 
0 0 0 0
Which is Echelon form of Matrix A. Therefore rank of matrix A, (A)=Total number of rows –
Number of rows containing all zeros =4 – 1=3.

–1 –3
 14 1
2
0 2
Example 2 : Find the rank of matrix  0 3 1 4  by Echelon form
0 1 0 2
–1 –3
 14 1
2
0 2 
Solution. : Given matrix is, A= 0 3 1 4 
0 1 0 2 
Applying R2 – 4R1
–1 –3
 10 5
2
–8 14 
 0 3 1 4
0 1 0 2
Applying R24
–1 –3
 10 1
2
0 2
 0 3 1 4
0 5 –8 14 
Applying R3 – 3R2 ; R4 – 5R2
–1
 10 1
2
0 2
3
 0 0 1 –2
0 0 –8 4
Applying R4 + 8R3
–1
 10 1
2
0 2
3
 0 0 1 –2
0 0 0 – 12 
This is echelon form of matrix A. (A)=4 – 0 = 4

Exercise :

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 5

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
 2 3 5  4
  1 2 1 2 
3 4 5 6 
i  

Ans:  A  2 (ii)  2 5 3 0  Ans:  A  2
4 5 6 7 
   1 0 1 10 
 9 10 11 12 

 2 2 0 6 

4 2 0 2 
 iii  

Ans:  A  3 (iv)[ ]Ans :  A  3.
1 1 0 3 
 
 1 2 1 2 

 3 –1 2
(v) – 6 2 4  Ans. = 2 (vi)[ ]
 
– 3 1 2

1.5 System of Linear equations

Consider system of m equations in unknowns

. . .

This system can be written as in matrix form

Where, A = , X = ,B=

[ ] [ ] [ ]

Matrix ‘A’ is called coefficient matrix.

The matrix [A | B] is called augmented matrix and is given by,

[ ]

(I) System of Non-homogeneous equations:

If  thenthe system of equation is said to be non- homogeneous.

If the system has some solution (unique or infinitely many) then it is said to be consistent, otherwise
inconsistent.

Consider the system of equations,

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 6

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
If  | , the system is inconsistent i.e. system has no solution.

If | the system is consistent and

(i) If the system has [n – number of unknowns] unique solution.

(If and | | then this unique solution is .

(ii) If the system has infinitely many solutions and unknowns can be expressed
as linear combination ofremaining – unknowns.

(II) System of Homogeneous Equations :

If (i.e. ) thenthe system of equation is said to be homogeneous.

Since the | the system is always consistent i.e. the system has always a solution,
This solution is , i.e. , is called trivial solution.

If | , and

(i) If the system has unique solution which is trivial solution. OR

(ii) If the system have infinite many solutions and r unknowns can be expressed as a
linear combination of remaining – unknowns.

Note: If is a matrix then of the system then | | implies the system has
nontrivial solution and | | implies the system has trivial solution.

Example 1: Examine for consistency and solve, if consistence

– – .

Solution: Given system of equations is,

 12 1
–1 3
1
 x 
 y ; B = 1
3

This system can be written as, where = 4 1 5; X =   2
3 –2 1
 z 4
Consider
| 1 1 1 3
2 –1 3 1
4 1 5 2
3 –2 1 4

Operate ; ;

1 1 1 3
0 –3 1 –5

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 7

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
0 –3 1 – 10
0 –5 –2 –5

Applying

1 1 1 3 
0  5  2  5 
 
0  3 1  10
 
0  3 1 5 
Operate
1 1 1 3 
0 1  4 15 
 
0  3 1  10
 
0  3 1 5 

Operate
1 1 1 3
0 1  4 15 
 
0 0  11 35
 
0 0  11 40
Operate ;

1 1 3
1
0 1 15 
4

0 0  11 35
 
0 0 0 5

This shows that (A) =3 and (A|B)=4  (A) (A|B). Hence system is inconsistence

Example 2 :Examine for consistency the system of equations

– – – – and solve them if found consistence.

Solution: Given system of equations,

– –

– –
This system can be written as AX=B
1 –1 – 1 x 2

Where A= 1 2 1 ; X =  y  ; B =  2 

4 –7 – 5  z 2

Consider, [A|B]
1 –1 –1 2

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 8

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
1 2 1 2
4 –7 –5 2
Operate, ; ; ;
1 –1 –1 2
 0 3 2 0
0 –3 –1 –6
Operate, ;

1 –1 –1 2
 0 3 2 0
0 0 1 –6

This shows that, | Hence system is consistent .

Also, | . Hence system has a unique solution.
From ,
From , and
From , – –
Solution is –

Example 3 : Examine for the following system of equations for consistency and if consistent then
solve it : – – –

Solution: Given system of equations,

–
–
–

It can be written in matrix form as, AX=B,

2 –3 5  1
Where,A= 3 1 – 1 ; =[ ]; 2
  
1 4 –6  1

2 –3 5 1
Consider, [A|B] = 3 1 –1 2
1 4 –6 1
Operate,
1 4 –6 1
3 1 –1 2
2 –3 5 1
Operate, ; ; ;

1 4 -6 1
0 -11 17 -1
0 -11 17 -1
Operate, ;

1 4 -6 1

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 9

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
0 -11 17 -1
0 0 0 11

This shows that, (A) = (A|B) = 2 System is consistent

since therefore, the system have infinite number of solutions.
Now, from and ,- – and –
Put where
– – ( )  Solution is,
where

Exercise

Q.Examine the following systems for the consistency and solve if consistent.

(i) 2x + y – z + 3w = 8 ; x + y + z – w = – 2 ; 3x + 2y – z = 6 ; 4y + 3z + 2w = – 8
(ii) 2x – y – z = 2 ;x + 2y + z = 2 ; 4x – 7y – 5z = 2.
(iii)
(Ans.: )

Non-Homogeneous System of Equations with unknown constants:

Example 1 : Investigate the values of  and  so that the equations

–  

have (i) No Solution (ii) Unique Solution (iii) An infinite number of solutions.

Solution: Given system of the equations,

–
 
2 3 5  x  9
We can write this in the form, AX=B, where A= 7 3 –2 ; X= y ;B= 8



 

 
2 3   z  

2 3 5 9
[A|B] = 7 3 –2 8
2 3  
Operate ,
7 3 –2 –8
 2 3 5 9
2 3  
Operate ;
1 –6 – 17 – 19
 2 3 5 9
2 3  
Operate ; ;

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 10

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
1 –6 – 17 – 19
 0 15 39 47
0 15  + 34  + 38
Operate R3 – R2
1 – 6 – 17 – 19
 0 15 39 47 …(1)
0 0  – 5  – 9 …(1)
(i) If  | then system has no solution
From (1), this is possible only if  = 5 and  9.

(ii) If | then system has unique solution.

This is possible only if  5 and is any arbitrary real number.

(iii) If | then system has infinite number of solution.

This is possible only if  = 5 and = 9

Example 2:Show that the system 3x + 4y + 5z =  ; 4x + 5y + 6z = , 5x + 6y + 7z =  is the

consistence only when , ,  in arithmetic progression.

Solution :Given system of equations,




3 4 5 x   
It can be written in matrix form as,AX=Bwhere, A=4 5 6;
 X=y  ; B=  
5 6 7 z   
Consider, 3 4 5 
[A|B] = 4 5 6 
5 6 7 
Operate
4 5 6 
~ 3 4 5 
5 6 7 
Operate
1 1 1 –
~ 3 4 5 
5 6 7 

Operate ;
1 1 1 –
~ 0 1 2 4 – 3
0 1 2 5 – 5 + 

Operate
1 1 1 –
~ 0 1 2 4 – 3

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 11

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
0 0 0  – 2 + 

The system is consistent only if (A)= (A|B), i.e. only if  – 2 +  = 0

+
which gives  = . This shows that , ,  are in arithmetic progression.
2

Example 3 : For what values of k , the equations

have infinite number of solutions
? Hence find the solutions.

Solution: Given system of equations,

1 1 1  x   1 
It can be written in matrix form as,AX =B, where,A=2 1 4 ;X=

y  ; B =  k 
   
4 1 10  z  k^2

1 1 1 1
[A|B] = 2 1 4
4 1 10
Operate ;
1 –1 1 1
~ 0 –1 2 –2
0 –3 6 –4
Operate
1 1 1 1
~ 0 –1 2 k–2
0 0 0

The system is consistent only if (A) = (A|B).

From above matrix, This is possible only if
– –

Exercise:

(i) Determine the values of and for which the system

Has (i) no solution (ii) unique solution (iii) infinite number of solutions. Find the
solutions in case (ii) and (iii).
Ans.: (i) no solution for
(ii)Unique solution for any value,
(iii) infinitely many solutions for .

(ii) Determine the values of and for which the system

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 12

LINEAR ALGEBRA F.Y. B. TECH.(COMP)

Has (i) no solution (ii) unique solution (iii) infinite number of solutions
Ans.: (i) no solution for
(ii)Unique solution for any value,
(iii) infinitely many solutions for .

(iii) Investigate the values of and. so that the equations 2x – y + 3z = 2 ; x + y + 2z = 2 ;

5x – y + z =  have (i) no solution (ii) unique solution (iii) infinite many solutions.

1.6 System of Linear Homogeneous Equations

Consider the homogeneous system of equations, Where

Since | , the system is always consistent i.e. the system has always a solution,

This solution is , and it is called as the trivial solution.

If | , and

(i) If the system has unique solution which is trivial solution.

(ii) If the system have infinite many solutions

Example 1: Examine the following set of equations for non-trivial solution and solve them :

Solution : Given system of equations,

 5 2 –3   x 0
It can be written as, AX =0,Where A= 3 1 1  ;
 X=  y  ; 0=  0 
 2 1 6   z 0

Consider,

1 2 –3 0

[A | 0] = 3 1 1 0
2 1 6 0

Operate R1 – 2R3
1 0 – 15 0
 3 1 1 0
2 1 6 0
Operate R2 – 3R1 ; R3 – 2R1
1 0 – 15 0
 0 1 46 0
0 1 36 0

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 13

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
1 0 – 15 0
Operate R3 – R2  0 1 46 0
0 0 – 10 0
This show that, (A) = 3 = number of unknowns i.e
Thus the system has only trivial solution,

Homogeneous System of Equations Containing Constants :

Example 1:For different values of k, discuss the following equations : x + 2y – z = 0 ;

3x + (k + 7)y – 3z = 0 ; 2x + 4y + (k – 3)z = 0.

Solution: Given equations, x + 2y – z =0

3x + (k + 7)y – 3z=0
2x + 4y + (k – 3)z=0
It can be written as,
1 2 –1  x 0
AX=0, where, A= 3 k + 7 – 3  ; X= y  ;
   Z= 0 
2 4 k – 3 z 0

Consider,
1 2 –1 0
[A | Z] = 3 k+7 –3 0
2 4 k–3 0
Operate, and
1 2 –1 0
 0 k+1 0 0 …(1)
0 0 k–1 0

The system has infinite number of solutions if (A) < 3

From the last matrix it is possible only if

k+1=0 or k – 1 = 0 i.e. k = – 1, or k = 1

Exercice:

(i) Determine such that the system of homogeneous equations:

has (i) trivial solution (ii) non-trivial solution. Find the non-trivial
solution. Ans.: (i) for , (ii) for ,
(ii) Solve the following:

. Ans.: .

(iii) Solve the following:

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 14

LINEAR ALGEBRA F.Y. B. TECH.(COMP)

2
3
-4 . Ans.: Trivial solution

1.7 Numerical solutions of Linear Equations by Gauss Elimination Method

(III) Gauss elimination method :

Consider, a1x + b1y + c1z = d1 …(1)
a2x + b2y + c2z = d2 …(2)
a3x + b3y + c3z = d3 …(3)
Eliminate x from (2) and (3) [assume a1  0]
  a2  a3 
Equation (3.2.2) –   Equaiton (3.2.1) and Equation (3.2.3) –   Equation (3.2.1)
  1
a  1
a 
Here Equation (1) is called pivotal Equation and a1 is called first pivot,
 a1x + b1y + c1z = d1 …(4)
b y + c  z = d 
2 2 2
…(5)
b3 y + c3 z = d3 …(6)
Eliminate y from Equation (6) [assume b2  0]
 b3  
Equation (6) –   Equation (5)
 b2  
Here Equation (3.2.5) is called pivotal equation and b2 is a new pivot
 a1x + b1y + c1z = d1
b2 y + c2 z = d2
c3 z = d3
The values of x, y, z can find by these equation.
Partial Pivoting :
In the first step arrange equations such that first equation with largest coefficient of x then
second equation with largest coefficient of y (except Ist choosen Equation), third with largest
coefficient of z (except first two choosen Equations).
In the second step :
This modified procedure is known as partial pivoting.
Examples based on Gauss Elimination Method :
Ex. : Use Gauss elimination method to solve the following system of equations.
10 x1 – 7 x2 + 3 x3 + 5 x4 = 6
– 6 x1 + 8 x2 – x3 – 4 x4 = 5
3 x1 + x2 + 4 x3 + 11 x4 = 2
5 x1 – 9 x2 – 2 x3 + 4 x4 = 7
Soln. : Given system of equations,
10 x1 – 7 x2 + 3 x3 + 5 x4 = 6 …(1)
– 6 x1 + 8 x2 – x3 – 4 x4 = 5 …(2)
3 x1 + x2 + 4 x3 + 11 x4 = 2 …(3)

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 15

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
5 x1 – 9 x2 – 2 x3 + 4 x4 = 7 …(4)
Step I : Divide Equation (1) by 10
 x1 – 0.7 x2 + 0.3 x3 + 0.5 x4 = 0.6 …(5)
Now, eliminate x1 from Equations (2), (3) and (4) by using Equation (5)
i.e. [Equations (2) + 6 Equation (5)]; [Equation (3) – 3 Equation (5)] and
[Equation (4) – 5 Equation (5)]
 3.8 x2 + 0.8 x3 – x4 = 8.6 …(6)
3.1 x2 + 3.1 x3 + 9.5 x4 = 0.2 …(7)
– 5.5 x2 – 3.5 x3 + 1.5 x4 = 4 …(8)
Step II : Divide Equation (6) by 3.8
x2 + 0.2105 x3 – 0.2632 x4 = 2.2632 …(9)
Now, eliminate x2 from Equations (7), (8) by using Equation (9).
i.e. [Equation (7) – (3.1) Equation (9)] ; [Equation (8) + (5.5) Equation (9)]
2.4475 x3 + 10.3159 x4 = – 6.8159 …(10)
– 2.3423 x3 + 0.0524 x4 = 16.4476 …(11)
Step III : Divide Equation (10) by 2.4475
 x3 + 4.2149 x4 = – 2.7848 …(12)
Now, eliminate x3 from Equation (11) by using Equation (12)
i.e. Equation (11) + (2.3423) Equation (12)
 9.9250 x4 = 9.9248 …(13)
Step IV : Now find the values of x1, x2, x3, x4 using Equations (5), (9), (12) and (13) by
back substitutions.
It gives, x4 = 1
x3 = – 2.7848 – 4.2149 x4 x3 = – 6.9997 ; x3  – 7
x2 = 2.2632 – 0.2105 x3 + 0.2632 x4 x2 = 3.999 ; x2  4
x1 = 0.6 + 0.7 x2 – 0.3 x3 – 0.5 x4 x1 = 5
 Solution is, x1 = 5, x2 = 4, x3 = – 7, x4 = 1
Ex. : Solve the following system of equations by Gauss elimination method
10 x + 2 y + z = 9 ; 2 x + 20 y – 2 z = – 44 ; – 2 x + 3 y + 10 z = 22
Soln. : Given system of equations,
10 x + 2 y + z = 9 …(1)
2 x + 20 y – 2 z = – 44 …(2)
– 2 x + 3 y + 10 z = 22 …(3)
Step I : Divide Equation (1) by 10
1 1 9
x + 5 y + 10 z = 10 …(4)
Now eliminate x from Equations (2) and (3) using Equation (4)
i.e. [Equation (2) – 2 Equation (4)]; [Equation (3) + 2 Equation (4)]
98 11 – 229
y – z = …(5)
5 5 5
17 51 119
y + z = …(6)
5 5 5
98
Step II : Divide Equation (5) by 5
11 – 229
y – 98 z = 98 …(7)

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 16

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
17
Now eliminate y from Equation (6) using Equation (7) [i.e. Equation (6) – 5 Equation (7)]
1037 3111
 98 z = 98 …(8)
Step III : From Equations (4), (7) and (8) by back substitution, it gives,
z = 3
229 11
y = – 98 + 98 (z) y=–2
9 1 1
x = 10 – 5 y – 10 z x=1

 Solution is, x = 1 , y = – 2, z = 3
Ex. : Use Gauss elimination method with partial pivoting to solve the following system of linear
equations
8 x2 + 2 x3 = – 7 ; 3 x1 + 5 x2 + 2 x3 = 8 ; 6 x1 + 2 x2 + 8 x3 = 26
Soln. : Given system of equations,
8 x2 + 2 x3 = – 7 …(1)
3 x1 + 5 x2 + 2 x3 = 8 …(2)
6 x1 + 2 x2 + 8 x3 = 26 …(3)
Step I : Since in Equation (1), x1 term is absent
 by partial pivoting method, interchange Equation (1) with Equation (3), since in Equation
(3) coefficient of x1 numerically large.
 6 x1 + 2 x2 + 8 x3 = 26 …(4)
3 x1 + 5 x2 + 2 x3 = 8 …(5)
8 x2 + 2 x3 = – 7 …(6)
Step II : Divide Equation (4) by 6
1 4 13
 x1 + x2 + x3 = …(7)
3 3 3
Eliminate x1 from Equations (5) by using Equation (7), i.e. Equation (5) –
4 x2 – 2 x 3 = – 5 …(8)
8 x2 – 2 x 3 = – 7 …(9)
Since coefficient of x2 is large in Equation (9)
 Rearrange Equations (8) and (9)
8 x2 + 2 x3 = – 7 …(10)
4 x2 – 2 x 3 = – 5 …(11)
Step III : Divide Equation (10) by 8
1 –7
 x2 + 4 x3 = 8 …(12)
Now eliminate x2 from Equation (11) i.e. [Equation (11) – 4 Equation (12)]
–3
– 3 x3 = 2 …(13)
Step IV : From Equations (7), (12) and (13) by back substitution, it gives,
1
x3 = 2
–7 1
x2 = 8 – 4 x3 x2 = – 1
13 1 4
x1 = 3 – 3 x2 – 3 x3 x1 = 4

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 17

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
1
 Solution is, x1 = 4 , x2 = – 1, x3 = 2.

Ex. : Use Gauss elimination method to solve the following system of equations.
x1 + 4 x2 – x3 = – 5 ; x1 + x2 – 6 x3 = – 12 ; 3 x1 – x2 – x3 = 4
Soln. : Given system of equations,
x1 + 4 x2 – x3 = – 5 …(1)
x1 + x2 – 6 x3 = – 12 …(2)
3 x1 – x2 – x3 = 4 …(3)
Step I : Eliminate x1 from Equations (2) and (3) using Equation (1)
i.e. [Equation (2) – Equation (1)]; [Equation (3) – 3 Equation (1)]
 – 3 x2 – 5 x3 = – 7 …(4)
– 13 x2 + 2 x3 = 19 …(5)
Step II : Divide Equations (4) by – 3
5 7
 x2 + 3 x3 = 3 …(6)
Now eliminate x2 from Equation (5) i.e. [Equation (5) + 13 Equation (6)]
71 148
x 3 = …(7)
3 3
Step III : Now find values of x1, x2, x3 from Equations (1), (6) and (7) by back substitution.
148
It gives,  x3 = 71
7 5 – 81
x2 = 3 – 3 x3 x2 = 71
117
x1 = – 5 – 4 x2 + x3 x1 = 71

117 – 81 148
 Solution is,x1 = 71 , x2 = 71 , x3 = 71

Ex. : Solve the following system of equations by Gauss elimination method.

2 x1 + x2 + x3 = 10 ; 3 x1 + 2 x2 + 3 x3 = 18 ; x1 + 4 x2 + 9 x3 = 16
Soln. : Given system of equations,
2 x1 + x2 + x3 = 10 …(1)
3 x1 + 2 x2 + 3 x3 = 18 …(2)
x1 + 4 x2 + 9 x3 = 16 …(3)
Step I : Divide Equation (1) by 2
1 1
x1 + 2 x2 + 2 x3 = 5 …(4)
Now eliminate x1 from Equations (2), (3) using Equation (4)
i.e. [Equation (2) – 3 Equation (4)], [Equation (3) – Equation (4)]
1 3
x + …(5)
2 2 2 x3 = 3
7 17
x + …(6)
2 2 2 x3 = 11
1
Step II : Divide Equation (5) by  2 
 
x2 + 3 x3 = 6 …(7)
Now, eliminate x2 from Equation (6) using Equation (7)
7
i.e. [Equation (6) – 2 Equation (7)]

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 18

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
– 2 x3 = – 10 …(8)
Step III : Now find the values of x1, x2, x3 from Equations (4), (7) and (8) by back substitutions.
It gives, x3 = 5
x2 = 6 – 3 x3 x2 = – 9
1 1
x1 = 5 – 2 x2 – 2 x3 x1 = 7

 Solution is,x1 = 7, x2 = – 9, x3 = 5

Ex. : Solve the following systems of equations by Gauss elimination method.

4 x1 + x2 + x3 = 4 ; x1 + 4 x2 – 2 x3 = 4 ; 3 x1 + 2 x2 – 4 x3 = 6
Soln. : Given system of equations,
4 x1 + x2 + x3 = 4 …(1)
x1 + 4 x2 – 2 x3 = 4 …(2)
3 x1 + 2 x2 – 4 x3 = 6 …(3)
Step I : Divide Equation (1) by 4
1 1
 x1 + 4 x2 + 4 x3 = 1 …(4)
Now, eliminate x1 from Equations (2) and (3) using Equation (4)
 [Equation (2) – Equation (4)] ; [Equation (3) – 3 Equation (4)]
15 9
 4 x2 – 4 x3 = 3 …(5)
5 19
x – …(6)
4 2 4 x3 = 3
15
Step II : Divide Equation (5) by 4
3 4
 x2 – 5 x3 = 5 …(7)
5
None eliminate x2 from Equation (6) using Equation (7) i.e. [Equation (6) – 4 Equation (7)]
– 4 x3 = 2 …(8)
Step III : Find the values of x1, x2, x3 from Equations (4), (7) and (8) by back substitution.
–1
It gives, – 4 x3 = 2 x3 = 2
4 3 1
x2 = 5 + 5 x3 x2 = 2
1 1
x1 = 1 – 4 x2 – 4 x3 x1 = 1

1 –1
 Solution is, x1 = 1, x2 = 2, x3 = 2

Ex. : Solve the system of equations 3x + y – z = 3, 2x – 8y + z = – 5, x – 2y + 9z = 8 using Gauss

elimination method.
Soln. : Given system of equation we can write as matrix form,
AX = B
 3 1 – 1   x   3 
 2 – 8 1   y  = – 5 
 1 – 2 9  z  8 
 The augmented matrix is

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 19

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
–1
3 1
  3 
[A | B] = 2 –8 1  – 5
1 –2 9  8 
Now obtain A as upper triangular choosing ‘3’ as pivot,
2 1
i.e. operate R2 – 3 R1 R3 – 3 R1

3 1 –1 3
– 26 5
 0 –7
3 3
–7 28
0 7
3 8
 26 
Now choose – 3  as the pivot from the second column,
 
7
Operate R3 – 26 R2

3 1 –1 3
– 26 5
 0 –7
3 3
693 231
0 0
78 26
It gives, 3x + y – z = 3,
26 5 693 231
– 3 y+3z = –7 and z =
78 26
z = 1,

Now by back substitution,

26 5 5 26
– y = –7– z=–7– =– y=1
3 3 3 3
and
1 1
x = 3 [3 – y + z] = 3 [3 – 1 + 1] = 1

 Solution is x = 1, y = 1, z = 1
1.8 Method of Factorization or Triangularization :
Consider the following system of equations :
a11 x1 + a12 x2 + a13 x3 = b1
a21 x1 + a22 x2 + a23 x3 = b2
a31 x1 + a32 x2 + a33 x3 = b3
We can write in matrix form as AX = B …(7)
a11 a12 a13
  x1   b1 
Where A =  a21 a22 a23 , X =  x2  and B =  b2 
 a31
a32 a33   x3   b3 
In this method we use the fact that the square matrix A can be factorized into the form LU,
where L is a unit lower triangular matrix and U, an upper triangular matrix if all the minors of A are
non-singular.
Consider, A = LU …(8)

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 20

LINEAR ALGEBRA F.Y. B. TECH.(COMP)

1 0 0
  u11 u12 u13

Where L =  l21 1 0 and U= 0 u22 u23 
 l31 l32 1  0 0 u33 
Now find L and U from LU = A, i.e.

 1 0 0

u11 u12 u13
  a11 a12 a13

 l21 1 0  0 u22 u23  =  a21 a22 a23 
 l31 l32 1  0 0 u33   a31 a32 a33 

 u11 u12 u13

  a11 a12 a13

 l21 u11 l21 u12 + u22 l21 u13 + u23  =  a21 a22 a23 
 l31 u11 l31 u12 + l32 u22 l31 u13 + l32 u23 + u33   a31
a32 a33 
Equating the corresponding coefficients, we get
u11 = a11, u12 = a12, u13 = a13
a21 a21
l21 u11 = a21 or l21 = a ; l31 u11 = a31 or l31 = a
11 11
1
l21 u12 + u22 = a22 or u22 = a (a11 a22 – a21 a12)
11
1
l21 u13 + u23 = a23 or u23 = a (a11 a23 – a21 a13)
11
1
l31 u12 + l32 u22 = a32 or l23 = u (a32 – l31 u12)
22
and l31 u13 + l32 u23 + u33 = a33 or u33 = a33 – l31 u13 – l32 – l32 u23
It gives L and U.
Now Equation (7) becomes LUX = B …(9)
put UX = Z …(10)
Equation (3.2.9) gives LZ = B …(11)
 1 0 0
 
z1 b1
 
i.e.  l21 1 0   z2  =  b2 
 l31 l32 1   z3   b3 
It gives, z1 = b1,
l21 z1 + z2 = b2
and l31 z1 + l32 z2 + z3 = b3
Find z1, z2 and z3 by forward substitution. Now, from Equation (10) we have

 u11 u12 u13

  x1   z1 
 0 u22 u23   x2  =  z2 
 0 0 u33   x3   z3 
i.e. u11 x1 + u12 x2 + u13 x3 = z1 ; u22 x2 + u23 x3 = z2
and u33 x3 = z3
we can find by back substitution, x1, x2, x3.
Examples Based on Triangularisation
Ex.: Solve the following system by the method of triangularisation :
2x – 3y + 10z = 3, – x + 4y + 2z = 20, 5x + 2y + z = – 12
Soln. : The given system we can write as, AX = B, ...(1)

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 21

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
– 3 10
 2  x  3 
A = –1 4 2 , X = y and B =  20 
 5 2 1  z  – 12 
Step I :
Consider A = LU i.e. LU = A, …(2)

1 0 0
  u11 u12 u13

L =  l21 1 0  and U= 0 u22 u23 
 l31 l32 1  0 0 u33 

 u11 u12 u13

  2 – 3 10


 l21u11 l21u12 + u22 l21u13 + u23 = –1 4 2 
   5 
 l31u11 l31u12 + l32u22 l31u13 + l32u23 + u33  2 1
Step II : By equating,
u11 = 2, u12 = – 3, u13 = 10.
1
l21u11 = – 1,  l21 = – ,
2
5
l31 u11 = – 5,  l31 =
2
 1 5
l21u12 + u22 = 4,  u22 = 4 – l21u12 = 4 – – 2 (– 3) = 2
 
 1
l21u13 + u23 = 2,  u23 = 2 – l12u13 = 2 – – 2 (10) = 7
 
1 19
l31u12 + l32u22 = 5,  l32 = u [5 – l31u12] = 5
22
253
l31u13 + l32u23 + u33 = 1,  u33 = 1 – l13u13 – l32u23 = – 5

 
1 0 0 2 – 3 10

 1
 5  
 
–2 1 0
 L = and U = 0 2 7
   
 
5 19 253
1 0 0 – 5
2 5
Step III : From equations (1) and (2) we can write,
z1
 
put UX = Z, where Z =  z2 , …(3)
 z3 
then LZ = B, LUX = B

 1  z 
1 0 0

 –2 0  3 
1
1
i.e. z  =  20 
5  z 
2
 – 12 
2 19
5 1 3

1 5 19
z1 = 3 ; – 2 z1 + z2 = 20 ; 2 z1 + 5 z2 + z3 = – 12
43 506
 z1 = 3, z2 = 2 z3 = – 5
Step IV : Now from equation (3)

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 22

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
2 –3
  x
10
5  z1 
and  0 2 7
 y =  z2 
z  z3 
0 0 –
5 
253

Now Equation (1) implies

5 43 253 506
2x – 3y + 10z = 3 ; 2 y + 7z = 2 , – 5 z = – 3
By back substitution, x = – 4, y = 3 and z = 2.
Ex. : Solve the following system by the method of factorisation x + 3y + 8z = 4,
x + 4y + 3z = – 2 and x + 3y + 4z = 1.
Soln. : The given system we can write as,
AX = B, …(1)

1 3 8
 x 4
   
Where, A = 1 4 3 , X = y and B = – 2
1 3 4  z  1 
Step I :
Consider A = LU i.e. LU = A …(2)

1 0 0
  u11 u12 u13 
Where, L =  l21 1 0 and U =  0 u22 u23 
 l31 l32 1  0 0 u33 
 u11 u12 u13
 1 3 8

 l21u11 l21u12 + u22 l21u13 + u23 = 1 4 3 
  1 3 4
 l31u11 l31u12 + l32u22 l31u13 + l32u23 + u33 
Step II : By equating,
u11 = 1, u12 = 3, u13 = 8,
l21u11 = 1 l21 = 1 ; l31u11 = 1 l31 = 1
l21u12 + u22 = 4 u22 = 4 – l21u12 = 4 – 1 (3) = 1
l21u13 + u23 = 3 u23 =3 – l21u13 = 3 – 1(8) = – 5
1
l31u12 + l32u22 = 3 l32 = u [3 – l31u12] = 0
22
l31u13 + l32u23 + u33 = 4
 u33 = 4 – l31u13 – l32u23
= 4 – 1.8 – 0  (– 5) = – 4
 1 0 0
  1 3 8

 L = 1 1 0  and U =  0 1 –5 
1
0 1 0
0 4 
Step III : From equations (1) and (2), we can write,
z1
 
Put UX = Z, where Z =  z2  , LUX = B
 z3 
then LZ = B …(3)

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 23

LINEAR ALGEBRA F.Y. B. TECH.(COMP)

1   1 4
0 0 z
0 1 0   z2  = – 2
1 0 1   z3  1
z1 = 4, z2 = – 2
and z1 + z3= 1 z3 = 1 – z1 = 1 – 4 = – 3
Step IV : From equation (3)

1 3 8 x
 
z1
 
and 0 1 –5  y =  z2 
0 0 – 4  z  z3 
x + 3y + 8z = 4
y – 5z = – 2
– 4z = – 3
By back substitution,
3
z =
4
3 7
y = – 4 + 5z = – 2 + 5 4 = 4
 
7 3 29
and x = 4 – 3y – 8x = 4 – 3 4 – 8 4 = 4
   
29 7 3
 x = – 4 , y=4, z=4

1.9 Gauss Jordan Method :

This is modification of Gauss elimination method.
If given system of equations
AX = D

 a1 b1 c1
  x
  d1

Where, A =
 a2 b2 c2 , X =
 y and D=
 d2
     
 a3 b3 c3   z  d3
st
In this method eliminate x from last two equations using I equation.
st rd nd
Then eliminate y from I and III equations by using II equation.
rd st nd
And using III equation eliminate z from I and II equations [i.e. diagonally]
From these equations by back we can find unknowns x, y, z.
Also, we can use matrix method
Given Equations write in the matrix form
AX = D
Consider augmented matrix [A | D]
By row transformation convert matrix A into identity matrix I,
It gives, [I | D1]
By back substitution we can find unknowns x, y, z.
Examples Based on Gauss Jordan Method :
Ex.: Solved the following system of equations using Gauss-Jordan method :
x1 + 2x2 + x3 = 8 ; 2x1 + 3x2 + 4x3 = 20 ; 4x1 + 3x2 + 2x3 = 16

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 24

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
Soln. : Given system of equations
x1 + 2x2 + x3 = 8 …(1)
2x1 + 3x2 + 4x3 = 20 …(2)
4x1 + 3x2 + 2x3 = 16 …(3)
Step I : Eliminate x1 from Equations (2) and (3) using Equation (1) :
 [Equation (2) – (2) Equation (1)] ; [Equation (3) – 4 Equation (1)]
x1 + 2x2 + x3 = 8 …(4)
– x2 + 2x3 = 4 …(5)
– 5x2 – 2x3 = – 16 …(6)
Step II : Eliminate x2 from (4), (6) using Equation (5)
[Equation (4) + 2 Equation (5)] ; [Equation (6) – 5 Equation (5)]
x1 + 5x3 = 16 …(7)
– x2 + 2x3 = 4 …(8)
– 12x3 = – 36 …(9)
Step IV : Eliminate x3 from Equation (7), (8) using Equation (9)
5 1
i.e. [Equation (7) + 12 Equation (9)] ; [Equation (8) + 6 Equation (9)]
   
x1 = 1
– x2 = – 2
– 12x3 = – 36
 Solution is, x1 = 1 ; x2 = 2, x3 = 3

1.10 Numerical Solution of System of linear Equations by Cholesky’s Method

Cholesky’s Method :
Given system of equations write in the form
AX = D ---(12)

 a1 b1 c1
  x
  d1

Where, A =
 a2 b2 c2 , X =
 y and D =
 d2
     
 a3 b3 c3  z  d3
Step I :
T T
Consider, A = LL or A = UU
  l11 0 0
  u11 u12 u13
 
Where L – Lower triangular matrix
= L =
 l21 l22 0 U =  0 u22 u23 
    
 U – Upper triangular Matrix 
l31 l32 l33  0 0 u33 
T
Find L by A = LL
 a1 b1 c1
 l11 0 0
   l11 l21 l31

i.e.
 a2 b2 c2 =
 l21 l22 0  0 l22 l32
     

a3 b3 c3 
l31 l32  l33  0 0 l33 
Then Equation (12) becomes,
T
LL X = D ...( 13)
T
Put L X = Z ...( 14)

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 25

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
Equation (13) becomes,
LZ = D

l11 0 0
  z1
  d1

i.e.
 l21 l22 0  z2 =
 d2
     

l31 l32 l33   z3  d3
and find Z
Put in Equation (14)
T
LX = Z
 l11 l21 l31
 x
  z1

 0 l22 l32  y =
 z2
    
 0 0 l33   z  z3
and find X.
Examples Based on Cholesky’s Method :
Ex.: Solve the following system by Cholesky’s method
9x1 + 6x2 + 12x3 = 17.4 ; 6x1 + 13x2 + 11x3 = 23.6
12x1 + 11x2 + 26x3 = 30.8
Soln. : Given system of Equations,
9x1 + 6x2 + 12x3 = 17.4
6x1 + 13x2 + 11x3 = 23.6
12x1 + 11x2 + 26x3 = 30.8
We can write as in matrix form AX = B …(1)
 9 6 12
 x1
 
 17.4 
Where A =  6 13 11  , X = x2 , B =  23.6 
 12 11 26  x3  30.8 
Step I :
T
Consider, A = LL …(2)
Where L is Lower triangular matrix,
 l11 0 0
l11 l21 l31
  6 12 
9
 l21 l22 0   0 l22 l32  =  6 13 11 
 l31 l32 l33   0 0 l33   12 11 26 
2
 11l l11 l12 l11 l31  9 6 12

 l21 l11 l + l
2 2
l21l31 + l22 l32  =  6 13 11 
 21 22
2 
 31 11 31 21 32 22 31 32 33   12 11 26 
2 2
l l l l + l l l + l + l
Step II : Equating corresponding elements,
2
 l11 = 9  l11 = 3
l11 l12 = 6  l12 = 2
l11 l31 = 12  l31 = 4
l21 l11 = 6  l21 = 2
2 2 2 2 2
 l21 + l22 = 13 l22 = 13 – l21 l22 = 9  l22 = 3

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 26

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
11 – l21 l31
l21 l31 + l22 l32 =11 l32 =  l32 = 1
l22
l31 l11 = 12  l31 = 4
2 2 2 2 2 2
 l31 + l32 + l33 =26 l33 = 26 – l31 – l32
2
 l33 = 9  l33 = 3

 3 0 0
  3 2 4

L =  2 3 0  and LT =  0 3 1 
4 1  3 0 0  3
Step III : Since from Equations (1) and (2)
T
LL X = B …(3)
T
Put L X = Z …(4)
 Equation (3) as LZ = B
 3 0 0
 z1  17.4 
 2 3 0  = z2 =  23.6 
 4 1 3  z3  30.8 
17.43z1 =
2z1 + 3z2
23.6 =
4z1 + z2 + 3z3
30.8 =
5.8, z1 =
1
z2 = 3 [23.6 – 2z1]  z2 = 4
1
z3 = 3 [30.8 – 4z1 – z2]  z3 = 1.2
z1 = 5.8,  z2 = 4  z3 = 1.2
Step IV : Now, from Equation (4),
 3 2 4  x1  5.8 
 0 3 1  x2 =  4 
 0 0 3  x3  1.2 
3x1 + 2x2 + 4x3 = 5.8
3x2 + x3 = 4
3x3 = 1.2
By back substitution, 3x3 = 1.2 x3 = 0.4
1
3x3 + x3 = 4 x2 = 3 [4 – x3]  x2 = 1.2
1
3x1 + 2x2 + 4x3 = 5.8 x1 = 3 [5.8 – 2x2 – 4x3]  x1 = 0.6

 Solution is, x1 = 0.6, x2 = 1.2, x3 = 0.4

Ex.: Solve the following system by Cholesky’s method

4 x1 – 2 x2 = 0 ; – 2 x1 + 4 x2 – x3 = 1 ; – x2 + 4 x3 = 0
Soln. :
Given system, 4 x1 – 2 x 2 = 0
– 2 x1 + 4 x2 – x3 = 1
– x2 + 4 x3 = 0
We can write as in matrix form,AX = B …(1)
Where,

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 27

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
 4 –2 0  x1  0
A =  – 2 4 – 1 X =  x2 
 B =  1 
  
 0 –1 4  x3  0
Step I :
T T
Consider, A = LL i.e. LL = A …(2)
Where L is lower triangular matrix,
 l11 0 0   l11 l21 l31   4 –2 0
l l 0   
l22 l32 = – 2  4 –1
 21 22   0   
 l31 l32 l33   0 0 l33   0 –1 4
2
 l11 l11 l21 l11 l31   4 –2 0
l l 2 2
l21 + l22 l l + l l  
=–2 4 –1
 21 11 2  
21 31 22 32

 l31 l11 l31 l21 + l32 l22 l31 + l32 + l33   0 – 1 4

2 2

Step II : Equating corresponding elements,

2
l11 = 4  l11 =2
–2
l11 l21 = – 2 l21 = l  l21 =–1
11
l11 l31 = 0 l31 = 0  l31 =0
2 2 2 2 2
l21 + l22 = 4 l 22 = 4 – l 21 l 22 = 3  l22 = 1.7321
– 1 – l21 l31
l21 l31 + l22 l32 = –1 l32 =  l32 = – 0.5773
l22
2 2 2 2 2 2 2
l 31 + l 32 + l 33 = l 33 = 4 – l 31 – l 32
4 l 33 = 3.6667  l33 = 1.9149
 2 0 0  2 –1 0 
 L =   – 1 1.7321 0  and T
L =  0 1.7321 – 0.5773  …(3)
 
 0 – 0.5773 1.9149  0 0 1.9149 
Step III : From Equations (1) and (2) we can write,
T
LL X = B …(4)
T
Put L X = Z …(5)
 LZ = B …(6)
 2 0 0   z1   0 
 – 1 1.7321 0   z2  =  1 
     
 0 – 0.5773 1.9149   z3   0 
2 z1 = 0
– z1 + 1.7321 z2 = 1
– 0.5773 z2 + 1.9149 z3 = 0
By forward substitution,
z1 = 0  z1 = 0
1 + z1
z2 = 1.7321 z2 = 0.5773  z2 = 0.5773
0.5773 z2
z3 = 1.9149 z3 = 0.1740  z3 = 0.1740
Step IV : Put these values in Equation (5)
T
i.e. L X = Z
2 –1 0   x1   0 
  0 1.7321 – 0.5773   x2  =  0.5773 

0 0 1.9149   x3   0.1740 
2 x1 – x2 = 0
1.7321 x2 – 0.5773 x3 = 0.5773
1.9149 x3 = 0.1740

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 28

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
0.1740
By back substitution, x3 = 1.9149 x3 = 0.0909
0.5773 + 0.5773 x3
x2 = x2 = 0.3636
1.7321
x2
x1 = 2 x1 = 0.1818

Hence solution is, x1 = 0.1818, x2 = 0.3636, x3 = 0.0909

Ex.: Solve the following system by Cholesky’s method
4 x1 + 6 x2 + 8 x3 = 0
6 x1 + 34 x2 + 52 x3 = – 160
8 x1 + 52 x2 + 129 x3 = – 452
Soln. :
Given system, 4 x1 + 6 x2 + 8 x3 = 0
6 x1 + 34 x2 + 52 x3 = – 160
8 x1 + 52 x2 + 129 x3 = – 452
We can write as in matrix form,AX = B …(1)
4 6 8   x1   0 
Where,   
A =  6 34 52  , X =  x2  , B =  – 160  
 8 52 129   x3   – 452 
Step I :
T T
Consider, A = LL i.e. LL = A …(2)
Where L is lower triangular matrix,
 l11 0 0   l11 l21 l31  4 6 8 

  21 22
l l 0   l22 32 l   6 34 52 
  0  =  
 l31 l32 l33   0 0 l33   8 52 129 
2
 l11 l11 l21 l11 l31  4 6 8 
l l l
2
+ l
2
l l + l l  =  6 34 52 
 21 11 2  
21 22 21 31 22 32
 8 52 129 
2 2
 l31 l11 l31 l21 + l32 l22 l31 + l32 + l33 
Step II : Equating corresponding elements,
2
l11 = 4 l11 = 2  l11 = 2
6
l11 l21 = 6 l21 = l l21 = 3  l21 = 3
11
8
l11 l31 = 8 l31 = l l31 = 4  l31 = 4
11
2 2 2 2 2
l21 + l22 = 34 l22 = 34 – l21 l22 = 25  l22 = 5
52 – l21 l31
l21 l31 + l22 l32 = 52 l32 = l32 = 8  l32 = 8
l22
2 2 2 2 2 2 2
l31 + l32 + l33 = 129 l33 = 129 – l31 – l32 l33 = 49 l33 = 7
2 0 0 2 3 4
 L =   3 5 0  L =  0 5 8 
T
…(3)
 and
4 8 7 0 0 7
Step III : From the Equations (1) and (2) we can write,
T
LL X = B …(4)
T
Put L X = Z …(5)
 LZ = B …(6)

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 29

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
2 0  z1 
0  0 
  3 5  z  =  – 160 
0
 2
  
4 8 7
 z3   – 452 
2 z1 = 0
3 z1 + 5 z2 = – 160
4 z1 + 8 z2 + 7 z3 = – 452
 By forward substitution,
z1 = 0  z1 = 0
– 160 – 3 z1
z2 = z2 = – 32  z2 = – 32
5
– 452 – 4 z1 – 8 z2
z3 = z3 = – 28  z3 = – 28
7
Step IV : Substitute these values in Equation (5)
 2 3 4   x1   0 
  0 5 8   x   – 32 
  2 =  
 0 0 7   x3   – 28 
 2 x1 + 3 x2 + 4 x3 = 0
5 x2 + 8 x3 = – 32
7 x3 = – 28
– 28
 By back substitution, x3 = 7 x3 = – 4
– 32 – 8 x3
x2 = x2 = 0
5
– 3 x2 – 4 x3
x1 = x1 = 8
2
Hence solution is, x1 = 8, x2 = 0, x3 = – 4

Ex. : Solve the following system by Cholesky’s method

4 x1 + 2 x2 + 14 x3 = 14 ; 2 x1 + 17 x2 – 5 x3 = – 101
14 x1 – 5 x2 + 83 x3 = 155
Soln. :
Given system, 4 x1 + 2 x2 + 14 x3 = 14
2 x1 + 17 x2 – 5 x3 = – 101
14 x1 – 5 x2 + 83 x3 = 155
We can write as in matrix form,AX = B …(1)
 4 2 14   x1   14 
Where, A =  2 17 – 5 
 X =  x2  B =  – 101 
 
 14 – 5 83   x3   155 
Step I :
T T
Consider, A = LL i.e. LL = A …(2)
Where L is lower triangular matrix,
 l11 0 0   l11 l21 l31   4 2 14 
  
  l21 22 0   0 l22 l32  =  2
l   17 – 5 

 l31 l32 l33   0 0 l33   14 – 5 83 
2
 l11 l11 l21 l11 l31   4 2 14 
l l 2
l21 + l22
2
l l + l l  
= 2 17 – 5 
 21 11 2  
21 31 22 32
 14 – 5 83 
2 2
 l31 l11 l31 l21 + l32 l22 l31 + l32 + l33 

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 30

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
Step II : Equating corresponding elements,
2
l11 = 4  l11 = 2
2
l11 l21 = 2 l21 = l l21 = 1  l21 = 1
11
14
l11 l31 = 14 l31 = l l31 = 7  l31 = 7
11
2 2 2 2 2
l21 + l22 = 17 l22 = 17 – l21 l22 = 16  l22 = 4
1
l21 l31 + l22 l32 = –5 l32 = l [ – 5 – l21 l31 ] l32 = – 3
22
2 2 2 2 2 2 2
l31 + l32 + l33 = 83 l33 = 83 – l31 – l32 l33 = 25 l33 = 5
2 0 0 2 1 7
 L =  1 4 0 0 4 –3
T
 and L =   …(3)
7 –3 5 0 0 5
Step III : From the Equations (1) and (2) we can write,
T
LL X = B …(4)
T
Put L X = Z …(5)
 LZ = B …(6)
2 0 0   z1   14 
 1 4 0   z   – 101 
  2 =  
 7 – 3 5   z3   155 
 2 z1 = 14
z1 + 4 z2 = – 101
7 z1 – 3 z2 + 5 z3 = 155
 By forward substitution,
z1 = 7  z1 = 7
1
z2 = [ – 101 – Z1 ] = – 27  z2 = – 27
4
1
z3 = 5 [ 155 – 7Z1 + 3Z2 ] = 5  z3 = 5
Step IV : Put these values in Equation (5)
T
i.e. L X = Z
2 1 7   x1   7 
  0 4 – 3   x   – 27 
  2 =  
0 0 5   x3   5 
 2 x1 + x2 + 7 x3 = 7
4 x2 – 3 x3 = – 27
5 x3 = 5
 By back substitution,
5 x3 = 5  x3 = 1
1
x2 = 4 [ – 27 + 3 x3 ]  x2 = – 6
1
x1 = 2 [ 7 – x2 – 7 x3 ]  x1 = 3

Hence solution is, x1 = 3, x2 = – 6, x3 = 1

Ex. : Solve the following system by Cholesky’s method

2 x1 – x2 = 1 ; – x1 + 3 x2 + x3 = 0 ; x2 + 2 x3 = 0

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 31

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
Soln. :
Given system, 2 x1 – x2 = 1
– x1 + 3 x2 + x3 = 0
x2 + 2 x3 = 0
We can write as in matrix form,AX = B …(1)
 2 –1 0  x1  1
Where, A = –1 3 1  X = x  B = 0
  2  
 0 1 2  x3  0
Step I :
T T
Consider, A = LL i.e. LL = A …(2)
Where L is lower triangular matrix,
 l11 0 0   l11 l21 l31   2 –1 0
  
  l21 l22 0   0 l22 l32  =  – 1  3 1

 l31 l32 l33   0 0 l33   0 1 2
2
 l11 l11 l21 l11 l31   2 –1 0
l l 2 2
l21 + l22 l l + l l  = –1 3 1
 21 11 2   
21 31 22 32
 0 1 2
2 2
 l31 l11 l31 l21 + l32 l22 l31 + l32 + l33 
Step II : Equating corresponding elements,
2
 l11 = 2  l11 = 1.4142
–1 –1
l11 l21 = – 1 l21 = l l21 = 1.4142  l21 = – 0.7071
11

l11 l31 = 0 l31 = 0  l31 = 0

2 2 2 2 2
l21 + l22 = 3 l22=3– l21 l22
= 2.5  l22 = 1.5811
1
l21 l31 + l22 l32 = 1 l32 = l [ 1 – l21 l31 ] l32 = 0.6325  l32 = 0.6325
22
2 2 2 2 2 2 2
l31 + l32 + l33 = 2 l33 = 2 – l31 – l32 l33 = 1.5999 l33 = 1.2649
 1.4142 0 0 
 L =   – 0.7071 1.5811 0 

 0 – 0.6325 1.2649 
 1.4142 – 0.7071 0 
and
T
L =   0 1.5811 0.6325  …(3)

 0 0 1.2649 
Step III : From the Equations (1) and (2) we can write,
T
LL X = B …(4)
T
Put L X = Z …(5)
 LZ = B …(6)
 1.4142 0 0   z1   1 

  – 0.7071 1.5811 0   z2  =  0 
    
 0 – 0.6325 1.2649   z3   0 
 1.4142 z1 = 1
– 0.7071 z1 + 1.5811 z2 = 0
0.6325 z2 + 1.2649 z3 = 0
 By forward substitution,
1.4142 z1 = 1  z1 = 0.7071
0.7071
– 0.7071 z1 + 1.5811 z2 = 0 z2 = 1.5811 z1  z2 = 0.3162

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 32

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
0.6325
0.6325 z2 + 1.2649 z3 = 0 z3 = – 1.2649 z2  z3 = – 0.1581

Step IV : Put these values in Equation (5)

 1.4142 – 0.7071 0   x1   0.7071 
  0 1.5811 0.6325  x  =  0.3162 
  2
 0 0 1.2649   x3   – 0.1581 
1.4142 x1 – 0.7071 x2 = 0.7071
1.5811 x2 + 0.6325 x3 = 0.3162
1.2649 x3 = – 0.1581
 By back substitution,
– 0.1581
1.2649 x3 = – 0.1581 x3 = 1.2649  x3 = – 0.1250

0.3162 – 0.6325 x3
1.5811 x2 + 0.6325 x3 = 0.3162 x2 =  x2 = 0.25
1.5811
0.7071 + 0.7071 x2
1.4142 x1 – 0.7071 x2 = 0.7071 x1 =  x1 = 0.6250
1.4142
Hence solution is, x1 = 0.6250, x2 = 0.25, x3 = – 0.1250

Ex.: Solve the following system by Cholesky’s method

4 x1 + 2 x2 + 4 x3 = 10 ; 2 x1 + 2 x2 + 3 x3 + 2 x4 = 18
4 x1 + 3 x2 + 6 x3 + 3 x4 = 30 ; 2 x2 + 3 x3 + 9 x4 = 61
Soln. :
Given system, 4 x1 + 2 x2 + 4 x3 = 10
2 x1 + 2 x2 + 3 x3 + 2 x4 = 18
4 x1 + 3 x2 + 6 x3 + 3 x4 = 30
2 x2 + 3 x3 + 9 x4 = 61
We can write as in matrix form,AX = B …(1)

 42 22 43 02   xx1   18 
10
Where, A = 4 3 6 3, X = x
 2 B =  30 
0 2 3 9  3  61 
 x4 
Step I :
T T
Consider, A = LL i.e. LL = A …(2)
Where L is lower triangular matrix,
 l l22 0 0   l011 ll21 ll31 ll41   42 22 43 02 
l11 0 0 0

 l
 21  22 32 42 
=4 3 6 3
 31 l l 0   0 0 l l 43 
 l41 l42 l43 l44   0 0 0 l44   0 2 3 9 
32 33 33

2
 l11 l11 l21 l11 l31 l11 l41

 l21 l11 l21 + l22 
2 2
l l
21 31 + l l
22 32 l 21 41 + l22 l42
l
 l31 l11 l31 l21 + l32 l22 33 43 
2 2 2
l31 + l32 + l33 l l
31 41 + l l
32 42 + l l
 l41 l11 l41 l21 + l42 l22 l41 l31 + l42 l32 + l43 l33 l412 + l422 + l432 + l442 
 42 22 4
3 2
0
=4 3 6 3
0 2 3 9

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 33

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
Step II : Equating corresponding elements,
2
 l11 = 4 l11 = 2  l11 = 2
2
l11 l21 = 2 l21 = l l21 = 1  l21 = 1
11
4
l11 l31 = 4 l31 = l l31 = 2  l31 = 2
11

l11 l41 = 0 l41 = 0  l41 = 0

2 2 2 2 2
l21 + l22 = 2 l22 =2– l21 l22
=1  l22 = 1
3 – l21 l31
l21 l31 + l22 l32 = 3 l32 = l32 = 1  l32 = 1
l22
2 – l21 l41
l21 l41 + l22 l42 = 2 l42 = l42 = 2  l42 = 2
l22
2 2 2 2 2 2 2
l31 + l32 + l33 = 6 l33 = 6 – l31 – l32 l33 = 1  l33 = 1
3 – l31 l41 – l32 l42
l31 l41 + l32 l42 + l33 l43 = 3 l43 = l43 = 1  l43 = 1
l33
2 2 2 2 2 2 2 2 2
l41 + l42 + l43 + l44 = 9 l44 = 9 – l41 – l42 – l43 l44 = 4 l44 =2

 21 0
1 0 0
0 0
 20 11 2
1 2
0
 L = 2 1 1 0 and L = 0 0
T
1 1
0 2 1 2 0 0 0 2
…(3)
Step III : From Equations (1) and (2) we can write,
T
LL X = B …(4)
T
Put L X = Z …(5)
 LZ = B …(6)

 21 01 00 00   z12 
z
 10
18 
 2 1 1 0 z =  30 
 0 2 1 2   z4   3
 61 
2 z1 = 10
z1 + z2 = 18
2 z1 + z2 + z3 = 30
2 z2 + z3 + 2 z4 = 61
 By forward substitution,
10
z1 = 2 z1 = 5  z1 = 5
z2 = 18 – z1 z2 = 13  z2 = 13
z3 = 30 – 2 z1 – z2 z3 = 7  z3 = 7
61 – 2 z1 – z3
z4 = z4 = 14  z4 = 14
2
Step IV : Substitute these values in Equation (5)

 20 11 21 02   x12   135 
x

 0 0 1 1 x =  7 
 3
 0 0 0 2   x4   14 
2 x1 + x2 + 2 x3 = 5
x2 + x3 + 2 x4 = 13

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 34

LINEAR ALGEBRA F.Y. B. TECH.(COMP)
x3 + x4 = 7
2 x4 = 14
 By back substitution,
14
x4 = 2 x4 = 7
x3 = 7 – x4 x3 = 0
x2 = 13 – x3 – 2 x4 x2 = – 1
5 – x2 – 2 x3
x1 = x1 = 3
2
Hence solution is, x1 = 3, x2 = – 1, x3 = 0, x4 = 7

Exercise

Solve the following equations by Gauss elimination method

1. 3x + 4y – z = 8, – 2x + y + z = 3, x + 2y – z = 2 Ans. : x =
1, y = 2, z = 3
2. x – y + z = 1, – 3x + 2y– 3z = – 6, 2x – 5y + 4z = 5 Ans. : x = – 2, y = 3, z = 6
3. 10x + y + z = 12, 2x + 10y + z = 13, 2x+ 2y + 10z = 14 Ans. : x = 1, y = 1, z = 1
4. 2x – y + 2z = 2, x + 10y – 3z = 5, x– y – z = 3 Ans. : x =
2, y = 0, z = – 1
5. 10x1 + x2 + x3 = 18.141, x1 + x2 + 10x3 = 38.139, x1 + 10x2 + x3 = 28.140
Ans. : x1 = 1.234, x2 = 2.348, x3 = 3.455
Solve the following equations by Gauss – Jordan method
6. 2x– 3y + z = – 1, x+ 4y + 5z = 25, 3x – 4y + z = 2Ans. : x = 8.7, y = 5.7, z = – 1.3
7. 2x + y + z = 12, 3x + 2y + 3z = 24, x + 4y + 9z = 34 Ans. : x = 7, y = – 9, z = 7
8. 10x + y + z = 12, x + 10y + z = 12, x + y + 10z = 12 Ans. : x = y = z = 1
Solve the following equations by Factorization (or Triangularization) method.
9. 3x + y + 2z = 16 ; 2x – 6y + 8z = 24; 5x + 4y – 3z = 2 Ans. : 1, 3, 5
10. 3x + 2y + 7z = 32 ; 2x + 3y + z = 40 ; 3x + 4y – z = 56 Ans. : 7, 9, – 1
11. 10x + y + z = 12 ; 2x + 10y + z = 13; x + y + 5z = 7 Ans. : 1, 1, 1

JSPM’s RAJARSHI SHAHU COLLEGE OF ENGINEERING, PUNE 35