..

8 infinite
(c) An Systems of Equations
number of solutions. .
am1 x1 + am2 x2 + · · · + amn xn = bm
Definition
where ai j 1.2: and System of Linear
b j are real numbers. Equations
The above is a system of m equations in the n variables,
A , x2 · · · , xof
x1system n . Written
linear more
equations simply
is a listinofterms of summation notation, the above can be written in
equations,
the form
1.2 Systems Of Equations, Algebraic Procedures a11n x1 + a12 x2 + · · · + a1n xn = b1
8 Systems of Equations ∑ ai j x j = bi , i = 1, 2, 3, · · · , m
21 x1 + a22 x2 + · · · + a2n xn = b2
aj=1
..
1.2.. Systems Of Equations, Algebraic Procedures 9
Definition 1.2: System of Lineara Equations
m1 x1 + am2 x2 + · · · + amn xn = bm
AThe relative
system size ofequations
of linear m and n is is not
a listimportant
of equations, here. Notice that we have allowed ai j and b j to be any
OutcomesIfnumber.
you think
realwhere ai j We and ofcaneach equation
are
b j also real as anumbers
call numbers.
these condition which is
Thescalars
above .must be satisfied
a system
We will m by
useofthis thethroughout
variables,
equations
term in theconsistent
text, sowould
n variables,
the keep
mean there
x1 , x2 that
in mind is some
· · · , xthe
n . termchoice
Written of
more
scalar variables
justsimply
meansa which
in
x terms
+
that we
a can
x satisfy
of
are+ summation
· ·
working· +all
a the
xwithconditions.
=notation,
real
b the
numbers.Inconsistent
above can would
be mean
written there
in
11 1 12 2 1n n 1
A.is no
Use
thechoice
Now,form of the variables whichtocan
elementary
suppose weoperations have a system afind
satisfy
21nx1 the
where +bai 22
=
all
x02 of
solution
+ the
for· ·all ·+ conditions.
to aInxlinear
i.a2n system
n = b2words
other of equations.
every equation equals 0. This is
The following sections provide methods for .
determining
. if a system is consistent or inconsistent, and
a special
B.finding
type of system.
Findsolutions
the row-echelon form andj=1
if they exist.
∑reduced
ai j x j = bi , .i = 1, 2, 3, · · · , m
row-echelon form of a matrix.
a x +a x +···+a x = b
m1 1 m2 2 mn n m
Definition 1.3:
C.1.2.1.
Determine Homogeneous
whether a system System
of linear of Equations
equations has no solution,
where Elementary
ai j and Operations
b j are real numbers. The above is a system of m equationsa in
unique
the n solution
variables, or an
A
The system
infinite number
relative of equations
size of
of m is
solutions
and called
n is from
nothomogeneous
its
importantrow-echelon
here. if each equation
form.
Notice that
x1 , x2 · · · , xn . Written more simply in terms of summation notation, the above can be written we in the
have system
allowed is
a i jequal
and bto 0.be
j to inAany
homogeneous
realthenumber.
form We system
can also has
callthe
theseformnumbers scalars . We will use this term throughout the text, so keep
n
D.We
in Solve
mind
beginthat athis
system
the termofwith
section equations
scalaranjust using
means
example. Gaussian
that
Recall wefrom Elimination
are working
Example with
1.1 that and
real Gauss-Jordan
solution to the Elimination.
numbers.
the given system was
(x, y)Now,
= (−1, suppose 4). we have a system where ∑a11axi1j x+j a=12bxi2, +
i =· ·1,· +2,a3, ·x·n· =
bi = 0 for all i. In1nother
, m0
words every equation equals 0. This is
j=1
E.a special
Modeltype a physical system with linear a21 x1 +equations
a22 x2 + · · ·and + a2nthen
xn =solve.
0
of system. ..
Example 1.5: Verifying an Ordered Pair is a Solution .
The relative size of m and n is nota x +
important a
m1 1 of Equationsx 2 · ·Notice
+
here. · + amnthat
xn =we 0 have allowed ai j and b j to be any
Definition 1.3:verify
Algebraically Homogeneous
that (x, y) =System
(−1, 4) is m2 a solution to the following system of equations.
real number. We can also call these numbers
We have taken an in depth look at graphical representations of systems scalars . We will use this term throughout
of equations, the text, so keep
as well as how to
Awhere
system
in mindsolutions a are
of
that the term
i j scalars
equationsandisx are
called
scalar just Our
i variables.
means homogeneous
that we are if
working each equation
with real in the
numbers. system is equal to 0 . A
find possible graphically. attention now
x+y = 3 turns to working with systems algebraically.
homogeneous system has the form
Now, suppose we have a system where bi = 0yfor − xall = i.5 In other words every equation equals 0. This is
a special
Recall typefrom of system.
the previous section athatx our + agoal + · · · +working
x when a x =with 0 systems of linear equations was to
11 1 12 2 1n n
find the point of intersection of the equations a21 x1 + awhen22 x2 + · · · + a2nInxnother
graphed. = 0 words, we looked for the solutions to
Solution. By graphing these two equations ofand identifying
.. the point of intersection, we previously found
theDefinition
system. We 1.3:
now Homogeneous
wish to find these System solutionsEquations
algebraically.
. We want to find values for x1 , · · · , xn which
that (x, y) = (−1, 4) is the unique solution.
solve all of the equations. If such
A system of equations is called homogeneous a set of values
am1 x1 + am2these exists,
if · each we call (x
· · + aequation , · ·in
· , xthe
n ) the solution
system set. to 0. A
is equal
2+ =
x1.2. Systems 1 0
xn Of Equations, Algebraic Procedures 9
We can verify algebraically by substituting values mn
into the original equations, and ensuring that
homogeneous
Recall the above system has the form
discussions about the types of solutions possible. We will see that systems of linear
the where
equations are hold. First,and
scalars we xsubstitute the values into the first equation and check that it equals 3.
equationsawill i j have one unique i are variables.
solution, infinitely many solutions, or no solution. Consider the following
definition.
If you think of each equation as a condition a 11 x1x +
+ aywhich
12 + ·must
=x2(−1) · ·++(4)
a1n
be= 3= 0
xnsatisfied by the variables, consistent would
a21 x1 + a22 x2 + · · · + a2n xn = 0
mean there is
ThisRecallsome
equalsfrom choice
3 as the of
needed, variables
previousso we which
see that
section that(−1, can
our 4) satisfy
goalis when all
a.. solution the conditions.
to the
working with firstsystems Inconsistent
equation. would
Substituting
of linear equations the mean to there
values
was
is no choice Definition
of second 1.4:
the variables Consistent
which and Inconsistent Systems
.
into
find the
the point ofequation
intersectionyieldsof can satisfy all
the equations
aym1−xx1 +
of the
when graphed. conditions.
a4mnIn other 05 words, we looked for the solutions to
A system of linear equations is called = a(4)
m2 x−
consistent2+
(−1) ···+
if =
there n1 =
+xexists
=want at least one solution.
the system. We now wish to find these solutions
The following sections provide methods for determining if a system algebraically. We to isfind values
consistent x1 It
for or · ·is· , called
,inconsistent,
xn which and
whichinconsistent
solve all
is of
true.
where aif theFor if there
equations.
(x, y) =is no
If
(−1, solution.
such
4) a set
each of values
equation exists,
is true we
and call (x
therefore, , · · · ,
thisx )isthe
a solution
solution toset.
the system. ♠
finding solutions i j are
they scalars
exist.and xi are variables. 1 n
Recall the above discussions about the types of solutions possible. We will see that systems of linear
Now, the interesting question is this: If you were not given these numbers to verify, how could you
equations will have one unique solution, infinitely many solutions, or no solution. Consider the following
algebraically
1.2.1. Elementary
Recall from determine
the the solution?
Operations
previous section that Linear
our goal algebra
whengives us the
working with tools needed
systems of to answer
linear this question.
equations was to
definition.
The following basic operations are important tools that we will utilize.
find the point of intersection of the equations when graphed. In other words, we looked for the solutions to
the Definition
system. We1.4: nowConsistent
wish to findand these solutions algebraically.
Inconsistent Systems We want to find values for x1 , · · · , xn which
We begin solveDefinition
this of the1.6:
allsection with Elementary
equations. If suchOperations
an example. a set of values
Recall from exists,
Example we call1.1 (x1that
, · · · ,the
xn ) the solution
solution to set.
the given system was
A system of linear equations is called consistent if there exists at least one solution. It is called
(x, y) = (−1, Elementary
Recall
4). the above operations
discussions are those
aboutoperations
the types of consisting
solutionsofpossible.the following. We will see that systems of linear
inconsistent if there is no solution.
equations will have one unique solution, infinitely many solutions, or no solution. Consider the following
1. Interchange the order in which the equations are listed.
definition.
Example 1.5: Verifying an Ordered Pair is a Solution
Definition 2. Multiply any equation by a nonzero number.
1.4: that
Consistent and Inconsistent Systemsto the following system of equations.
Algebraically verify (x, y) = (−1, 4) is a solution
3. Replace
A system any equation
of linear equationswith itself added
is called to a multiple
consistent of another
if there exists equation.
at least one solution. It is called
inconsistent if there is no solution. x+y = 3
y−x = 5
It is important to note that none of these operations will change the set of solutions of the system of
equations. In fact, elementary operations are the key tool we use in linear algebra to find solutions to
systems of equations.
Solution. By graphing these two equations and identifying the point of intersection, we previously found
that (x, y) = (−1, 4) is the unique solution.
We can verify algebraically by substituting these values into the original equations, and ensuring that
the equations hold. First, we substitute the values into the first equation and check that it equals 3.
x + y = (−1) + (4) = 3
This equals 3 as needed, so we see that (−1, 4) is a solution to the first equation. Substituting the values
into the second equation yields
 be a solution 22 = 2kb2 . Hence, (x1 , · · · , xn ) is also a solution
1 to1 kE 1 to
1 1.3.
2 2
the only difference between these two systems is that the second involves multiplying the equation,
ESimilarly, let (x1 ,k.· · Recall
2 = b2 by the scalar
· , xn ) that
be awhen
solutionyou1.2. E1 =both
of Systems
multiply b1 ,sides = an 2 . Then
kE2Equations,
Of of kbequation bywe
thecan
Algebraic multiply
same number, the equation
Procedures 13
kE =
the sides
2 kb by the scalar 1/k, which is possible only because we have required
are2 still equal to each other. Hence if (x1 , · · · , xn ) is a solution to E2 = b2 , then it will that
alsok ̸= 0. Just as
beabove, thistoaction
a solution kE2 =preserves
kb2 . Hence, equality
(x1 , · · · ,and we obtain
n ) isSystems
x1.2. also Of the
a solution equation E2 = bProcedures
to 1.3. Algebraic
Equations, 2 . Hence (x111
, · · · , xn ) is also
replace theSimilarly,
asecond equation
solution (xE, 1· ·=
letto
by (−2) times
E2 a=
· , xb1), be b2 . theoffirst
solution
equation added to the second. This yields the system
E = b , kE = kb . Then we can multiply the equation
1 n 1 1 2 2
kE2 = kb2 by the scalar 1/k, which is possible only because we have required that k ̸= 0. Just as
Theorem
3. above, 1.8: Elementary
Finally, Operations and
x +weSolutions
+1.1 =
thiswe willpreserves
action prove that the systems
equality and 6z and
3yobtain the251.4 haveE the
equation same solution set. We will show that
2 = b2 . Hence (x1 , · · · , xn ) is also
a any
Suppose solution
you to
solution E1 aof
have bE1 ,1E=
=system b1btwo
of
2=
= b2equations
,2E. 2linear + 2za =
is yalso solution
8 of 1.4. Then, we will show that any solution (1.6) of
1.4 is also a solution of E1 = b1 , E2y
2= +b5z
2. = 19(x1 , · · · , xn ) be a solution to E1 = b1 , E2 = b2 . Then
Let
3. Finally, we will itprove
in particular solves thatEthe=systems b1 . Hence, =itb1.4
1.1E1and 1solveshave the
the same solution set.
first equation in We 1.4.will show thatit also solves
Similarly,
1 (1.1)
any solution of = , = is also a E =
solutionb of 1.4. Then, we will show that any solution
Now, replace E2the
= bthird E
2 . Byequation
our1 proof b E
1 with b
(−2)
2of 1.3, 2 it times the
2 second
also solves 2
kE1 =added to the third.
kb1 . Notice that ifThis
we add yieldsE2 and kEof1 , this is equal
the system
1.4 is also a solution of E1 = b1 , E2 = b2 . Let (x1 , · · · , xn ) be a solution to E1 = b1 , E2 = b2 . Then
Then tothe + kb1it. Therefore,
b2following
in particular systems
solves E1 have
if (x
= b1the
· · · , xsolution
,same
. 1Hence, n )solves
it
solves setEas 1= b1 , E2 = b2 it must also solve E2 + kE1 = b2 + kb1 .
1.1:
x+ 3y + 6zthe=first 25 equation in 1.4. Similarly, it also solves
ENow
1.2
= b2suppose (x1 , ·of
. By our proof · · 1.3,
, xn )itsolves
also solves theykE
system
+12z = kbE
= 18.1Notice= b1 , that
E2 +if kE = bE22 +
we1add and 1 .1 ,Then
kbkE this isin particular
equal
(1.7) it is a
b2 + kb1 .of
tosolution Therefore, if (x ,
E1 = b1 . Again by our proof· · · , x ) solves E =
of b , E = b it must also solve E + kE = b
b2 it is also a solution to kE1 = kb1 . Now if we subtract + kb .
1 n
zE=
1 2= 31.3,
1 2 2 2 1 2
(1.2)
1
these
Now equal (x
suppose quantities
1 , · · · , x n ) from
solves both
the sides
system E E
of
1 1==
E 2b
b 1+,1EkE
2+ = 1b=
1 kE 2+ +1kbwe
b2kb obtain
1 . Then E2 = b2 , itwhich
in particular is a shows that
solution of E1 =also
the solution
At this point, b1 . Again
satisfies by E our1= proof of = bit2take
E21.3,
b1 ,Simply .is alsoz = a solution to kE1 = kbthis
1 . Now if weintosubtract
2. we can easily find the solution. 3 and substitute back the previous
these equal quantities from both
equation to solve for y, and similarly to solve for x.E1 = b1 sides of E 2 + kE 1 = b 2 + kb 1 we obtain E 2 = b 2 , which shows that
the solution also satisfies E1 = b1 , E2 = b2 . kE = kb (1.3) ♠
2 2

for simply,
any scalar k, above
provided xk + 0. +
̸= 3y 6 (3) = x + 3y + 18 = 25 ♠ solution set
Stated the theorem shows that the elementary operations do not change the
y + 2 (3) = y + 6 = 8
of aStated
system of equations.
3. simply, the above theorem shows that thez = elementary operations do not change the solution set
3
of a system
We will of now
equations.
look at an example of a system E1 = ofb1 three equations and three variables. Similarly to the
(1.4)
E2 + kE1 = b2 + kb1
The second
previous equation
We will is now
examples,
now look atthe
angoal is tooffind
example values
a system of for y, z such and
threex,equations thatthree
eachvariables.
of the given equations
Similarly to the are satisfied
previous for any
examples,scalar
the k (including
goal is
when these values are substituted in.to k
find = 0 ).
values fory+6 = 8
x, y, z such that each of the given equations are satisfied
when these values are substituted in.
You can see from this equation that y = 2. Therefore, we can substitute this value into the first equation as
Example
Before 1.9: Solving
we proceed with theaproof
System of Equations
of Theorem with
1.8, let us Elementary
consider Operations
this theorem in context of Example
follows:Example 1.9: Solving a System of Equations with Elementary Operations
1.7. Then,
Find the solutions to the system, Ex +=3x(2)+ y,+ 18
b == 25
7
Find the solutions to the system, 1 1
By simplifying this equation, we find that Ex2= = 1.
2x − y, b2 =
Hence, the8 solution to this system is (x, y, z) = (1, 2, 3).
x + 3y x++ =+
6z3y 256z = 25
This Recall
processtheiselementary operations
called back that we used to modify
substitution. 2x + 7y2x++ 7ythe
14z =+58
system
14z in the solution to the example. First,
= 58 (1.5) (1.5)
we added (−2) times the first equation to the second equation. In terms of Theorem 1.8, this action is
Alternatively, in 1.7 you could have continued 2y
2y + 5zas +
195z = 19
=follows. Add (−2) times the third equation to the
given by
second and then add (−6) times the second to the
E2 + (−2) E1first. This
= b2 + (−2)yields
b1
Solution.
Solution.We
or We cancan
relate thisthis
relate system to Theorem
system x1.8
to Theorem above.
+ 3y = 7In
1.8 this case,
above. we have
In this case, we have
2x − y + (−2) (x + y) = 8 + (−2) 7
E1 = x + 3yy+= 6z,2 b1 = 25
This gave us the second system in Example E
1.7,
1 = x
given+ 33y +b26z,
by
E2 = 2x + 7yz+=14z, = 58 1
b = 25
=+
E3E=2 2y 2x + 7y +b314z,= 19 b2 = 58
E1 5z,
= b1
Now add (−3) times the second to the Efirst. = 2y + 5z,
E3 yields
This b3 = 19
2 + (−2) E1 = b2 + (−2) b1
Theorem 1.8 claims that if we do elementary operations on this system, we will not change the solution
Theorem
set. From 1.8point,
Therefore,
this claims
we can that if
solve
we were weto
this
able do
findelementary
system using
the operations
the elementary
solution onTheorem
this system,
operations
to the system.x=1 given in we uswill
that not
Definition
1.8 tells change
the1.6. First, the solution
solution
set.found
we Therefore,
is in fact we can solve
a solution to thethis system
original using
y = the
system. 2 elementary operations given in Definition 1.6. First,
14 We will
Systems now prove Theorem 1.8.
of Equations z=3
Proof.
a system which has the same solution set as the original system. This avoided back substitution and led
1.2.2. Gaussian
to the 1. The
same proof
solution Elimination
that
set.the systems
It is 1.1 and 1.2
your decision haveyou
which the prefer
same solution
to use, set
as is as follows.
both methods Suppose that correct
lead to the
(x , · · · , x ) is
solution, (x, 1y, z) =n(1, 2, 3). a solution to E 1 = b ,
1 2E = b 2 . We want to show that this is a solution to the system ♠
in 1.2 above. This is clear, because the system in 1.2 is the original system, but listed in a different
The work weorder.
did inChanging the order
the previous does not
section willeffect the solution
always so (x1 , · ·to
find theset,solution · , xthe
n ) is a solution
system. Intothis
1.2.section, we
will explore a less cumbersome way to find the solutions. First, we will represent a linear system with
an augmented matrix. A matrix is simply a rectangular array of numbers. The size or dimension of a
matrix is defined as m × n where m is the number of rows and n is the number of columns. In order to
construct an augmented matrix from a linear system, we create a coefficient matrix from the coefficients
of the variables in the system, as well as a constant matrix from the constants. The coefficients from one
equation of the system create one row of the augmented matrix.
For example, consider the linear system in Example 1.9

x + 3y + 6z = 25
2x + 7y + 14z = 58
2y + 5z = 19
 For a linear system of the form
a11 x1 + · · · + a1n xn = b1
.. a11 x1 + · · · + a1n xn = b1
. ..
am1 x1 + · · · + amn xn = bm .
am1 x1 + · ·1.2.
· + aSystems
mn xn = bOf
m Equations, Algebraic Procedures 15
les and the ai j and bi are constants, the augmented matrix of this system is
where the xi are variables and the ai j and bi are constants, the augmented matrix of this system is
⎡ Definition
given by 1.10: Augmented
⎤ Matrix of a Linear System
⎡ ⎤
a11For ·a· ·linear
a1nsystem
b1 of the form a11 · · · a1n b1
⎢ .. .. .. ⎥ ⎢ .. .. .. ⎥
⎣ . . . ⎦ ⎣ . . . ⎦
a11 x1 + · · · + a1n xn = b1
am1 · · · amn bm am1 · · · . amn bm
..
am1 x1 + · · · + amn xn = bm
ary operations in the Now, consider
context elementary operations in the
Thecontext of the augmented matrix. The elementary opera-
where
tions the xiofarethe
in Definition 1.6
augmented
variables
can be used
matrix.
and the i j and
on athe are
rowsbijust
elementary
constants,
as we used the
opera-
themaugmented matrix
on equations of this system
previously. Changesis to
be used on the rowsgivenjust by
as we used them on equations previously. Changes to
a system of equations in as a result of an ⎡ elementary operation are ⎤ equivalent to changes in the augmented
a result of an elementary operation are equivalent
matrix resulting from the corresponding row to changes
a 11 · · ·in the
a 1n augmented
b 1
orresponding rowrow 14 Systems ofthat
operation. Equations ⎢ operation. Note that⎥Theorem 1.8 implies that any elementary
.. that any.. elementary
..
operationsNote
used on an Theorem
augmented1.8matrix
implies
⎣ will . not change . ⎦ solution to the corresponding system of
.the
augmented matrix will notWe
equations. change the solution
now formally define to the corresponding
elementary am1 row amn bsystem
· · · operations.m These ofare the key tool we will use to find
ly define elementary row
solutions tooperations.
systems of
1.2.2. Gaussian Elimination These are
equations. the key tool we will use to find
uations.
Now, consider
Definition 1.11:elementary
Elementary operations in the context of the augmented matrix. The elementary opera-
Row Operations
tions in Definition 1.6 can be used on the rows just as we used them on equations previously. Changes to
The
Thework
ntary Row Operations
a system
we didrow
elementary
of equations
in the previousansection
operations
in as a result of(also knownwill
elementary
always
as row find
operations)
operation
theconsist
solution
are equivalent
tofollowing
theinsystem.
of changes
to the In this section, we
the augmented
will explore
matrix resulting afrom
lessthe
cumbersome
corresponding way to find theNote
solutions. First,1.8weimplies
will represent a linear system with
perations (also known 1. row operations)
asSwitch two rows. consist ofrow theoperation.
following that Theorem that any elementary
an augmented matrix. A matrix is simply a rectangular array of numbers.
row operations used on an augmented matrix will not change the solution to the corresponding system The size or dimension
of of a
matrix
equations. isWe
defined as m
nowa formally
2. Multiply row by× n where
a define
nonzero m is therow
elementary
number. number of rows
operations. Theseand is key
are nthe thetool
number
we willofuse
columns.
to find In order to
solutions to systems of equations.
construct an augmented matrix from a linear system, we create a coefficient matrix from the coefficients
of the3. variables
y a nonzero number. Replace ain rowthebysystem,
any multiple of another
as well row added
as a constant to it. from the constants. The coefficients from one
matrix
Definition 1.11: Elementary Row Operations
equation of the system create one row of the augmented matrix.
any multiple of another row addedrow
to it.
Forelementary
The
Recall example, operations
consider
how we solved (also
the linear
Example known
1.9. system
We dorow
canasin theoperations)
Example 1.9 consist
exact same steps asofabove,
the following
except now in the
context of an augmented matrix and using row operations. The augmented matrix of this system is
1. Switch two rows. ⎡ x + 3y +⎤6z = 25
Example 1.9. We can do the exact same steps as above, 1 3except6 25now in the
2. MultiplyThe
a rowaugmented
by a nonzero number. 2x
⎣of2 this + 7y + 14z = 58
matrix and using row operations. matrix 14 58 ⎦
7 system is
0 2
2y
5
+
19
5z = 19
⎡ 3. Replace a row⎤ by any multiple of another row added to it.
1 3 6 25
⎣This
Thus system
2 the
7 first
caninbe written as an augmented matrix, as follows
58 ⎦ solving the system given by 1.5 would be to take (−2) times the first row of the
14 step
augmented matrix
weand add itExample
to the second row,can⎡do the exact same⎤steps as above, except now in the
0Recall
2 how5 19 solved 1.9. We
1 3 ⎤ 6 25
Systems of Equations
context of an augmented matrix and using row ⎡ operations. The augmented matrix of this system is
1 3⎣ 2 6 257 14 58 ⎦
ing the system given 16 by Systems be to take (−2) times
of Equations
1.5 would ⎡⎣ 0 the first8 ⎦⎤5 of19the
row
1 31 0 26 225
teit how
to thethis
second row, to 1.6. Next take (−2) times the
corresponds 0 72 14
⎣ 2second 5 row
19
58 and
⎦ add to the third,
⎡ Notice
Note it⎤has exactly
⎡ to 1.6. 0⎤ (−2)
2 5 times 19 the second row and add to the third,
1 how3 this corresponds
6 that
25 Next take
1 3the 6same 25 information as the original system.⎡Here⎤ it is understood that the
⎣ 0 the1 first ⎡ ⎤ 1
Thus 2 step8 ⎦in solving ⎣ the
0 system
1 2 given 8 ⎦ by11.53would 6 25 be to take (−2) times the first row of the
first column contains the coefficients from x in each equation, in order, ⎣ 2 ⎦ . Similarly, we create a
0 2 5matrix
augmented 19 and add it to0the0second
1 row,3 ⎣ 0 1 2 8 ⎦
⎡ 0
0 0 1⎡⎤ 3 ⎤
s augmented matrix corresponds to the system 1 3 6 25 3
⎣ 0 1 2 8 ⎣⎦ ⎦
This augmented
column from the matrix corresponds
coefficients to each
on y in the system
equation, 7 and a column from the coefficients on z in each
x + 3y + 6z = 250 2 5 19
2
⎡ ⎤ y + 2z = 8 x + 3y + 6z = 25
6
z=3 y + 2z = 8
equation, 14 ⎦ . For a system of more than three
⎣
z = 3variables, we would continue in this way constructing
ich is the same as 1.7. By back 5substitution you obtain the solution x = 1, y = 2, and z = 3.
a column
which
Through a systematic is thefor each
same
procedure variable.
as row
of 1.7. By backSimilarly,
operations, forsimplify
substitution
we can ayou
system anofaugmented
obtain less
the solution x = 1,variables,
than three
matrix = 2,carry
yand zit=simply
andwe 3. construct a
row-echelon form orcolumn
Through for
reduced each variable.
a systematic
row-echelon procedure of rowwe
form, which operations,
define next.we can simplify
These formsanare
⎡augmented matrix and carry it
used⎤to find
to row-echelon form or reduced row-echelon
solutions of the system of equations corresponding to the augmented matrix. form, which we define next. 25
These forms are used to find
Finally, we construct a column from the constants of the equations, ⎣ 58 ⎦ .
In the followingthedefinitions,
solutions ofthe theterm
system of equations
leading corresponding
entry refers to the firstto the augmented
nonzero entry matrix.
of a row when
In the following definitions, the term leading entry refers to the first 19
nonzero entry of a row when
nning the row from left to right.
The rows of the augmented
scanning the row from left to right. % matrix correspond & to the equations in the system. For example, the top
row in the augmented matrix, 1 3 6 | 25 corresponds to the equation
Definition 1.12: Row-Echelon Form
Definition 1.12: Row-Echelon Form
An augmented matrix is in row-echelon form if x + 3y + 6z = 25.
An augmented matrix is in row-echelon form if
the Through
solutions aofsystematic
the systemprocedure of row
of equations operations, to
corresponding wethe
can simplify an
augmented augmented matrix and carry it
matrix.
to row-echelon form or reduced row-echelon form, which we define next. These forms are used to find
In the following definitions, the term leading entry refers to the first nonzero entry of a row when
the solutions of the system of equations corresponding to the augmented matrix.
scanning the row from left to right.
In the following definitions, the term leading entry refers to the first nonzero entry of a row when
scanning the row
Definition from
1.12: left to right.Form
Row-Echelon
An augmented matrix is in row-echelon form if
Definition 1.12: Row-Echelon Form
An1.augmented matrix
All nonzero in row-echelon
rowsisare form
above any rows if
of zeros.

2.
1. Each leadingrows
All nonzero entryare
of above
a row is
anyin rows
a column to the right of the leading entries of any row above
of zeros.
it.
2. Each leading entry of a row is in a column to the right of the leading entries of any row above
it. leading entry of a row is equal to 1.
3. Each

3. Each leading entry of a row is equal to 1.

We also consider another reduced form of the augmented matrix which has one further condition.

We also consider
Definition another reduced
1.13: Reduced form ofForm
Row-Echelon the augmented matrix which has one further condition.
1.2. Systems Of Equations, Algebraic Procedures 17
An augmented matrix is in reduced row-echelon form if
Definition 1.13: Reduced Row-Echelon Form
An1.
form. augmented
However, it matrix
All nonzero rows
often in reduced
isare
happens abovethatany
therow-echelon
rows of zeros.
row-echelon form
formifis sufficient to provide information about the
solution of a system. 1.2.
2.
1. Each leadingrows
All nonzero entryare
of above
a row isany in rows
a column to Systems
of zeros. the right Of Equations,
of the leading Algebraic Procedures
entries of any rows above17
The following examples describe matrices in these various forms. As an exercise, take the time to
it.
carefully Each
verify that they areofin athe specified form. to the right of the leading entries of any rows above
form. 2.
However,leading
it oftenentry
happens row
that is
theinrow-echelon
a column form is sufficient to provide information about the
3. Each
it. leading
solution of a system. entry of a row is equal to 1 .
Example 1.14: Not in Row-Echelon Form
The3. following
4. All
Eachentries examples
leading in entry describe
a column
of a above
row ismatrices
and
equal toin1.these
below various
a leading forms.
entry As an exercise, take the time to
are zero.
The following
carefully verify thataugmented
they are inmatrices are not
the specified in row-echelon form (and therefore also not in reduced
form.
row-echelon form).
4. All entries in a column above and below a leading entry are zero.
Notice that the first three ⎡ conditions on ⎤a reduced row-echelon form matrix are the same as those for
Example 1.14: Not in Row-Echelon Form ⎡ ⎤
row-echelon form. 0 0 0 0 ⎡ ⎤
Notice that theaugmented ⎢ conditions
first three on⎥ a reduced row-echelon 0 form
2 3 matrix
3 are thenot
same as those for
The following ⎢ matrices
1 2 3 are
3 not in 1row-echelon
2
⎥ matrix is also in 3 form (and therefore
⎢ row-echelon ⎥form. also in reduced
Hence, every reduced row-echelon form The converse is not
row-echelon
row-echelon form.
form). ⎢ 0 1 0 2 ⎥ , ⎣ 2 4 −6 ⎦ , ⎢ 1 5 0 2 ⎥
necessarily true; we cannot⎢assume that every ⎥ matrix in row-echelon ⎣ 7 5 form 0 1is ⎦also in reduced row-echelon
Hence, every reduced ⎡ ⎣ 0 0 0 1form
row-echelon ⎤⎦ matrix 4 0 is also 7 in row-echelon form. The converse is not
00 00 00 00 ⎡ ⎡ 0 0 1 0 ⎤
necessarily true; we cannot⎢assume that every ⎤
⎥ matrix in row-echelon
0 2 form3 3 is also in reduced row-echelon
⎢ 1 2 3 3 ⎥ 1 2 3 ⎢ 1 5 0 2 ⎥
⎢ 0 1 0 2 ⎥ , ⎣ 2 4 −6 ⎦ , ⎢ ⎥
⎢ ⎥ ⎣ 7 5 0 1 ⎦
⎣ 0 0 0 1 ⎦ 4 0 7
0 0 1 0
Example 1.15: Matrices in 0 Row-Echelon
0 0 0 Form
The following augmented matrices are in row-echelon form, but not in reduced row-echelon form.
⎡ ⎤
⎡ ⎤ 1 3 5 4 ⎡ ⎤
Example 1.15: Matrices1 in
0 Row-Echelon
6 5 8 2 Form ⎢ 0 1 0 7 ⎥ 1 0 6 0
⎢ 0 0 1 2 7 3 ⎥ ⎢ ⎥ ⎢ 0 1 4 0 ⎥
⎢
The following augmented ⎥ , ⎢ 0 0 1form,
matrices are in row-echelon 0 ⎥but
⎢not in reduced⎥row-echelon form.
⎣ 0 0 0 0 1 1 ⎦ ⎢ ⎥,⎣ 0 0 1 0 ⎦
⎣
⎡ 0 0 0 1⎤ ⎦
⎡ 0 0 0 0 0 0⎤ 10 30 50 40 ⎡ 0 0 0 0⎤
1 0 6 5 8 2 ⎢ 0 1 0 7 ⎥ 1 0 6 0
⎢ 0 0 1 2 7 3 ⎥ ⎢ ⎥ ⎢ 0 1 4 0 ⎥
⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎣ 0 0 0 0 1 1 ⎦,⎢ 0 0 1 0 ⎥,⎣ 0 0 1 0 ⎦
⎣ 0 0 0 1 ⎦
Notice that we could 0apply
0 0further
0 0 row0 operations to these matrices
0 0 0to carry
0 them to reduced row-
echelon form. Take the time to try that on your0 own. 0 0 Consider
0 the following matrices, which are in
reduced row-echelon form.
Notice that we could apply further row operations to these matrices to carry them to reduced row-
Example 1.16: Matrices in Reduced Row-Echelon Form
echelon form. Take the time to try that on your own. Consider the following matrices, which are in
The following
reduced augmented
row-echelon form. matrices are in reduced row-echelon form.
⎡ ⎤
⎡ ⎤ 1 0 Form
0 0
Example 1.16: Matrices in Reduced
1 0 0 5 0 0 Row-Echelon ⎡ ⎤
⎢ ⎥
⎢ 0 0 1 2 0 0 ⎥ ⎢ 0 1 0 0 ⎥ 1 0 0 4
⎢
The following augmented ⎥ ⎢ row-echelon
matrices are in reduced ⎥ ⎣form. ⎦
⎣ 0 0 0 0 1 1 ⎦,⎢ 0 0 1 0 ⎥, 0 1 0 3
⎡⎣ 0 0 0 1 ⎤⎦ 0 0 1 2
⎡ 0 0 0 0 0 0⎤ 1 0 0 0
1 0 0 5 0 0 0 0 0 0 ⎡ ⎤
⎢ ⎥
⎢ 0 0 1 2 0 0 ⎥ ⎢ 0 1 0 0 ⎥ 1 0 0 4
⎢ ⎥ ⎢ ⎥ ⎣ ⎦
⎣ 0 0 0 0 1 1 ⎦,⎢ 0 0 1 0 ⎥, 0 1 0 3
⎣ 0 0 0 1 ⎦ 0 0 1 2
One way in which the0row-echelon
0 0 0 0form 0 of a matrix is useful is in identifying the pivot positions and
pivot columns of the matrix. 0 0 0 0

One way in which the row-echelon form of a matrix is useful is in identifying the pivot positions and
pivot columns of the matrix.
 For example consider the following.
18 Systems of Equations
Definition 1.17: Pivot Position and Pivot Column
18 Example
A Systems 1.18: Pivot
in a Position
of Equations
pivot position matrix is the location of a leading entry in the row-echelon formof a matrix.
Definition 1.17: Pivot Position and Pivot Column
A pivot
Let column is a column that contains a pivot position. ⎤
A pivot position in a matrix is the location of a ⎡
leading entry in the row-echelon formof a matrix.
ADefinition 1.17:isPivot
pivot column Position
a column thatand PivotaColumn
contains
1 2 3 4
pivot position.
⎣
A pivot position in a matrix is the locationAof=a leading
3 2 entry ⎦ row-echelon formof a matrix.
1 in6the
For example consider the following. 4 4 4 10
A pivot column is a column that contains a pivot position.
For example consider the following.
Where are the pivot positions and pivot columns of the augmented matrix A?
Example 1.18: Pivot Position
For example
Example 1.18:consider the following.
Pivot Position
Let ⎡ ⎤
Let
Solution. The row-echelon form of this matrix ⎡ is1 2 ⎤3 4
Example 1.18: Pivot Position
A1=⎡
2⎣ 3 4
3 2 1 ⎤6 ⎦
Let A = ⎡ 3 21 12 63 ⎦
⎣
4 4 4⎤ 4 10
41⎣ 42 43 104 3 ⎦
0 1 2 ⎦2
= ⎣ columns
3 2 1 of6 the
Wherearearethethe
Where pivot
pivot positions
positions andApivot
and pivot columns augmented
of the augmented matrix A?matrix A?
4 40 40 10 0 0
Where are the pivot positions and pivot columns of the augmented matrix A?
This is
Solution.
Solution. allrow-echelon
The
The we need inform
row-echelonthis example,
form
of thisbut
of matrix
this isnoteisthat this matrix is not in reduced row-echelon form.
matrix
⎡ in the original
In order to identify the pivot positions ⎤ matrix, we look for the leading entries in the
Solution. The row-echelon form of this matrix1is ⎡
2 13 24 3 4 ⎤
row-echelon form of the matrix. Here,⎣the entry
1 02 in
3 the
⎦ first row and first column, as well as the entry in
⎡0 ⎣ 2 ⎤2 3 ⎦
1
the second row and second column are the1 leading 2 3 4 entries.
2 Hence, these locations are the pivot positions.
0 0 0 03
⎣ 0 1matrix,
We identify the pivot positions in the original 02 02 ⎦as
0 in0 the following:
This is all we need in this example, but note0⎡that 0 this
0 matrix
0 is not⎤in reduced row-echelon form.
This is all we need in this example, but note1 that2 this 3 matrix
4 is not in reduced row-echelon form.
In order to identify the pivot positions in the original matrix, we⎦look for the leading entries in the
This is all we need in this example, but note⎣that3this 2matrix is
1 and not in
6matrix, reduced row-echelon form.
In orderform
row-echelon to identify the pivot
of the matrix. Here,positions
the entry in the thefirst
original
row we look
first column, for the
as well leading
as the entries in the
entry in
the In orderrow
second
row-echelon toform
identify
and of the
second pivot
thecolumn positions
matrix. areHere, inthe
the leading 4 in4 the
theentry
original
entries. 4 first
matrix,
Hence, 10these
werowlook forfirst
theare
locations
and leading entries
the pivot
column, in the
aspositions.
well as the entry in
row-echelon
We identify form ofpositions
the matrix. Here, the entry in the first
therow and first column, as well as the entry in
the secondthe row pivot
and second in the
column original
are thematrix, as in
leading following:
entries. Hence, these locations are the pivot positions.
Thus the pivot
the second row andcolumns
secondincolumn
the matrix
are theare the1.2.
leading first twoHence,
entries.
Systems columns.
Of these locations
Equations, are the
Algebraic pivot positions.
Procedures 19 ♠
We identifythe thepivot
pivot positions ⎡
in original
the original matrix, ⎤
as in the following:
We identify positions in the 1matrix, 2 as
3 in 4the following:
The following is an algorithm for⎣⎡carrying 3 ⎡ 2 1a matrix 6 ⎦ ⎤ to row-echelon
⎤ form and reduced row-echelon
Algorithm
form. You may 1.19:wish
Reduced
to useRow-Echelon
this algorithm Form
14 to 1Algorithm
2 3
carry2 4 the
3 above4 matrix to row-echelon form or reduced
4 4 10
⎣ 3⎣ 2 3 1 2 6 1 ⎦ 6 ⎦
row-echelon
This algorithm formprovides
yourself for practice.
a method for using row operations to take a matrix to its reduced row-
Thus the pivot
echelon form.columns
We beginin the matrix
with are theinfirst
the matrix 4 original
its two44columns.
4 form.
410 4 10 ♠

ThusThe
Thus thefollowing
the
1. pivot
Starting is anthe
pivotcolumns
columns
from algorithm
in the
in matrix
left,the
find for carrying
are
matrix
the the a two
arefirst
first the
nonzeromatrix to columns.
row-echelon
firstcolumns.
two
column. This is theform and reduced
first pivot and the ♠
column,row-echelon ♠
form. You may
position wish
at the to
topuse
of this
this algorithm
column is to
the carry
first the
pivot above matrix
position. to
Switch row-echelon
rows if form
necessary orplace
to reduced
The following
row-echelon is an algorithm
form yourself for for carrying a matrix to row-echelon form and reduced row-echelon
practice.
The following
a nonzero is
number an algorithm
in the first
form. You may wish to use this algorithm for position.
pivot carrying
to carry the a matrix to row-echelon
above matrix to row-echelonformform
and orreduced
reducedrow-echelon
form. You may yourself
row-echelon wish to use this algorithm to carry the above matrix to row-echelon form or reduced
2. Use form
row operationsfor to practice.
make the entries below the first pivot position (in the first pivot column)
row-echelon form yourself for practice.
equal to zero.

3. Ignoring the row containing the first pivot position, repeat steps 1 and 2 with the remaining
rows. Repeat the process until there are no more rows to modify.

4. Divide each nonzero row by the value of the leading entry, so that the leading entry becomes
1. The matrix will then be in row-echelon form.

The following step will carry the matrix from row-echelon form to reduced row-echelon form.

5. Moving from right to left, use row operations to create zeros in the entries of the pivot columns
which are above the pivot positions. The result will be a matrix in reduced row-echelon form.

Most often we will apply this algorithm to an augmented matrix in order to find the solution to a system
of linear equations. However, we can use this algorithm to compute the reduced row-echelon form of any
matrix which could be useful in other applications.
Consider the following example of Algorithm 1.19.

Example 1.20: Finding Row-Echelon Form and

Reduced Row-Echelon Form of a Matrix
Let ⎡ ⎤
0 −5 −4
A=⎣ 1 4 3 ⎦
 Most often we will apply this algorithm to an augmented matrix in order to find the solution to a system
Most often we will apply this algorithm to an augmented matrix in order to find the solution to a system
of linear equations. However, we can use this algorithm to compute the reduced row-echelon form of any
of linear equations. However, we can use this algorithm to compute the reduced row-echelon form of any
matrix which could be useful in other applications.
matrix which could be useful in other applications.
Consider the following example of Algorithm 1.19.
Consider the following example of Algorithm 1.19.

Example
Example1.20:
1.20:Finding
FindingRow-Echelon
Row-Echelon Form
Form and
and
Reduced
Reduced Row-Echelon Form
Row-Echelon Form ofof aa Matrix
Matrix
Let ⎡
Let ⎡ 0 −5 −4 ⎤ ⎤
0 −5 −4
AA = ⎣ 33 ⎦
= ⎣ 11 44 ⎦
55 10 10 77
Find
Find the row-echelon form of A. Then complete the process until
the row-echelon form of A . Then complete the process is in
A is
until A in reduced
reduced row-echelon
row-echelonform.
form.

Solution.
Solution.InInworking
workingthrough
throughthis
thisexample,
example, we
we will
will use
use the
the steps outlined in
steps outlined in Algorithm
Algorithm 1.19.
1.19.

1.1. The
Thefirst
firstpivot
pivotcolumn
columnisis the
the first
first column
column of of the
the matrix,
matrix, asas this is the
this is the first
first nonzero
nonzero column
column from
fromthethe
left. Hence the first pivot position is the one in the first row and first column. Switch
left. Hence the first pivot position is the one in the first row and first column. Switch the first two the first two
rows
rowstotoobtain
obtainaanonzero
nonzeroentry
entry inin the
the first
first pivot
pivot position, outlined in in aa box
box below.
below.
⎡
⎡ ⎤
11 4 3
⎣
⎣ 00 −5 −4 ⎦
20 Systems of Equations
55 10 7

2. Step two involves creating zeros in the entries below the first pivot position. The first entry of the
second row is already a zero. All we need to do is subtract 5 times the first row from the third row.
The resulting matrix is ⎡ ⎤
1 4 3
⎣ 0 −5 −4 ⎦
0 10 8

3. Now ignore the top row. Apply steps 1 and 2 to the smaller matrix
% &
−5 −4
10 8
In this matrix, the first column is a pivot column, and −5 is in the first pivot position. Therefore, we
need to create a zero below it. To do this, add 2 times the first row (of this matrix) to the second.
The resulting matrix is % &
−5 −4
0 0
Our original matrix now looks like ⎡ ⎤
1 4 3
⎣ 0 −5 −4 ⎦
0 0 0
We can see that there are no more rows to modify.

4. Now, we need to create leading 1s in each row. The first row already has a leading 1 so no work is
needed here. Divide the second row by −5 to create a leading 1. The resulting matrix is
⎡ ⎤
1 4 3
⎣ 0 1 45 ⎦
0 0 0
This matrix is now in row-echelon form.
5. Now create zeros in the entries above pivot positions in each column, in order to carry this matrix
all the way to reduced row-echelon form. Notice that there is no pivot position in the third column
so we do not need to create any zeros in this column! The column in which we need to create zeros
is the second. To do so, subtract 4 times the second row from the first row. The resulting matrix is
⎡ ⎤
1 0 − 15
⎢ 4 ⎥
⎣ 0 1 5 ⎦
0 0 0

This matrix is now in reduced row-echelon form. ♠

 2xOur+ 4yoriginal
− 3z =Example −1 now1.21:
matrix looks Finding
like ⎡ Solution to
0 ⎤0 0
the a System
1
5x + 10y We−can 7z = see−2 that there are no more⎣rows to4 modify. 3
⎦
3x + 6y + 5z = 9 Give the complete solution to
0 the
−5 following
−4 system of equations
0 0
4. Now, we need to create leading 1s in each row. The first row already has a leading 1 so no work is 0
We needed
can see that here.there are no
Divide themore secondrows1.2. Systems
to modify.
row by −5 toOf 2x +a4y
Equations,
create − 3z
leading =1.−1
Algebraic Procedures
The resulting matrix21is
5x + 10y − 7z = −2
matrix for this system 4. Now, is we need to create leading 1s in each row. The⎡first row already ⎤
1 43x + 3 6y + has 5z = a leading
9 1 so no work is
⎡ needed here. Divide ⎤ the second
In the next example, we look at how to solve a system of equations row by −5 to create a leading 1. The resulting matrix is augmented
2 4 −3 −1 ⎣ 0 1 using 4 ⎦ the corresponding
5
matrix. ⎡ ⎤
⎣ 5 10 −7 −2 ⎦ 1 4 30 0 0
3 6 Solution. 9 The augmented matrix ⎣ 0for1 this system is
4 ⎦
5 5
Example 1.21:This Finding
1.2.
matrix theisSolution
Systems now Of in to a System
Equations,
row-echelon Algebraicform. 0 0Procedures 0 ⎡ 21 ⎤
2 4 −3 −1
ution toGivethisthe complete
system, wematrix
ThisNow
solution
wish to carry
is now
to thethe following
in row-echelon
augmented system matrix of equations
form.above pivot positions
to reduced ⎣ 5 row- −2 ⎦ in order to carry this matrix
−7 column,
10in each
5. create zeros in the entries
so using
we look atAlgorithm
how to solve 1.19.a systemNotice of that the first using
equations column theiscorresponding
nonzero, so this is
augmented our
3that 6there5is no9pivot position in the third column
5. Now all the way toinreduced row-echelon
2x + 4y − 3z form.
= in −1Notice
st entry in the first row,create 2, is zeros
the first the entries
leading above
entry pivot
and itpositions
is theinfirst each column, in order to carry this matrix
pivot
all so we
the way doto not
reduced needrow-echelon
to create 5x + any10y
form. zeros
− 7z
Notice in= this
−2that column!
there is noThepivotcolumn
positionininwhichthe third wecolumn
need to create zeros
operations to create zeros in the In entries
order below
to find the
the 2. First, replace
solution to this the second
system, we wish to carry the augmented
so we is thedo notsecond.
need toTocreate do so, anysubtract
3x + in
zeros 6y4this
+times =the
5zcolumn! 9 second
The column row fromin whichthewe first
needrow. zeros matrix
The resulting
to create matrixtoisreduc
ingrowthe
plus 2 timesto
Solution thea second
System row.
echelon This
form. yieldsWe will do sothe using Algorithm
is the second. To do so, subtract 4 times second ⎡row from 1.19. ⎤ Notice
the first
1
row. The thatresulting
the firstmatrixcolumn is is nonzero, so thi
⎡ ⎤ 1 0 −
lution to the following 2 4system −3first−1 ofpivot
equationscolumn. The first ⎡ entry in the
1 0 ⎢− 15
⎤first row, 5 2, is the first leading entry and it is in the fir
4 ⎥ in the entries below the 2. First, replace the
Solution. The ⎣ augmented
0 0 position.
1 matrix We will use row
1 ⎦ for this system is ⎢ 0 1 ⎣ 04 ⎥1 operations to create zeros
5 ⎦
2x + 4y − row = −1−5 times the first row
3z with ⎣ plus 2 times ⎦ the second row. This yields
3 6 5 9 ⎡ 1.2. Systems ⎤0 Of0Equations,
5
0 Algebraic Procedures 21
5x + 10y − 7z = −2 2 4 −30 −1 0 0 ⎡ ⎤
with −3 times the 3x
first +row6y + 5z = 9 ⎣ 5 10 −7 −2 ⎦ 2 4 −3 −1
This Inmatrix
the nextplus to 2inwe
isexample,
now times lookthe
reduced at howthird row.aThis
to solve
row-echelon system
form. yields
of equations using the corresponding augmented ♠
This matrix is now in reduced row-echelon 3 form.
6 5 9 ⎣ 0 0 1 1 ⎦ ♠
⎡ matrix. ⎤
The2The 4 above
above −3 −1
algorithm
algorithm gives gives
you a simpleyou a simpleway to obtain way to theobtain 3 the
row-echelon 6 row-echelon
5 and
form 9 reduced
form and reduced row-echelon
row-echelon
d matrixInfor this
order ⎣ system
to 0of
form
form Example
find is
the
0 matrix.
aof 1.21:
1solution
a matrix. The1 main⎦
Finding to this
The the
main Solution
ideasystem,
isidea to
to doiswe a
row System
to wish
do row
operations to carry
operations
in suchthe augmented
a in
way as toa matrix
such wayupaswith
end totoreduced
end up row-
a matrix with
in a matrix in
echelon form. ⎡ WeGive
row-echelon
0 0 willthe Now,
complete
do
form
1 so
or
21 replace
using
reduced ⎤
solution theto
Algorithm third
the
row-echelon row
following
1.19.
form.with
system
Notice −3ofprocess
This times
equations
that the
the is firstcolumn
first
importantrow plus is
becauseto 2the
nonzero,times the
so
resulting third
this is row. This yields
our
matrix
row-echelon form or reduced row-echelon form. This process is important because the resulting matrix
2 4 −3 −1 ⎡ system
first pivot will allow
column.
will
⎣ allow you
Theyou tofirst
describe
toentry the
describe
⎦ in solutions
the
thefirst to
row,
solutions the 2,
2x + 4y
corresponding
tois−the the=
3z first linear
−1 leading
corresponding 2 4 entry
linear
−3
ofand it⎤is of
equations
system
−1 ininequations
a meaningful
the first pivot
in a meaningful
column below the
way. 5
pivot 10 −7
position −2
are zeros. We now look for the second pivot
position. Weway. will use row operations to create zeros 5x + 10y in −the 7z = entries
−2 ⎣below 0pivot 0 the 12. First, 1 ⎦replace the second
is column
row with three.
−5 times
3 the
Here, the6 first
1 in5row 9plus 2 times
the second row and thethird 3x +column
second 6y row.
+ 5z =This is9 in the yields
st one row operation to create a zero below the⎡1. 0 0 1 21
⎤
solution to this system, we wish to carry the augmented 2 4 column −3matrix −1below to reduced row-
cond row and adding it toThe
Solution. the
Now third the
augmented row yields
entries
matrix inforthethisfirst
system is the pivot position are zeros. We now look for the secon
o so using Algorithm 1.19. Notice that the first ⎣ column
0 0 is1 nonzero, 1 ⎦ so this is our
⎡ column, ⎤which in this case ⎡ is column three.
in the Here, the 1 in the second row and third column is in th
⎤
first entry in the first 2 4row, 2, is
−3position.−1the We first leading3entry 6 2 and 45 −3 it is9 −1 first pivot
need to do just
⎣ one row
−7replace −2 ⎦ the second a zero below the 1.
operation to create
w operations to⎣create 0 0 zeros 1 in the 1 ⎦entries below the 52. 10 First,
Now,
rst row replace
plus 2 timesthe
0 thethird
0 second 0 Taking
row row.−3
with
20 −1
This times
times
yields thethe first 3row6 plus
second row 5 to and92 times
adding theit third
to therow.
thirdThis rowyields
yields
⎡ In order to find the solution ⎤ to ⎡ this system, we wish to⎤carry the
⎡ augmented matrix⎤to reduced row-
2 4 −3 −1 2 4 −3 −1 2 4 −3 −1
the algorithm to carry echelon thisform.
matrixWe will to do so using Algorithm
row-echelon form 1.19.
or Notice that
reduced the first column is nonzero, so this is our
row-echelon
⎣ 0pivot 0 column. 1 The 1 ⎦first entry ⎣ 0 first
0 row, 1 2, is1the⎦ first leading ⎣ 0 0 entry1and it1is⎦in the first pivot
that we are looking 22first for the solutions
Systems of Equations to the system in the
of equations. Take another
3 6We will
position. 5 use9row operations to0 create 0 zeros 1 in21the entries below 0 0 the 2.0 First, 20 replace the second
matrix. Notice that rowitwith corresponds
−5 times thetofirst therow equation
plus 2 times the second row. This yields
Now−3
ow with thetimes 22
entriestheinSystems
There isthe
firstnorowfirstofplus
solution Equations
column
We to 2 below
tothis
could equation
times
proceed thebecause
the pivot
third
with ⎡ the position
for all
row. This x, y,are
z, ⎤
yields zeros.
the left sideWewill nowequal
look0 andfor the0 ̸= second
20. This pivot
shows
0x + 0y + 0z = 20 2 4 algorithm −3 −1 to carry this matrix to row-echelon form or reduced row-
there
column, which⎡in this caseis no solution
form. to
is column the given
⎤ three.
However, system
remember Here, of equations.
⎣ the
0that 01 we in1the In other
are1second
looking
⎦ words,
row this
forand system
the third is inconsistent.
column
solutions to the is in of♠equations. Take
the pivot
system
position. WeThere need 2istono4do −3
just at
solution −1the
one row operation tohow create ax,
5zero below the 1.system ̸= 20.
The look
following istoanother
this equation
third rowbecause
example ofofthe for6toall
3matrix. find z,9the
y,the
Notice left
solutionthatside awill equalof0equations
it corresponds
to and
to 0the byThis
equation shows
carrying the
⎣
there 0
is no 0 solution1 to 1the⎦ given system of equations. In other words, this system is inconsistent. ♠
Taking −1 times
corresponding the second
augmented row and
matrix adding
to reducedit to the
row-echelon third
Now, replace the third row with −3 times the first row plus to 2 times the third row. This yields row form. yields
0 0 1 21 0xto+a0y + 0z of
= equations
20
The following is another example ⎡ of how ⎡ to find the ⎤ solution
⎤ system by carrying the
Example 1.22:
corresponding augmentedAn InfinitematrixSet 2 4
of Solutions
to reduced 2 −34 −3 −1
row-echelon form. −1
rst column below the pivot position are zeros.⎣We 0 now ⎣ 0 look
0 of01equations 1for 1 the1 ⎦second pivot
⎦
se is column three. Give Here, thethecomplete
1 in the solution
secondtorow the system
and 0third0 column 1 21 is in the pivot
Example 1.22: An Infinite Set of 0Solutions 0 0 20
just one row operation to create a zero below the 1.
Now the entries in the first column below the pivot 3x −positiony − 5z = 9
Give the complete solution to the system of equations are zeros. We now look for the second pivot
second We rowcould
and adding
column, it to
which the
in third
this caserow is
proceed with the algorithm to carry this matrix yields
column three. Here, the
y − 110z
into = 0second row andform
therow-echelon third column
or reducedis in the pivot(1.8)
row-echelon
position. We need to do just one row operation3x to−2x create + ya =zero−6 below the 1.
form. However, ⎡ remember that we ⎤ are looking for the−solutions y − 5z = 9to the system of equations. Take another
2
Taking 4 −1−3 times −1 the second row and adding ity −
look at the third row of the matrix. Notice that it corresponds to 10z
the third
=to 0 rowthe yields
equation (1.8)
⎣ 0 0 1 1 ⎦ ⎡ −2x + y = −6 ⎤
Solution. The augmented matrix of this system 2 4 is−3 −1
0 0 0 20 0x +⎣0y0 +00z = 1 201 ⎦
⎡ ⎤
0 30 −10 20 −5 9
Solution. The augmented matrix of this system is
th the algorithm to carry this matrix to row-echelon⎡form ⎣ 0 or reduced 0⎤⎦
1 −10 row-echelon
We could proceed with the algorithm to carry this matrix to row-echelon form or reduced row-echelon
ber that we are looking for the solutions to the system −2
form. However, remember that we are looking
of for
−11 −50 Take
3 equations.
the solutions
−6
9 another
to the system of equations. Take another
he matrix. Noticelook
thatatitthe
corresponds to matrix.
the equation ⎣ 0 1 −10 0 ⎦
third row of the Notice
In order to find the solution to this system, that it corresponds to the equation
−2we will 1 carry0 −6 the augmented matrix to reduced row-echelon
form, using Algorithm 1.19. The first column 0x + 0yis +the
0zfirst
= 20pivot column. We want to use row operations to
In
0x +to0yfind
create
orderzeros
+ 0z
beneath
= 20
the firsttoentry
the solution in this column,
this system, whichthe
we will carry is in the first pivot
augmented position.
matrix Replace
to reduced the third
row-echelon
 ⎣ first
0 1column, −10matrix 0 ⎦we
order to findNow, we have to
the solution created
this system, zeros beneath we willthe carry 3 inthe the augmented
1.2.3x −
Systemsy − 5z Of=
so tomove
Equations,
9column, reduced on to
Algebraic
the second pivot
row-echelon
Procedures 23
column
matrix of this system
(which is Now,
is
the we
second have
column)
rm, using Algorithm 1.19. The first column is the first pivot column. created and zeros
repeat beneath
the the
procedure. 0 3 0in theTake first
0 −1 0 times theso we
second moverow on to
and the
add second
to the pivot colu
third
(which is the second column) and repeat the 10zWe
y −procedure. = want
0 Take to use row operations to
−1 times the second row and (1.8)
add to the th
eate zeros beneath ⎡ 1 the first entry in this
row. ⎤ column, which ⎡ is in the −2x first
+ y pivot
= −6 ⎤position. Replace the third
times the second row to the first row.
w with 2 times the first−1
3 3 This row.
row is in−5reduced
added to9 3 timesrow-echelon the thirdform, row. 3 which−1 ⎡
This you
gives −5should
−1
9 verify using
−5
⎤ Definition 1.13. The equations
⎣ corresponding
0 1 −10to this 0 ⎦reduced row-echelon ⎡⎣ 0 form are 3
−10 0 ⎤ ⎦ 9
⎡ 1 0⎤ 1−5 ⎣ 03 1 −10 0 ⎦
−2 1 0 −6 −1 of−5 ⎣ 0 9 1 −10 00⎦ 0
0 0
Solution. The augmented3matrix this systemxis−05z =03 0 0
⎣ 0 1 the −10 0 0 0⎦column 0 0 is now a zero. Notice that we have no more pivot
The
on to this system, we will entry below carry the pivot
the position
augmented in
matrix second
to ⎡reduced y −row-echelon
10z = 0 ⎤
columns The entry
because we below
have the0pivot
only two −10 entries.
position
1leading in
0 the 3 second−1 column −5 9is now a zero. Notice that we have no more pi
19. The first columnThis is columns
theis infirst pivot
reducedbecause column.
row-echelon
we have Weform,wantwhich
only twoto⎣use you
leading row shouldoperationsverify using to ⎦Definition 1.13. The equations
or 0 entries. 1 −10 0
first
ow, entry
we havein this
createdAt this
column,
corresponding
zeros stage,
which
beneath wethis
to isalsoin want
the
reduced
the 3 infirst thepivot
leading
row-echelon
the first position.entries
form
column, are
so to
Replace
we
x be
= move
3equal
+ the5zonto
third
toone.
theTo do so,pivot
second divide the first row by 3.
column
At this stage, we also want the leading −2 entries 1 0 be−6
to equal to one. To do so, divide the first row by 3
row added to 3 times the third row.
which is the second column) and repeat the procedure. Take This gives ⎡ x−1 y =the ⎤ row and add to the third
1 −−5z1times =
⎡ 3− 53 Of
10zsecond
3 ⎤
w. ⎡ In order to find the solution ⎤ ⎡ to this system, 1.2.
y −⎤10z we =3 Systems
will01 carry − 1 Equations,
the− Algebraic
5augmented
3 Procedures
matrix to reduced 23 row-echelon
−1 −5
3 Observe that z9 is not restrained 3 −1 by
−5 ⎢
any 9 equation. In fact,
3 ⎥ z can
3 equal any number. For example, we can
form, using 1.2. Algorithm
Systems Of ⎣ 1.19. The first
Equations, ⎣column
Algebraic 0 Procedures 1is⎢ −10first0 pivot
the ⎦23 column. ⎥ We want to use row operations to
⎣or
let 0 z = 1
t, −10
where we 0 ⎦
can choose
0first row.t to
1 −10 be any ⎦
number.
0x column,
1.2. ⎣ In
0 which
Systems thisOf context
−10
1 Equations, t ⎦
is called
0 Algebraic a parameter
Procedures .
23Therefore, the
1 create zeros beneath the first entry in this0 = 30 + 5z 0 0 is in the first pivot position. Replace the third
times the second row to the
0
3solution 1 set−10 of this 0 system is 0
row with 2 times the first 0row added 0 0 y = 10z
to
⎡ 3 times the0third 0
⎤ row. This gives
0 0
the first row. x = 33 + 5t
⎡0 aIn−5
1
he
os entry below
beneath the 3the This times
in3pivot
the matrix
Observe
the second
position
first is now
column,
that z inrow
is in
the
not so
tosecond
we
the firstcolumn
row-echelon
restrained move
row. form.
by onany to is1now
the
equation.
⎣ second zero. =0 Notice
ypivot
fact, column
⎦can
z10t equal that⎤any
wenumber.
have noFor more pivotwe can
example,
lumns because we⎡ have only
This matrix
two ⎤
leading
is now in row-echelon
entries. ⎡ 0 1 form. 3−10 −1 ⎤ −5 9
mn) and repeat the Let’s 1z =continue
letprocedure.0t, where −5Take we 3can
with −1rowchoose
times to be
t the
operations anyuntil
second number.row ⎣00and
01 the In −5this
add zcontext
toistthe t reduced
inthird parameter . Therefore,
is called arow-echelon the involves
solution set Let’s
of this continue
system is with row operations ⎣
0matrix
until01 03=
the −10
⎦ matrix 0 ⎦is in reduced row-echelon form. This
form. This invol
At this stage, ⎡⎣ also
creating
we 0 zeros
1 want −10abovethe 0leading ⎦⎤pivot
the positions
entries to beinequal each 0 1to pivot
0
−10column.
one. 1 To 0 do so,
−10 This0 requires
divide onlyrow
the first oneby step,
3. which is to add
where3This t is arbitrary.
creating
−1 −5zeros The system
above
9 row-echelon the pivot haspositions
an 0infinite +set
0= 3inshould
x you 5t ofpivot
each solutionsusing which
column. This arerequires
givenThe by these
only oneequations.
step, whichFor is to a
⎣ any 0 value0 is inofreduced
0t we 0 select, ⎦ ⎡x, y, and form,
1 z
which
will 5 be ⎤= 10t0by 0the
given
y
verify
above
Definition
equations. For
1.13.
example,
equations
if we choose t = 4
Now,0 we
corresponding 1 have −10 to this0 reduced 1row-echelon
created zeros − 3 −the
beneath form3 are
3 3 in the first column, so we move on to the second pivot column
z = t
-echelon form, which
then 0 This you
is in reduced
the0 corresponding
should 0second 0verify row-echelon
⎢solution would
using
form, which
Definition
be you ⎥ should verify using Definition 1.13. The equations
1.13. The −1 times the second row and add to the third
(which
corresponding is the to this reduced
where t is arbitrary. The system has an infinite set ⎣
column) 0 and
row-echelon
1 repeat
−10 form the
0x −⎦procedure.
are 5z 3 equations
of=solutions
Take
which are given by these equations. For
ced row-echelon
position in the second form
row.value
any are
column of t weisselect,now ax,zero. 0 Notice
y, and 0 bethat
z will 0 we
given y0⎡ − have
x10z
by =the3 =no+0 5(4)
above more =pivot
23 ⎤ For example, if we choose t = 4
equations.
x −35z = −1 3 −5
only two leading entries.
then the corresponding solution would be
or −
y ==10(4) = 409
x − 5z = 3 y⎣ 10z 0
03 + 5zz1 =−10 ⎦
want
Thisthe leading
matrix is nowentries in to be
row-echelon equal to one.
form. To do so, divide x= the first row 4 by 3.0
or y − 10z = 0 x = 3y+=05(4) 10z = 0 23 0 0
Let’s continue⎡with row operations ⎤ until the matrix is y =inx = 3
reduced
10(4) + 5z
= 40row-echelon form. This involves ♠
The − 13 that
1Observe
entry − 53z is3the
below pivot
not restrained positionby any in the
equation. second
yz== 10z column
In4fact, z canis nowany
equal a zero.
number.Notice that we have
For example, we canno more pivot
eating zeros above ⎢ thex pivot positions ⎥ in each pivot column. This requires only one step, which is to add
⎣letcolumns
0z = t,=1where
InObserve
Example
+ 5z
3because
−10 we can
that z1.22
we
0is not
have
⎦choose
the only
t to be
solution
restrained
two
any
by any
leading
number.
involved
equation.
entries.
InInthis
one context t isItcalled
parameter.
fact, z can equal
parameter
ahappen
maynumber.
any Forthat
. Therefore,
the solution
example,
the
we can♠ to a system
solution y =
set 10z
of this system is
0 zAt= 0this
involves
let t, more stage,
where we0we
0than can one also
choose want
parameter, theany
t to be leading
as shownentries
number. in thisto
In the be equal
following
context to one.
example.
t is called To do so,. divide
a parameter thethe
Therefore, first row by 3.
In
solution Example set of 1.22
this the
system solution
is involved one = 3 + 5t It may happen that the solution to a system
xparameter.
strained by any equation. In fact, z can equal any number. For ⎡ example, we can⎤
row tand
oose to add
be to number.
any the
involves second. more
In thisThen
than one
context take −1
parameter,
t is times
called the
as shown
a first
parameter in the xy==
row 10t
3and
.+−
1following5t13addexample.
Therefore, −to53the 3
the
row-echelon form. Example 1.23: A Two Parameter Set ⎢of yz=Solutions
=10t t ⎥
is operations ⎡ ⎤ ⎣ z0= t This 1 −10 0 ⎦
ow until
1where the
2 t is
Find
Example matrix
−1 the
arbitrary.
1.23: is
1solution in
A3The reduced
Two tosystem
the row-echelon
system
Parameter has anSet infinite
of Solutionsform. involves
set of solutions which are given by these equations. For
24positions
Systems of x =
Equations 3 + 5t
ivot ⎣ in0anyeach−1
where pivot
value
Findy tthe 0 column.
of
is t0 we select,
arbitrary.
solution −2toThe ⎦ x, y,
This
the system
requires
and z willonly
system has an infinite set
one step,
be given x 0+
by the
of2ywhich−0 z +is
above
solutions
0=add
wto
equations.
which
03 areFor example,
given by these if we =4
choose tFor
equations.
then the = 10t
corresponding solution would be
0 any1 value0 of0t we select, 2 x, y, and z will be given x + 2yby x+
− the −=
z +yabove
w z+ w = 1 For example, if we choose t = 4
3 equations.
thenThis =
thez corresponding
t
matrix is now in row-echelon
solution would be xform.
Take −1 times the first row and add to the second. x= +3Then
y+ x+
− z+
5(4)3y w=
take− =23 + wtimes
z1−1 = 5 the first row and add to the
wystem
to thehas
third anrow andLet’s
infinite divide
set of the
solutions
continue second with row
which row by are−1. given
operations by
x + these
until
3y − the
z +equations.
w matrix
= 5 is Forin reduced row-echelon form. This involves
third. This yields
⎡ ⎤ ⎡ xy= =310(4)+ 5(4)=⎤=4023
y, and z will be given creating by the zeros above above equations.
the pivot 1 For example,
positions
2 −1 y = in if
each we choose
pivot
z = 34 = 40
1 10(4) column.t = 4 This requires only one step, which is to add
1 2 −1 1 3
ution would be⎣ Solution. The augmented
ions ⎣matrix0 is −1is 0 ⎡ =4⎦
0 1
Solution. 0The0augmented 2 ⎦ matrix 0 z−2 (1.9) ⎤ ♠
⎡ 1 2 −1 ⎤1 3
0x =03 + 5(4) 0 0 =023 0 1 01 02 −12 1 3 ♠
⎣ one ⎣ 1−11 1 −1 ⎦
1 1happen that the solution to a system
toyIn Example 1It ⎦
first rowNowandadd add the =
the
second second.
10(4) row =1.22 40
to Then
the
the solution
thirdtake row−1involved
times
and dividethe 1 first1parameter.
the row
second
3 1 −1 and row
may
add
1byto 5−1.the that the solution to a system
chelon form andinvolves In
we canzmore Example
see than one
that 1.22
x andthe solution
parameter,
y correspond involved
as shown to one
1in 3 1following
the
pivot parameter.
−1
columns, It may
5 example.
while happen
involves= 4 than one parameter, ⎤ ⎡z and ⎤
we will assign ⎡
more as shown in the following example.
parameters
We1wish
We wish2totocarry to the
carry
−1 1this
this variables
matrixmatrix
3 to row-echelon 2w. −1
to1row-echelon Assign
form. 1 form. the
3 parameter
Here, Here,
we will we s the
will
outline outline the row operations
row operations used. However,used. However,
Example 1.23: A Two Parameter
x +⎣ Set
− 1sSet+ of Solutions
w. Then the first⎣make row
make yields
sure
−1surethat
0 Example the
that you
01.23: equation
0understand
you A −2 Two⎦Parameter
understand the2y0steps
the steps of0=Solutions
intterms in3,0terms
while
of ⎦ofthe
2Algorithm second
1.19. ♠ 1.19.
Algorithm (1.9)
2. Since y = 2, the0Find firstthe equation
solution
1the solution
0 0 It tomay becomes
to the system
2the system x + 0 4 −0 s + t0 = 0 3 showing
0 that the
olution involved oneFind parameter. happen that the solution to a system
x + 2y − z + w = 3
rameter, as shown x + 2y − z + w = 3
row to the
This matrix xin
third row the
is=in −1
and following
+ s − t theexample.
divide
row-echelon second
form androwwe −1.see
bycan x+y−z+w = 1
x +that
y − z x+and
w = 1y correspond to pivot columns, while
⎡ y = 2 ⎤ x + 3y − z + w = 5
z and w do not. Therefore, x + 3y − z + w = 5
Parameter Set of 1z = 2 s−1we1 will
Solutions 3 assign parameters to the variables z and w. Assign the parameter s
to z and the parameter
⎣ 0 1t to 0w. Then 0 2 the ⎦ first row yields the equation x + 2y(1.9) − s + t = 3, while the second
row yields theSolution.
system equationw= Theyt =augmented
2. Since y = 2,
matrix the
is first equation becomes x + 4 − s + t = 3 showing that the
0 0 The0augmented
Solution. 0 0 matrix is ⎡⎡ ⎤
⎤
solutioninisthe
solution x+
given
form by2y − z + w = 3 1 2 −1
1 2 −1 1 3 1 3
⎣⎣ +11 s1− t−1 11 ⎦
⎡ ⎤
x + y⎡− z + w = 1 ⎤
w-echelon form and we can see that x and y correspond x = −1 to −1 11columns,
1 pivot ⎦
while
x x + 3y −−1 z + +w s=− t
5 y = 11
2 33 −1
−1 1
1 55
ore, we will⎢assignyWe⎥
parameters
⎢to carry2 this to the⎥ variables z and w. Assign the parameter s
to w. Then ⎢ the ⎥
Wewish
first
⎣ zmake
⎦ = ⎢ yields
wish
row
⎣
to carry the
s
⎥ totorow-echelon
thismatrix
matrix
equation
⎦
row-echelon
x + 2y −z= sform.
st=
form.
+ Here,
Here, we
we will
3, while outline the
willthe
outline
(1.10) therow
second rowoperations
operationsused.
used.However,
However,
sure that you understand the steps
you understand the stepswin=in terms of Algorithm 1.19.
y = 2. Since yw= make2, thesurefirstthat
equation
t becomes x + 4 −terms st+ t of
=Algorithm
3 showing 1.19.
that the
matrix is ⎡ ⎤
It is customary to write this solution in the form
1 2 −1 1 3
⎣ 1 1x =−1 −1 + s − t⎦
1 1
⎡ ⎤ ⎡
x −1 + s − t
⎤ ♠
y=2
ystem of equations1 with 3 an−1infinite
1 5 ⎢ set ⎥ ⎢ depends 2 on two ⎥ param-
z = s solution ⎢ y which
⎥ ⎢ ⎥
 ⎡ ⎤ ⎡ ⎤ ⎡ ⎤
solution is given by −4 −3 −18
3 ⎣ 1 ⎦ + 2 ⎣ x0=⎦−1
= ⎣+ s −3 t⎦
0 1 y=2 2
z=s
You should take a moment to verify that
w=t
⎤ ⎡ ⎡ ⎤
It is customary to write this solution in x
the form−18
⎣ y ⎦=⎣ 3 ⎦
⎡ ⎤ ⎡ ⎤
z x 2 −1 + s − t
⎢ ⎥ 2 ⎢ ⎥
⎢ y ⎥=⎢
is in fact a solution to the system in Example 1.31. ⎥ (1.10)
⎣ z ⎦ ⎣ s ⎦
Another way in which we can find out more information about the solutions of a homogeneous system
w matrix. We now
is to consider the rank of the associated coefficient t define what is meant by the rank of a
matrix.
♠
Definition 1.33: Rank of a Matrix
This example shows a system of equations with an infinite solution set which depends on two param-
Let A be a matrix and consider any row-echelon form of A. Then, the number r of leading entries
eters. It can be less confusing in the case of an infinite solution set to first place the augmented matrix in
of A does not depend on the row-echelon form you choose, and is called the rank of A. We denote
reduced row-echelon
it by rank (A). form rather than just row-echelon form before seeking to write down the description
of the solution.
32 Systems of Equations
In the above steps, this means we don’t stop with the row-echelon form in equation 1.9. Instead we
Similarly, we could count the number of pivot positions (or pivot columns) to determine the rank of A.
first place it in reduced row-echelon form as follows.
rank of the coefficient matrix of this system, we can find out even more about the solution. Note that we
are Example
looking at1.34:
just the coefficient
Finding matrix,
the Rank of not the⎡entire
a Matrix 1 0augmented
−1 1 matrix.
−1
⎤
Consider the matrix ⎡⎣ 0 1 ⎤ 0 0 2 ⎦
Theorem 1.35: Rank and Solutions to a Homogeneous
1 0 2 03 0System
⎣ ⎦
0 0
1 5 9 to a homogeneous system of equations, and
Let A be the m × n coefficient matrix corresponding
Then the A
suppose solution = 2 from
has rankisr.yThen, to the2 corresponding
the second
the solution 4 6and x = system
row −1 + zhas
− wn −from the first. Thus letting z = s and
r parameters.
= t, the
w What is itssolution
rank? is given by 1.10.
You can
Consider oursee here
above that there
Example 1.31 in arethetwo paths
context to the
of this correct
theorem. Theanswer,
system in which both yield
this example has mthe = 2 same answer.
Hence,
Solution. either approach may be used. The process which we first used
equations in n = 3 variables. First, because n > m, we know that the system has a nontrivial solution, find
First, we need to find the reduced row-echelon form of A. Through the usualin the above
algorithm, we solution
and is called
that this
Gaussian
therefore is Elimination
infinitely This process
many solutions. This tellsinvolves
⎡ us that carrying ⎤ thewill
the solution matrix
containto at
row-echelon form, converting
least one parameter. The back to
rank of the coefficient matrix can tell us even more 0 −1the solution! The rank of the coefficient matrix
1 about
32equations,
Systemsand using back substitution⎣ to find the solution.
of Equations When you do row operations until you obtain
ofreduced
the system is 1, as it hasform,
one leading 0 1
entry inisrow-echelon 2 ⎦form. Theorem 1.35 tells us that the solution
row-echelon the process called Gauss-Jordan Elimination.
will have n − r = 3 − 1 = 2 parameters. You can0check 0 that0 this is true in the solution to Example 1.31.
rank of
wethe
HereNotice coefficient
that
have =m matrix
if n leading
two n< ofm,
this
or entries, itorsystem,
istwo
possible weto
pivot can findeither
have
positions,out even moresolution
a unique
shown above about the solution.
(which
in boxes.The Note
willof
rank isthat
beAthe r =we
trivial
2.
are looking at just the coefficient
solution) or infinitely many solutions. matrix, not the entire augmented matrix. ♠
We are not limited to homogeneous systems of equations here. The rank of a matrix can be used to
Notice that
Theorem we Rank
1.35: wouldand have achievedtothe
Solutions same answer ifSystem
a Homogeneous we had found the row-echelon form of A
learn about the solutions of any system of linear equations. In the previous section, we discussed that a
instead of the reduced row-echelon form.
system the m × can
Let Aofbeequations have no solution,
n coefficient a unique solution,
matrix corresponding or infinitely many
to a homogeneous systemsolutions. Suppose
of equations, and the
Suppose
system Awehashave
rankawhether
is consistent,
suppose rhomogeneous
. Then,itthe systemtoofthe
is homogeneous
solution equations
mor not. Theinfollowing
correspondingn variables,
hasand
n −suppose
systemtheorem rtells usthat > m.
hown we
parameters. canFrom
use
our abovetodiscussion,
the rank learn about wethe know
type ofthat this system
solution we have.will have infinitely many solutions. If we consider the

Consider our above Example 1.31 in the context of this theorem. The system in this example has m = 2
Theorem 1.36: Rank and Solutions to a Consistent System of Equations
equations in n = 3 variables. First, because n > m, we know that the system has a nontrivial solution, and
therefore be the m ×
Let A infinitely (n +solutions.
many 1) augmented
Thismatrix
tells uscorresponding to awill
that the solution consistent
contain system
at leastof
oneequations in nThe
parameter.
rankvariables, and suppose
of the coefficient A has
matrix canrank . Then
tell rus even more about the solution! The rank of the coefficient matrix
of the system is 1, as it has one leading entry in row-echelon form. Theorem 1.35 tells us that the solution
1. the system has a unique solution if r = n
will have n − r = 3 − 1 = 2 parameters. You can check that this is true in the solution to Example 1.31.
Notice
2. thethat if n =has
system m or n < m, many
infinitely it is possible
solutions r < either
to ifhave n a unique solution (which will be the trivial
solution) or infinitely many solutions.
We are not limited to homogeneous systems of equations here. The rank of a matrix can be used to
We will not present a formal proof of this, but consider the following discussions.
learn about the solutions of any system of linear equations. In the previous section, we discussed that a
system of equations can have no solution, a unique solution, or infinitely many solutions. Suppose the
1. No
system consistent,The
is Solution above ittheorem
whether assumes that
is homogeneous the system
or not. is consistent,
The following that tells
theorem is, that
us it
howhaswe
a solution.
can use
It turns out that it is possible for the
the rank to learn about the type of solution we have. augmented matrix of a system with no solution to have any
rank r as long as r > 1. Therefore, we must know that the system is consistent in order to use this
theorem!
Theorem 1.36: Rank and Solutions to a Consistent System of Equations
2. Unique
Let m × (n +Suppose
A be theSolution r = n. matrix
1) augmented Then, there is a pivot to
corresponding position in every
a consistent column
system of the coefficient
of equations in n
matrix of A. Hence, there is a
variables, and suppose A has rank r. Thenunique solution.

3. 1.
Infinitely Manyhas
the system Solutions solution rif<
a uniqueSuppose r=n. nThen there are infinitely many solutions. There are less
pivot positions (and hence less leading entries) than columns, meaning that not every column is a
pivot
2. thecolumn. Theinfinitely
system has columnsmanywhichsolutions if r <columns
are not pivot n correspond to parameters. In fact, in this
 2. the system has infinitely many solutions if r < n

We will not present a formal proof of this, but consider the following discussions.
44 Systems of Equations
1. No Solution The above theorem assumes!that the system" is consistent, that is, that it has a solution.
1 2 matrix
It turns out that it is possible for the augmented 2 of a system with no solution to have any
rank r as long as r > 1. Therefore, we must2 know
h k that the system is consistent in order to use this
theorem!

2. Unique
Exercise Solution
1.2.17 Suppose
Determine r=
if the n. Then,
system there is aIfpivot
is consistent. so, isposition in every
the solution column of the coefficient
unique?
matrix of A. Hence, there is a unique solution.
x + 2y + z − w = 2
3. Infinitely Many Solutions Suppose r <xn. − yThen
+ z +there
w = 1are infinitely many solutions. There are less
44 pivotSystems
positionsof(and
Equations 2x + y −than
hence less leading entries) z = 1columns, meaning that not every column is a
pivot column. The columns which are not 4x + 2y +columns
pivot z = 5 correspond to parameters. In fact, in this
case we have n − r parameters. ! "
1 2 2
Exercise 1.2.18 Determine if the system is consistent. If 2so, h k solution unique?
is the

x + 2y + z − w = 2
x − yis+consistent.
Exercise 1.2.17 Determine if the system z + w = 0 If so, is the solution unique?
2x + y − z = 1
4x + 2yx++ 2y
z =+3z−w = 2
x−y+z+w = 1
2x + y − z = 1
Exercise 1.2.19 Determine which matrices are in reduced row-echelon form.
4x + 2y + z = 5
! "
1 2 0
(a)
0 1 7
⎡ 1.2.18 Determine
Exercise ⎤ if the system is consistent. If so, is the solution unique?
1 0 0 0
(b) ⎣ 0 0 1 2 ⎦
0 0 0 0
x + 2y + z − w = 2
⎡ ⎤ x−y+z+w = 0
1 1 0 0 0 5 2x + y − z = 1
(c) ⎣ 0 0 1 2 0 4 ⎦ 4x + 2y + z = 3
0 0 0 0 1 3

Exercise
Exercise1.2.20 Row
1.2.19 reduce the following
Determine matrix toare
which matrices obtain the row-echelon
in reduced form. Then
row-echelon continue to obtain
form.
the reduced row-echelon form. ⎡ ⎤
! " 2 −1 3 −1
1 2 0 ⎣ 1 0 2 1 ⎦
(a)
0 1 7 1 −1 1 −2
⎡ ⎤
1 0 0 0
Exercise 1.2.21 Row reduce the following matrix to obtain the row-echelon form. Then continue to obtain
(b) ⎣ 0 0 1 2 ⎦
the reduced row-echelon form. ⎡ ⎤
0 0 0 0 0 0 −1 −1
⎡ ⎤ ⎣ 1 1 1 0 ⎦
1 1 0 0 0 5 1 1 0 −1
(c) ⎣ 0 0 1 2 0 4 ⎦
0 0 0 0 1 3

Exercise 1.2.20 Row reduce the following matrix to obtain the row-echelon form. Then continue to obtain
the reduced row-echelon form. ⎡ ⎤
2 −1 3 −1
⎣ 1 0 2 1 ⎦
1 −1 1 −2

Exercise 1.2.21 Row reduce the following matrix to obtain the row-echelon form. Then continue to obtain
the reduced row-echelon form. ⎡ ⎤
0 0 −1 −1
 1.2. Systems Of Equations, Algebraic Procedures 45

Exercise 1.2.22 Row reduce the following matrix to obtain the row-echelon form. Then continue to obtain
the reduced row-echelon form. ⎡ ⎤
3 −6 −7 −8
⎣ 1 −2 −2 −2 ⎦
1 −2 −3 −4

Exercise 1.2.23 Row reduce the following matrix to obtain the row-echelon form. Then continue to obtain
the reduced row-echelon form. ⎡ ⎤
2 4 5 15
⎣ 1 2 3 9 ⎦
1 2 2 6

Exercise 1.2.24 Row reduce the following matrix to obtain the row-echelon form. Then continue to obtain
the reduced row-echelon form. ⎡ ⎤
4 −1 7 10
⎣ 1 0 3 3 ⎦
1 −1 −2 1

Exercise 1.2.25 Row reduce the following matrix to obtain the row-echelon form. Then continue to obtain
the reduced row-echelon form. ⎡ ⎤
3 5 −4 2
⎣ 1 2 −1 1 ⎦
1 1 −2 0

Exercise 1.2.26 Row reduce the following matrix to obtain the row-echelon form. Then continue to obtain
the reduced row-echelon form. ⎡ ⎤
−2 3 −8 7
⎣ 1 −2 5 −5 ⎦
1 −3 7 −8

Exercise 1.2.27 Find the solution of the system whose augmented matrix is
⎡ ⎤
1 2 0 2
⎣ 1 3 4 2 ⎦
1 0 2 1

Exercise 1.2.28 Find the solution of the system whose augmented matrix is
⎡ ⎤
1 2 0 2
⎣ 2 0 1 1 ⎦
3 2 1 3
Exercise 1.2.29 Find the solution of the system whose augmented matrix is
! "
1 1 0 1
1 0 4 2

Exercise 1.2.30 Find the solution of the system whose augmented matrix is
⎡ ⎤
1 0 2 1 1 2
⎢ 0 1 0 1 2 1 ⎥
⎢ ⎥
⎣ 1 2 0 0 1 3 ⎦
1 0 1 0 2 2

Exercise 1.2.31 Find the solution of the system whose augmented matrix is
⎡ ⎤
1 0 2 1 1 2
⎢ 0 1 0 1 2 1 ⎥
⎢ ⎥
⎣ 0 2 0 0 1 3 ⎦
1 −1 2 2 2 0

Exercise 1.2.32 Find the solution to the system of equations, 7x + 14y + 15z = 22, 2x + 4y + 3z = 5, and
3x + 6y + 10z = 13.

Exercise 1.2.33 Find the solution to the system of equations, 3x − y + 4z = 6, y + 8z = 0, and −2x + y =
−4.

Exercise 1.2.34 Find the solution to the system of equations, 9x − 2y + 4z = −17, 13x − 3y + 6z = −25,
and −2x − z = 3.

Exercise 1.2.35 Find the solution to the system of equations, 65x + 84y + 16z = 546, 81x + 105y + 20z =
682, and 84x + 110y + 21z = 713.

Exercise 1.2.36 Find the solution to the system of equations, 8x + 2y + 3z = −3, 8x + 3y + 3z = −1, and
4x + y + 3z = −9.

Exercise 1.2.37 Find the solution to the system of equations, −8x + 2y + 5z = 18, −8x + 3y + 5z = 13,
and −4x + y + 5z = 19.

Exercise 1.2.38 Find the solution to the system of equations, 3x − y − 2z = 3, y − 4z = 0, and −2x + y =
−2.

Exercise 1.2.39 Find the solution to the system of equations, −9x + 15y = 66, −11x + 18y = 79, −x + y =
4, and z = 3.
 30 30
20 y w 0
20 x z 0
10 10

Exercise 1.2.46 Find the rank of the following matrix.

⎡ ⎤
4 −16 −1 −5
⎣ 1 −4 0 −1 ⎦
1 −4 −1 −2

Exercise 1.2.47 Find the rank of the following matrix.

⎡ ⎤
3 6 5 12
⎣ 1 2 2 5 ⎦
1 2 1 2

Exercise 1.2.48 Find the rank of the following matrix.

⎡ ⎤
0 0 −1 0 3
⎢ 1 4 1 0 −8 ⎥
⎢ ⎥
⎣ 1 4 0 1 2 ⎦
48 Systems of Equations
−1 −4 0 −1 −2

Exercise 1.2.49 Find the rank of the following matrix.

⎡ ⎤
4 −4 3 −9
⎣ 1 −1 1 −2 ⎦
1 −1 0 −3

Exercise 1.2.50 Find the rank of the following matrix.

⎡ ⎤
2 0 1 0 1
⎢ 1 0 1 0 0 ⎥
⎢ ⎥
⎣ 1 0 0 1 7 ⎦
1 0 0 1 7

Exercise 1.2.51 Find the rank of the following matrix.

⎡ ⎤
4 15 29
⎢ 1 4 8 ⎥
⎢ ⎥
⎣ 1 3 5 ⎦
3 9 15

Exercise 1.2.52 Find the rank of the following matrix.

⎡ ⎤
0 0 −1 0 1
⎢ 1 2 3 −2 −18 ⎥
⎢ ⎥
⎣ 1 2 2 −1 −11 ⎦
−1 −2 −2 1 11

Exercise 1.2.53 Find the rank of the following matrix.

⎡ ⎤
1 −2 0 3 11
⎢ 1 −2 0 4 15 ⎥
⎢ ⎥
⎣ 1 −2 0 3 11 ⎦
0 0 0 0 0