Lecture 8

Solve the following system of equations

x  2 y  z  1
x  y  2z  3
4 x  y  5z  8
5 x  7 y  2 z  1

Augmented matrix of the system is

 1  2 1  1
   1  2 1  1
 
1 1  2 3  0 3  3 4 
4 1 5 8   
  , R2  R2  R1 , R3  R3  4 R1 , R4  R4  5R1
  
0 9 9 12

 5  7 2  1 0 3  3 4 
   

 1  2 1  1
 
 0 3  3 4  1
0 3  3 4  , R3  R3 , R4  R4  R2
 3
 
0 0 0 
 0

 1  2 1  1
 
0 3  3 4 
0 0  , R3  R3  R2
 0 0
 
0 0 0 
 0

HVS de Silva Dept. of Mathematics Page 1

  1  2 1  1
 4
 0 1  1  1
 3  , R2  R2
 0 0 0 0 3
0 0 0 0 


Therefore reduced system is

x  2 y  z  1
4
yz
3
Let z t where t is a parameter.

Then z t
1
x  (3t  5)
3
The solution of the system is
1
x  (3t  5)
3
1
y  (4  3t )
3
z t

Since t is a parameter the system has infinite

number of solutions.

HVS de Silva Dept. of Mathematics Page 2

2.11 Rank of a Matrix
Number of non zero rows(columns) of the
row(column) echelon form of a matrix is defined as
the rank of a matrix.
1 0 0
 
Note:  0 1 0 is in column echelon form.
 2 3 1
 

A matrix is in column echelon form if its transpose is

in row echelon form.

For a non- homogeneous system Anxn X nx1  Bnx1

If A  0 the system has a unique solution

( i.e. Rank of A = Rank of augmented matrix =n)

If A  0 the system has infinite no of solutions or no

solutions
(i.e. Rank of A  Rank of the augmented matrix then
the system has no solutions

Rank of A = Rank of augmented matrix < n

then the system has infinite number of solutions)

HVS de Silva Dept. of Mathematics Page 3

If the system has solutions we say it is consistent
otherwise inconsistent.

Examples

1. Show that the following system of equations has a

unique solution unless a  2.

x  y  2 z  3
2x  3 y  4z  5
5x  az  b
 1  1 2  3
 
 2 3  4 5 
Augmented matrix = 5 0
 a b 

1 1 2 3 
 
 0 5  8 11  , R  R  2 R , R  R  5R
 0 5 (a  10) (b  15)  2 2 1 3 3 1

 

1 1 2 3 
 
 0 5  8 11 , R3  R3  R2
 0 0 (a  2) (b  4) 
 

HVS de Silva Dept. of Mathematics Page 4

 1 1 2 3 
 8 11 
0 1  1
 5 5 , R2  R2
 5
 0 0 (a  2) (b  4) 

If a  2
 
 
1 1 2 3 
0 1  8 11 
 5 5 , R3 
1
R3
 b4 (a  2)
0 0 1 
 a2
If a  2 ,
Rank of the coefficient matrix = 3
Rank of the augmented matrix = 3
The system has a unique solution.
2. For what values of  ,  the simultaneous equations
x  2y  z  8
2 x  y  3z  13
3 x  4 y  z  
have (i) no solutions
(ii) a unique solution
(iii) infinitely many solutions

HVS de Silva Dept. of Mathematics Page 5

 1 2 1 
 
A  2 1 3
Let 3 4  
 

1 2 1
A2 1 3  3  11
3 4 

If A  0 i.e. if  3  11  0 , 
11
the system has a
3

unique solution for any value of  .

If A  0 i.e. if 
11
the system has no solutions or
3

infinite number of solutions.

 
1 2 1 8 
Augmented matrix =  2 1 3 13
 11 
3 4 
 3 

 
1 2 1 8 
0  3 1 3  , R2  R2  2 R1 , R3  R3  3R1
 2 
0  2   24 
 3 

1 2 1 8 
  2
0  3 1 3  , R3  R3 
3
R2
 0 0 0   22 
 

HVS de Silva Dept. of Mathematics Page 6

 1 2 1 8 
 1  1
0 1 1  , R2  R2
 3  3
0 0 0   22 

If   22

1 2 1 8
 1 
0 1 1
 3 
0 0 0 0 

Rank A = Rank (Aug. matrix)=2 < 3

The given system has infinite number of solutions.
If   22

1 2 1 8
 1 
Augmented matrix 0 1 1
 3 
0 0 0 1 

Rank A  Rank (Aug. matrix)

The given system has no solutions.
Thus we get that
11
(i) If   3 ,   22 then the system has no
solutions.

HVS de Silva Dept. of Mathematics Page 7

 11
(ii) If  then the system has a unique
3
solution for any value of  .
11
(iii) If   3 ,   22 then system has infinite
number of solutions.

2.12 Solution of homogeneous systems

Homogeneous system is of the form AX  O . Always
X  O is a solution of the system.
X  O is called the trivial solution. Therefore
homogeneous system is always consistent.

For a homogeneous system, where A is a square

matrix of order nxn,
If A 0 , the system has only the trivial solution.
If A  0 , the system has infinite solutions other than
the trivial solution.
In other words,
If Rank(A) = Rank(Aug. matrix)=n only trivial
solution exists.
HVS de Silva Dept. of Mathematics Page 8
If Rank(A) = Rank(Aug. matrix) <n solutions other
than the trivial solution exists.
Examples
Solve the following systems
1.
x yz 0
3x  y  z  0
2 x  2 y  3z  0

 1 1 1 0  1 1 1 0
   
 3  1  1 0    3  1  1 0  , R  R
 2 2  3 0   2 2  3 0 1 1
  

1 1 1 0
 
   , R  R  3R , R  R  2R
 0 2 4 0
 0 4  5 0 2 2 1 3 3 1
 

 1 1 1 0
 
 0 2  4 0 , R3  R3  2R2
0 0 3 0 


 1 1 1 0
 
  , 1 1
 0
0 0
1 2 0 R2  R2 , R3  R3
 1 0  2 3

HVS de Silva Dept. of Mathematics Page 9

Rank of the coe. matrix=Rank of the aug.matix =3
= No. of variables

System has only the trivial solution, x  y  z  0 .

2. x1  x2  x3  3x4  0

x1  x2  2 x3  x4  0

4 x1  2 x2  4 x3  x4  0

 1 1 1 3 0
 
 1 1  2 1 0
 4  2 4 1 0
 
 1 1 1 3 0
 
  0 2  1  2 0 , R2  R2  R1 , R3  R3  4 R1
 0 2 8  11 0 
 

 1 1 1 3 0
 
  0 2  1  2 0 , R4  R4  R2
0 0 9  9 0
 

HVS de Silva Dept. of Mathematics Page 10

  1 1 1 3 0
 1  1 1
 0 1   1 0 , R2  R2 , R3  R3
 2  2 9
0 0 1 1 0

Rank(coe. Matrix)=Rank(Aug.matrix)=3< 4=no.of

variables
The system has infinite no. of solutions other than the
trivial solution.
The reduced system is,

x1  x2  x3  3x4  0

1
x2  x3  x4  0
2
x3  x4  0

Let x3  x4  t where t is a parameter.

1 3
x  x  x4  t
Then 2 2 3 2
t
x1  x2  x3  3 x4  
2
HVS de Silva Dept. of Mathematics Page 11
 The solution is given by,
t
x1  
2

3
x2  t
2
x3  x4  t

Note:
System of linear Equations ( Amxn X nx1  Bmx1 ), mn

Non Homogeneous( B  o ) Homogeneous( B  o )

consistent inconsistent consistent

(has solutions) ( no solutions) always trivial solution exists

Rank(A)=Rank(AIB)=n Rank(A) Rank(AIB) Rank(A)=Rank(AIB)=n

a unique solution no solutions only trivial solution

Rank(A)=Rank(AIB)<n Rank(A)=Rank(AIB)<n

infinite no. of solutions infinite no. of solutions other

than the trivial solution

HVS de Silva Dept. of Mathematics Page 12