Algebraic solution of

simultaneous linear
equations
Objectives

At the end of this section you should be able to:

• Solve a system of two simultaneous linear equations in two unknowns using elimination.
• Detect when a system of equations does not have a solution.
• Detect when a system of equations has infinitely many solutions.
• Solve a system of three simultaneous linear equations in three unknowns using elimination.

In Section 1.1 a graphical method was described for the solution of simultaneous linear equa- tions. Both
lines are sketched on the same piece of graph paper and the coordinates of the point of intersection are
then simply read off from the diagram. Unfortunately this approach has several drawbacks. It is not
always easy to decide on a suitable scale for the axes. Even if the scale allows all four points (two from
each line) to fit on the diagram, there is no guarantee that the point of intersection itself also lies on it.
You may have encountered this difficulty when solv- ing Practice Problem 14(d) in the previous section.
When this happens you have no alternative but to throw away your graph paper and to start again,
choosing a smaller scale in the hope that the solution will now fit. The second drawback concerns the
accuracy of the graphical solution. All of the problems in Section 1.1 were deliberately chosen so that the
answers had nice numbers in them; whole numbers such as −1, 2 and 5 or at worst simple fractions such
as /2, 2 /2 and − /4. In practice, the coefficients of the equations may well involve decimals and we might
1 1 1

expect a decimal solution. Indeed, even if the coefficients are whole numbers the solution itself
Linear Equations 36

could involve nasty fractions such as 7/8 or perhaps something like 231/571. A moment’s thought should
convince you that in these circumstances it is virtually impossible to obtain the solution graphically, even
if we use a really large scale and our sharpest HB pencil in the pro- cess. The final drawback concerns the
nature of the problem itself. Quite frequently in eco- nomics we need to solve three equations in three
unknowns or maybe four equations in four unknowns. Unfortunately, the graphical method of solution
does not extend to these cases.
In this section an alternative method of solution is described which relies on algebra. It is called the
elimination method, since each stage of the process eliminates one (or more) of the unknowns. This
method always produces the exact solution and can be applied to systems of equations larger than just two
equations in two unknowns. In order to illustrate the method, we return to the simple example considered
in the previous section:
4x + 3y = 11 (1)
2x + y = 5 (2)
The coefficient of x in equation (1) is 4 and the coefficient of x in equation (2) is 2. If these numbers had
turned out to be exactly the same then we could have eliminated the variable x by subtracting one
equation from the other. However, we can arrange for this to be the case by multiplying the left-hand side
of the second equation by 2. Of course, we must also remember to multiply the right-hand side of the
second equation by 2 in order for this operation to be valid. The second equation then becomes
4x + 2y = 10 (3)
We may now subtract equation (3) from (1) to get
y=1
You may like to think of this in terms of the usual layout for the subtraction of two ordinary numbers: that
is,
4x + 3y = 11 4x + 2y = 10 −
y=1

This number can now be substituted into one of the original equations to deduce x. From equation (1)
4x + 3(1) = 11 (substitute y = 1)
4x + 3 = 11
4x = 8 (subtract 3 from both sides)
x = 2 (divide both sides by 4)
Hence the solution is x = 2, y = 1. As a check, substitution of these values into the other ori- ginal
equation (2) gives
2(2) + 1 = 5 /
The method of elimination can be summarized as follows.

Step 1
Add/subtract a multiple of one equation to/from a multiple of the other to eliminate x.
1.2 • Algebraic solution of simultaneous linear equations 37
Advice

The following example involves fractions. If you have forgotten how to perform calcula- tions with
fractions, you might like to read through Section 1.4.3 (pp. 76–81) now, since this section contains a
review of their basic properties.

Example

Solve the system of equations

3x + 2y = 1 (1)
−2x + y = 2 (2)

Solution
Step 1

The coefficients of x in equations (1) and (2) are 3 and −2 respectively. We can arrange for these to be the
same size (but of opposite sign) by multiplying equation (1) by 2 and multiplying (2) by 3. The new equa-
tions will then have x coefficients of 6 and −6, so we can eliminate x this time by adding the equations
together. The details are as follows.

Doubling the first equation produces

6x + 4y = 2 (3)
Tripling the second equation produces
−6x + 3y = 6 (4)
If equation (4) is added to equation (3) then
6x + 4y = 2 −6x + 3y = 6 +
7y = 8 (5)

Step 2

Equation (5) can be solved by dividing both sides by 7 to get

y = 8/7
Step 2
Solve the resulting equation for y.

Step 3
Substitute the value of y into one of the original equations to deduce x.
Step 4
Check that no mistakes have been made by substituting both x and y into the other original equation.
Linear 38
Equations Step 3
If 8/7 is substituted for y in equation (1) then
3x + 2 AC
87DF= 1 3x + 167= 1
3x = 1 − 167
(subtract 16/7 from both sides)
3x = 7 − 716
(put over a common denominator)
3x = − 97x = 13× AC − 97
DF(divide both sides by 3)
x=−
37
The solution is therefore x = −3/7, y = 8/7.
Step 4
As a check, equation (2) gives
−2 − 37
DF+ 87
= 67
+ 87
= 6 + 7 AC
8
= 147 = 2 / Advice
In the general description of the method, we suggested that the variable x is eliminated in step 1.
There is nothing special about x. We could equally well eliminate y at this stage and then solve the
resulting equation in step 2 for x.
You might like to solve the above example using this alternative strategy. You need to double
equation (2) and then subtract from (1).
Practice Problem
1 (a) Solve the equations
3x − 2y = 4 x − 2y = 2
by eliminating one of the variables.
(b) Solve the equations
3x + 5y = 19 −5x + 2y = −11
by eliminating one of the variables.
Example
Solve the system of equations
x − 2y = 1 2x − 4y = −3
Solution
Step 1
The variable x can be eliminated by doubling the first equation and subtracting the second:
2x − 4y = 2 2x − 4y = −3 −
0=5
The statement ‘0 = 5’ is clearly nonsense and something has gone seriously wrong. To understand what is
going on here, let us try and solve this problem graphically.
The line x − 2y = 1 passes through the points (0, −1/2) and (1, 0) (check this). The line 2x − 4y = −3
passes through the points (0, 3/4) and (−3/2, 0) (check this). Figure 1.11 shows that these lines are parallel
and so they do not intersect. It is therefore not surprising that we were unable to find a solution using
algebra, because this system of equations does not have one. We could have deduced this before when
subtracting the equations. The equation that only involves y in step 2 can be written as
0y = 5
and the problem is to find a value of y for which this equation is true. No such value exists, since
zero
× any number = zero and so the original system of equations does not have a solution.
Figure 1.11
1.2 • Algebraic solution of simultaneous linear equations 39
The following examples provide further practice in using the method and illustrate some special cases
which may occur.
Linear Equations 40

Example

Solve the equations

2x − 4y = 1
5x − 10y = 5/2

Solution

Step 1

The variable x can be eliminated by multiplying the first equation by 5, multiplying the second equation
by 2 and subtracting

10x − 20y = 5 10x − 20y = 5 −

0=0

Again, it is easy to explain this using graphs. The line 2x − 4y = 1 passes through (0, −1/4) and (1/2, 0).
The line 5x − 10y = 5/2 passes through (0, −1/4) and (1/2, 0). Consequently, both equations represent the
same line. From Figure 1.12 the lines intersect along the whole of their length and any point on this line is
a solution. This particular system of equations has infinitely many solutions. This can also be deduced
algebraically. The equation involving y in step 2 is

0y = 0

which is true for any value of y.

Figure 1.12
1.2 • Algebraic solution of simultaneous linear equations 41
These examples show that a system of equations can possess a unique solution, no solution or infinitely
many solutions. Algebraically, this can be detected in step 2. If the equation result- ing from the
elimination of x looks like
any non-zero number
× y = any number then the equations have a unique solution, or if it looks like
zero
× y = any non-zero number then the equations have no solution, or if it looks like
zero
× y = zero then the equations have infinitely many solutions.
It is interesting to notice how the graphical approach ‘saved the day’ in the previous two examples. They
show how useful pictures are as an aid to understanding in mathematics.
Practice Problem
2 Attempt to solve the following systems of equations
(a) 3x − 6y = −2 (b) −5x + y = 4
−4x + 8y = −1 10x − 2y = −8
Comment on the nature of the solution in each case.
We now show how the algebraic method can be used to solve three equations in three unknowns. As you
might expect, the details are more complicated than for just two equations, but the principle is the same.
We begin with a simple example to illustrate the general method. Consider the system
x + 3y − z = 4 (1)
2x + y + 2z = 10 (2)
3x − y + z = 4 (3)
The objective is to find three numbers x, y and z which satisfy these equations simultaneously. Our
previous work suggests that we should begin by eliminating x from all but one of the equations.
The variable x can be eliminated from the second equation by multiplying equation (1) by 2 and
subtracting equation (2):
2x + 6y − 2z = 8 2x + y + 2z = 10 −
5y − 4z = −2 (4)
Similarly, we can eliminate x from the third equation by multiplying equation (1) by 3 and sub- tracting
equation (3):
3x + 9y − 3z = 12 3x − y + z = 4 −
10y − 4z = 8 (5)
Linear Equations 42

At this stage the first equation is unaltered but the second and third equations of the system have changed
to equations (4) and (5) respectively, so the current equations are
x + 3y − z = 4 (1) 5y − 4z = −2 (4) 10y − 4z = 8 (5)
Notice that the last two equations constitute a system of just two equations in two unknowns, y and z.
This, of course, is precisely the type of problem that we already know how to solve. Once y and z have
been calculated, the values can be substituted into equation (1) to deduce x.
We can eliminate y in the last equation by multiplying equation (4) by 2 and subtracting equation (5):
10y − 8z = −4 10y − 4z = 8 −
−4z = −12 (6)
Collecting together the current equations gives
x + 3y − z = 4 (1) 5y − 4z = −2 (4) −4z = −12 (6)
From the last equation,

z = −12 −4
= 3 (divide both sides by −4)

If this is substituted into equation (4) then

5y − 4(3) = −2 5y − 12 = −2
5y = 10 (add 12 to both sides)
y = 2 (divide both sides by 5)
Finally, substituting y = 2 and z = 3 into equation (1) produces
x + 3(2) − 3 = 4 x + 3 = 4
x = 1 (subtract 3 from both sides)
Hence the solution is x = 1, y = 2, z = 3.
As usual, it is possible to check the answer by putting these numbers back into the original equations (1),
(2) and (3)
1 + 3(2) − 3 = 4 / 2(1) + 2 + 2(3) = 10 / 3(1) − 2 + 3 = 4 /
The general strategy may be summarized as follows. Consider the system
?x + ?y + ?z = ? ?x + ?y + ?z = ? ?x + ?y + ?z = ?
where ? denotes some numerical coefficient.
1.2 • Algebraic solution of simultaneous linear equations 43

Step 1
Add/subtract multiples of the first equation to/from multiples of the second and third equa- tions to
eliminate x. This produces a new system of the form
?x + ?y + ?z = ? ?y + ?z = ? ?y + ?z = ?

Step 2
Add/subtract a multiple of the second equation to/from a multiple of the third to eliminate y. This
produces a new system of the form
?x + ?y + ?z = ? ?y + ?z = ? ?z = ?

Step 3
Solve the last equation for z. Substitute the value of z into the second equation to deduce y. Finally,
substitute the values of both y and z into the first equation to deduce x.

Step 4
Check that no mistakes have been made by substituting the values of x, y and z into the original
equations.

It is possible to adopt different strategies from that suggested above. For example, it may be more
convenient to eliminate z from the last equation in step 2 rather than y. However, it is important to notice
that we use the second equation to do this, not the first. Any attempt to use the first equation in step 2
would reintroduce the variable x into the equations, which is the last thing we want to do at this stage.

Example

Solve the equations

4x + y + 3z = 8 (1) −2x + 5y + z = 4 (2) 3x + 2y + 4z = 9 (3)

Solution
Step 1

To eliminate x from the second equation we multiply it by 2 and add to equation (1):
4x + y + 3z = 8 −4x + 10y + 2z = 8 +

11y + 5z = 16 (4)*
Linear Equations 44

To eliminate x from the third equation we multiply equation (1) by 3, multiply equation (3) by 4 and
subtract:
12x + 3y + 9z = 24 12x + 8y + 16z = 36 −
−5y − 7z = −12 (5)
This produces a new system:
4x + y + 3z = 8 (1)
11y + 5z = 16 (4)
−5y − 7z = −12 (5)

Step 2

To eliminate y from the new third equation (that is, equation (5)) we multiply equation (4) by 5, multiply
equation (5) by 11 and add:
55y + 25z = 80 −55y − 77z = −132 +
−52z = −52 (6)
This produces a new system
4x + y + 3z = 8 (1)
11y + 5z = 16 (4)
−52z = −52 (6)

Step 3

The last equation gives

z = −52 −52
= 1 (divide both sides by −52)

If this is substituted into equation (4) then

11y + 5(1) = 16
11y + 5 = 16
11y = 11 (subtract 5 from both sides)
y = 1 (divide both sides by 11)
Finally, substituting y = 1 and z = 1 into equation (1) produces
4x + 1 + 3(1) = 8
4x + 4 = 8
4x = 4 (subtract 4 from both sides)
x = 1 (divide both sides by 4)
Hence the solution is x = 1, y = 1, z = 1.

Step 4

As a check the original equations (1), (2) and (3) give

1.2 • Algebraic solution of simultaneous linear equations 45

4(1) + 1 + 3(1) = 8 /
−2(1) + 5(1) + 1 = 4 /
3(1) + 2(1) + 4(1) = 9 /
respectively.

Practice Problem

3 Solve the following system of equations:

2x + 2y − 5z = −5 (1)
x − y + z = 3 (2)
−3x + y + 2z = −2 (3)

As you might expect, it is possible for three simultaneous linear equations to have either no solution or
infinitely many solutions. An illustration of this is given in Practice Problem 8. The method described in
this section has an obvious extension to larger systems of equations. However, the calculations are
extremely tedious to perform by hand. Fortunately there are many computer packages available which are
capable of solving large systems accurately and efficiently (a matter of a few seconds to solve 10 000
equations in 10 000 unknowns).

Advice

We shall return to the solution of simultaneous linear equations in Chapter 7 when we describe how
matrix theory can be used to solve them. This does not depend on any sub- sequent chapters in this
book, so you might like to read through this material now. Two techniques are suggested. A method
based on inverse matrices is covered in Section 7.2 and an alternative using Cramer’s rule can be
found in Section 7.3. rms

Elimination method The method in which variables are removed from a system of simul- taneous
equations by adding (or subtracting) a multiple of one equation to (or from) a mul- tiple of another.

Linear 46
Equations Practice Problems

4 Use the method of elimination to solve the systems of equations given in Section 1.1, Problem 14.
5 Sketch the following lines on the same diagram:

2x − 3y = 6, 4x − 6y = 18, x − 32
y=3

Hence comment on the nature of the solutions of the following systems of equations:
(a) 2x − 3y = 6 (b) 4x − 6y = 18

x − 32
y = 3 x − 32
y=3
y=3

6 Use the elimination method to attempt to solve the following systems of equations. Comment on
the
nature of the solution in each case.
(a) −3x + 5y = 4 (b) 6x − 2y = 3
9x − 15y = −12 15x − 5y = 4
7 Solve the following systems of equations:
(a) x − 3y + 4z = 5 (1) (b) 3x + 2y − 2z = −5 (1)
2x + y + z = 3 (2) 4x + 3y + 3z = 17 (2)
4x + 3y + 5z = 1 (3) 2x − y + z = −1 (3)
8 Attempt to solve the following systems of equations. Comment on the nature of the solution in
each
case.
(a) x − 2y + z = −2 (1) (b) 2x + 3y − z = 13 (1)
x + y − 2z = 4 (2) x − 2y + 2z = −3 (2)
−2x + y + z = 12 (3) 3x + y + z = 10 (3)