GEC220 NOTE

§1.0 PARTIAL DIFFERENTIATION

1.1 FUNCTIONS OF SEVERAL VARIABLES

Functions of several variables occur frequently in Engineering. For example, if we

have a right circular cone of altitude h and radius of base r , its volume and lateral
π
V = r2h
surface area are, respectively 3
2 2
and S=πr r +h . √
Here V and S are functions of the two variables r , h . Also the law of cosine

expresses the length c of the third side of a triangle as a function of the lengths a , b
of the other two sides and of the included angle θ . The formula expressing c as a
1

function of three variables isc=( a + b −2 ab cos θ ) .

2 2 2

A function of two variables is defined as follows. Let D be a collection of number

pairs ( x , y ) , and suppose that with each pair is associated a unique number z,

thus giving us a certain collection of a number triples ( x , y , z ) . This collection is

called a function of two variables. If we denote the function by a single letter, say f ,

then we write z=f ( x , y ) and call z the value of f at ( x , y ) . Where x and y are
independent variables and z is called the dependent variable. The collection of all

the values of f is called the range of f , while the collection D of allowable pairs
( x , y ) is called the domain of f .
Functions of three or more variables are defined in a similar manner.

1.2 PARTIAL DERVATIVES

The ideas of calculus apply to functions of several variables as well as to functions of
one variable. Of course, because more variables are involved, the notation and
technical detail are more complicated but the essential ideas are the same.

For function of one variable, f ( x ) , the derivative is defined by

1
df lim
=
Δf lim
=
dx Δx →0 Δx Δx →0 Δx [
f ( x+ Δx )−f ( x )
.
]
and this is a measure of the rate of change of the value of the function f ( x ) with
respect to its variable x .
For a function of several variables, it is also useful to know how the function changes
when one, some or all of the variables change. To achieve this we define the partial
derivative of a function.

First, we consider a function f ( x , y ) of the two variables x and y . The partial

∂f
derivative, ∂ x , of f ( x , y ) with respect to x is its derivative with respect to x treating

the values of y as being constant.

∂f
=
df
Thus, ∂ x dx
[ ] y =CONSTANT
=lim
Δx→0 [ Δx ]
f ( x+ Δx , y )−f ( x , y )
.

Likewise, the partial derivative, , of f ( x , y ) with respect to y is its derivative

with respect to y treating the value of x as being constant, so that:

∂f
= [ ]
df
∂ y dy
= lim
x=CONSTANT Δy →0
[ f ( x , y +Δy )−f ( x , y )
Δy ]
The process of obtaining the partial derivatives is called partial differentiation.
Note the use of ‘curly dees’, is to distinguish between partial differentiation and
ordinary differentiation.
A concise notation is sometimes used for partial derivatives; as an alternative to the
∂f ∂f
f x= f y=
‘curly dees’, written as ∂ x and ∂y.
If we write z=f ( x , y ) then the partial derivatives may also be written as
∂ z ∂z
, or z x , z y
∂x ∂ y .

Example1:

2
 ∂f ∂f
Find and where f ( x , y ) is given by
∂x ∂y
a) 3 x 2+ 2 xy + y 3
b) ( y 2 + x ) e−xy
SOLUTION
a) f ( x , y )=3 x 2+ 2 xy + y 3
∂f
To find , we differentiate f ( x , y ) with respect to x regarding y as a constant.
∂x
Thus, we obtain
∂f ∂
= ( 3 x 2) + ∂ ( 2 xy ) + ∂ ( y3 )
∂x ∂x ∂x ∂x
d 2
¿3 ( x )+ 2 y d ( x ) +0
dx dx
¿ 6 x +2 y
Similarly,
∂f ∂
= ( 3 x 2 ) + ∂ ( 2 xy )+ ∂ ( y 3 )
∂y ∂y ∂y ∂y
d d
¿ 0+2 x ( y )+ ( y 3 )
dy dy
2
¿ 2 x+3 y

b) f ( x , y )=( y 2 + x ) e−xy
Using the product rule, let u=( y 2+ x ) and v=e− xy
Differentiating with respect to x taking y as a constant gives:
∂f ∂v ∂u
=u +v
∂x ∂x ∂x
∂ −xy
¿( y +x)
2
( e ) + ( e−xy ) ∂ ( y 2+ x )
∂x ∂x
¿ ( y 2 + x ) (− y e−xy ) + ( e−xy ) ( 1 )
¿ e− xy ( 1− y 3−xy )
Similarly
∂f ∂v ∂u
=u +v
∂y ∂y ∂y
∂ −xy
¿( y +x)
2
( e ) + ( e−xy ) ∂ ( y 2 + x )
∂y ∂y

3
 ¿ ( y 2 + x ) (−x e−xy ) + ( e−xy ) ( 2 y )
¿ e− xy ( 2 y−xy 2−x 2)

Although, partial derivatives have been introduced here in the context of function of
two variables, the concept may be readily extended to obtain the partial derivatives of
a function of as many variables as we please.
Finding partial derivatives is no more difficult than finding derivatives of functions of
one variable, with the constant multiplication, sum, product and quotient rules having
counterparts for partial derivatives.

1.3 HIGHER-ORDER PARTIAL DERIVATIVE

For function of two variables, there are four different second-order partial derivatives.

The partial derivative with respect to x

∂f
( )
∂ ∂f
of ∂ x is ∂ x ∂ x , usually abbreviated as
∂2 f
∂ x 2 or f xx . Similarly, taking two successful partial derivatives with respect to y

( )
2
∂ ∂ f = ∂ f =f
yy
gives us ∂ y ∂ y ∂ y
2
.
For mixed second-order partial derivative, one derivative is taken with respect to
each variable. If the first partial derivative is taken with respect to x , we have

∂2 f
∂ ∂f
( )
∂ y ∂ x , abbreviated as ∂ y ∂ x or ( f x ) y=f xy . If the first partial derivative is taken

∂ ∂f
( ) ∂2 f
with respect to y , we have ∂ x ∂ y , abbreviated as ∂ x ∂ y , or ( y ) x
f =f yx
.

Example 2
Find all the second-order partial derivatives of f ( x , y )=x 2 y − y 3+ ln x .
Solution

4
First we compute all the first-order partial derivatives
∂f 1
=2 xy + and
∂x x
∂f 2 2
=x −3 y
∂y
Therefore

( )
2
∂ f ∂ ∂f ∂
=f xx = ( 2 xy + x )
−1
2
=
∂x ∂ x ∂ x ∂ x
1
¿ 2 y− 2
x

( ) ( )
2
∂ f ∂ ∂f ∂ 1
= =f xy = 2 xy + =2 x
∂ y∂ x ∂ y ∂ x ∂y x

= ( )
2
∂ f ∂ ∂f ∂ (
x −3 y )=2 x
2 2
=f = yx
∂ x∂ y ∂x ∂ y ∂x

( )
2
∂ f ∂ ∂f ∂ (
x −3 y )=−6 y
2 2
= 2
=f = yy
∂ y ∂ y ∂y ∂y

Theorem 1

If f xy (x , y) and f yx ( x , y ) are continuous on an open set containing(a , b ) , then

f xy (a , b)=f yx (a , b ) .
Third -, fourth -, or even higher-order partial derivative can be taken. Theorem 1 can
be extended to show that as long as the partial derivative are all continuous in an open
set, the order (arrangement) of differentiation of higher mixed-order derivative does
not matter. Thus f xxy =f xyx =f yxx or f xyy =f yxy=f yyx as can be seen in the example 2
above.

∂3 f
With higher-order partial derivative, notation such as ∂x ∂ y ∂x becomes quite awkward
f
and so, xyx is usually used instead.

Assignment 1
∂f ∂f
1.) Find ∂ x and ∂ y when f(x,y )is
(a) Sin (x2 -3y)
(b) b) exy cos x

5
 x
(c) 2 2
x +y
1
(d) (3 x 2+ y 2+2 xy ) 2

∂f
∂f ∧∂ f
2.) a) Find ∂ x , ∂ y if f(x,y,z) =xyz2 +3xy –z
∂z
∂z ∂z
b) Show that ∂ x + ∂ y = 1 if z=ln (e x +e y ¿ )¿
3.) The ideal gas law relating pressure, temperature and volume is
CT ∂ P ∂V
P= for some constant C . Show that T =C .
V ∂ T ∂T
4.) Suppose that the concentration of some pollutant in a river as a
function of position x and time t is given by:
P (x,t) = PO (x-ct) e-μt for constants Po, c and μ.
∂p ∂p
Show that ∂ t = -c ∂ x –μP
5.) In a chemical reaction, the temperature T , entropy S, Gibbs free
energy G and enthalpy H are related by G=H −TS. Show that
∂ ( G/T ) −H
= 2 .
∂T T
6.) For f(x,y) = Cos (xy) –x3 + y4, compute fxyy , fyxy and fxyyy

1.4 TOTAL DIFFERENTIAL

Consider a function u=f ( x , y ) of two variables x and y . Then the corresponding

increment in u is given by Δu=f ( x + Δx , y + Δy )−f ( x , y ) .

This can be rewritten as:
Δu= [ f ( x+ Δx , y + Δy )−f ( x , y + Δy ) ] + [ f (x , y + Δy )−f ( x , y ) ] .

Dividing the first bracketed term by Δx and the second by Δy , gives:

f ( x + Δx , y + Δy )−f ( x , y + Δy) f (x , y +Δy )−f ( x , y )

+ Δy
Δx Δx Δy
From the definition of the partial derivative we may approximate this expression by
∂f ∂f
Δu≈ Δx+ Δy
∂x ∂y .
This could be written as

6
 ∂f ∂f
du= dx + dy
∂x ∂y .
We see that the differential du is an approximation to the change Δu in u=f ( x , y )

resulting from small changes Δx and Δy in the independent variables x and y, that
is
∂f ∂f ∂f ∂f
dx+ dy Δx+ Δy
Δu ¿ du = ∂x ∂y = ∂x ∂y
This extends to functions of as many variables as we please, provided that the partial
derivative exists. For example, for a function of three variables x , y , z defined by
u=f ( x , y , z ) we have

∂f ∂f ∂f
Δu≈du= dx+ dy + dz
∂x ∂y ∂z
∂f ∂f ∂f
= Δx+ Δy+ Δz
∂x ∂y ∂z
The differential of a function of several variables is often called a total differential,
emphasizing that it shows the variation of the function with respect to small changes
in all independent variables.
Example 3
The function z is defined by z ( x , y )=x 2 y −3 y . Find
a) the actual change in z , ∆ z and
b) by total differential, the approximate change in z , dz
When x=4, y=3 , ∆ x=−0.01 and ∆ y =0.02
Solution
2
z ( x , y )=x y −3 y
a) ∆ z=f ( x+ ∆ x , y + ∆ y )−f ( x , y )

¿ [ ( x +∆ x ) ( y+ ∆ y )−3 ( y + ∆ y ) ]−[ ( x y−3 y ) ]

2 2

¿ [ ( 4−0.01 )2 ( 3+0.02 )−3 ( 3+0.02 ) ]− [ 4 2 ( 3 ) −3 ( 3 ) ]

¿ [ ( 3.99 )2 ( 3.02 )−3 ( 3.02 ) ]− [ 48−9 ]

¿ 0.018702
b) For total differential z=f ( x , y ),
∂z ∂z
dz= dx + dy
∂x ∂y

7
 2
z=x y−3 y
∂z
=2 xy
∂x
∂z 2
=x −3
∂y
∴ dz=( 2 xy ) dx+ ( x2 −3 ) dy
¿ ( 2 ) ( 4 )( 3 ) [ −0.01 ] + ( 42−3 ) [ 0.02 ]
¿−0.24+ 0.26
¿ 0.02
Example 4
4
πP r
In a coal processing plant the flow V of slurry along a pipe is given by: v= . If r
8 ŋl
and l both increase by 5% and P and ŋ decrease by 10% and 30% respectively, using
total differential, find the change in v and hence, approximate percentage change in V.
Solution
4
πP r
v=
8 ŋl
∴ v=f (r , l , P , ŋ)

Using total differential

∂v ∂v ∂v ∂v
dv = dr + dl+ dP+ d ŋ
∂r ∂l ∂P ∂ŋ
3
∂ v 4 πP r
=
∂r 8 ŋl
4
∂ v −πP r
= 2
∂l 8 ŋl
4
∂v π r
=
∂ P 8 ŋl
4
∂ v −πP r
= 2
∂ŋ 8ŋ l
d ( r )=0.05 r
d ( l )=0.05 l
d ( P )=−0.1 P
d ( ŋ )=−0.3 ŋ
Hence, the total differential equation becomes:
3 4 4 4
4 πP r ( πP r ( πr ( πP r
dv = 0.05 r )− 2
0.05 l ) + −0.1 P )− 2 (−0.3 ŋ )
8 ŋl 8 ŋl 8 ŋl 8ŋ l
4 4 4 4
πP r πP r πP r πP r
∴ dv=4 ( 0.05 ) −0.05 −0.1 +0.3
8 ŋl 8 ŋl 8 ŋl 8 ŋl

8
 4
πP r
¿ ( 0.2−0.05−0.1+ 0.3 )
8 ŋl
4
πP r
¿ 0.35
8 ŋl
4
πP r
But v=
8 ŋl
∴ dv=0.35 v

dv
ii) Thus, % change in v ¿ x 100
v

0.35 v
¿
x 100=35
v
∴The approximate % change in v is 35 %

1.5 CHAIN RULE

The rules and results of ordinary differentiation are carried over to partial
differentiation. In particular, the composite-function rule still hold, but in a modified
form.

Consider the two-variable case where z=f ( x , y ) and x and y are themselves
function of two independent variables, s and t . Then z itself is also a function of s

and t say F (s ,t ) , and we can find its derivative using a composite-function rule that
gives the rate of change of z with respect to s and t in terms of the rate of change of
z with respect to x and y and the rate of change of x and y with respect to s and t
. Thus
∂z ∂z ∂x ∂z ∂ y ∂z ∂z ∂x ∂z ∂ y
= + = +
∂s ∂x ∂s ∂ y ∂s and ∂t ∂ x ∂ t ∂ y ∂t

Example 5:
∂T ∂T
Find and when
∂r ∂θ
3 3
T ( x , y )=x −xy + y and
x=r cos θ and y=r sin θ
Solution
T =f ( x , y )
x=f ( r ,θ )∧ y=f ( r ,θ )

9
T is not expressed directly as a function of r and θ but wish to differentiate T with
respect to r and θ .
By chain rule
∂T ∂T ∂ x ∂T ∂ y
= +
∂r ∂ x ∂ r ∂ y ∂ r
∂T 2 ∂T 2
=3 x − y and =−x +3 y and
∂x ∂y
∂x ∂y
=cos θ and =sin θ
∂r ∂r
Thus,
∂T
=( 3 x − y ) cos θ+ (−x+ 3 y ) sin θ
2 2
∂r
Substituting for x=r cos θ and y=r sin θ interms of r and θ gives:
∂T
=3 r ( cos θ+sin θ )−2 r cos θ sin θ
2 3 3
∂r
Similarly
∂T ∂T ∂ x ∂T ∂ y
= +
∂ θ ∂ x ∂θ ∂ y ∂ θ
∂x ∂y
=−r sin θ and =r cos θ
∂θ ∂θ
∂T
=( 3 x − y ) (−r sinθ ) + (−x +3 y ) r cos θ
2 2
∴
∂θ

Example 6:
dH
Find when
dt
H=sin (3 x− y )
1 2
x=2 t −3 and y= t −5 t+ 1
2
2
Solution
H=f ( x , y )
x=f ( t ) and y=f ( t )
By the chain rule, noting that x and y are functions of t only, we have
dH ∂ H dx ∂ H dy
= +
dt ∂ x dt ∂ y dt
∂H
=3 cos ( 3 x− y )
∂x

10
∂H
=−cos ( 3 x− y )
∂y
dx
=4 t
dt
dy
=t−5
dt
Substituting for the derivatives involved, we have
dH
=3 [ cos ( 3 x− y ) ] 4 t− [ cos ( 3 x− y ) ] ( t−5 )
dt
¿ ( 11t+ 5 ) cos ( 3 x− y )

¿ ( 11t+ 5 ) cos ( 112 t +5 t−10)

2

Example 7
The base radius r cm of a right – circular cone increases at 2 cm.sec -1 and its height h
cm at 3 cm.sec-1. Find the rate of increase in its volume, V, where r = 5 and h = 15.
1 2
[Note V = π r h]
3

Solution
V =f (r , h)
r =f ( t)
h=f (t)
∂V
=¿ ?
∂t
Using chain rule
dV ∂V dr ∂ V dh
= . + .
dt ∂ r dt ∂ h dt
∂V 2
=¿ πrh
∂r 3
∂V 1 2
= πr
∂h 3
dr
=2 cm/s
dt
dr
=2 cm/s
dt
dh
=3 cm/s
dt
dV
dt (2
) ( 13 π r ) ∙ (3 )
= πrh ∙ ( 2 ) +
3
2

11
 4 2
¿πrh+ π r
3
When r =5 cm and h=15 cm
3 3
dV cm cm
=125 π =392.7
dt s s

1.6 IMPLICIT DIFFERENTIATION

For a function of single variable, sometimes f is defined as a function of x , not by

giving the value of y explicitly in terms of x , but by giving an equation in x and y .
Such an equation may not determine y uniquely in terms of x , but in the situations
we commonly meet, the equation determines one or more distinct functions.

2 2
For instance, the equation x + y −16=0 determines two functions of x :

y= √16−x 2 and y=− √16−x 2

2 2
Also, the equation 5 x −6 xy +5 y =128 determine two function of x , which we find
by solving for y by the quadratic formula:

6 x± √36 x 2 −4 ( 5 ) ( 5 x 2 −128 )
y=
10
1 1
y=
5
[ 3 x +4 √ 40−x2 ] y=
5
[ 3 x−4 √ 40−x 2 ]
or and

When an equation in x and y determines y as a function of x (or as one of several

functions of x ), but when we do not have a direct explicit formula of y in terms of
x we say that y is defined implicitly as a function of x by the equation.

When y is a differentiable function of x which is defined implicitly by an equation

dy
of a suitable type, we can calculate the derivative dx directly from the equation
without solving for y explicitly. The derivative will usually be expressed in terms of
dy
both x and y rather than in terms of x alone. In finding dx we use the composite
function (chain) rule and regard y as a function of x whenever it appears.

12
For instance
d 2
( y )=2 y dy
dx dx
d 2 3
( x y )=x 2 3 y 2 dy +2 xy 3
dx dx

Example 8
∂z ∂z
Find and for the equation:
∂x ∂y
3
z +3 xz−2 y=0
Solution
∂z
To find , note that z is a function of x , although implicitly and not expressed
∂x
explicitly, thus we use function of function where z appears and when it appears as a
product with x , we also use product rule. Because we are differentiating with respect
to x , any other variable, apart from x and z (which is a function of x ), is regarded as a
constant.
Thus, we have,
2 ∂z ∂z
3z +3 x +3 z ( 1 ) −0=0
∂x ∂x
Collecting like terms and re-arranging gives:

( 3 z 2+ 3 x ) ∂ z =−3 z
∂x
∂ z −z
∴ =
∂ x z2 + x
∂z
Likewise, for ,
∂y
2 ∂z ∂z
3z +3 x −2=0
∂y ∂y
Collecting like terms and re-arranging gives:

( 3 z 2+ 3 x ) ∂ z =2
∂y
∂z 2
∴ = 2
∂ x 3(z +x)

13
Now, we desire to obtain a general formula for obtaining the differentials of implicit

function. Consider the implicit function of two variables. The relation f ( x , y , z ) may
define any one of the variables as an implicit function of the other two. Let z be the
dependent variable. Then
∂z ∂z
dz= dx+ dy
∂x ∂y -------------------------------------------- (1)

∂f ∂f ∂f
df = dx + dy+ dz=0
∂x ∂y ∂z ------------------------------------------- (2)

Substituting dz in equation (1) into equation (2) gives

( ∂∂ fx + ∂∂ fz ∂∂ zx ) dx+( ∂∂ fy + ∂∂fz ∂∂ zy ) dy=0 ----------------------------------------- (3)

Inasmuch as x and y are independent and equation (3) must hold for all values of dx

anddy , it follows that the coefficient (in parentheses) of dx and dy must individually
∂f
≠0
be zero. If ∂ z , this condition gives
df
dx Fx
∂z df
∂x = - dz = - Fz
df
dy Fy
∂z df
Similarly, ∂y = - dz = - F z

Example 9
∂z ∂z
Using the general implicit formula, find and , given that:
∂x ∂y
2 3
x y + z +sin ( xyz )=0
Solution
f ( x , y , z )=x y 2+ z3 +sin ( xyz ) =0

14
 Using general implicit formula,

∂ z −F x
=
∂ x Fz

∂ z −F y
=
∂y Fz

2
F x = y + yzcos ( xyz )

F y =2 xy+ xzcos ( xyz )

2
F z =3 z + xycos ( xyz )

2
∂ z − y + yzcos ( xyz )
∴ = 2
∂ x 3 z + xycos ( xyz )

∂ z −2 xy + xzcos ( xyz )
=
∂ y 3 z 2 + xycos ( xyz )

Note that, much like implicit differentiation with two variables, implicit
differentiation with three variables yields expression for the derivatives that depend
on all three variables.

GEC 220 ASSIGNMENT 2

1.) Find the total differential of f(x,y)
a) f(x,y) = y ex + sin x
b) f(x,y) = √ x+ y

2.) In a coal processing plant the flow V of slurry along a pipe is

given by:
4
πP r
v= . If r and l both increase by 5% and P and ŋ decrease by
8 ŋl
10% and 30% respectively, using total differential, find the change
in v and hence, approximate percentage change in V.

15
3.) The base radius r cm of a right – circular cone increases at 2
cm.sec-1 and its height h cm at 3 cm.sec-1. Find the rate of increase
1 2
in its volume, V, where r = 5 and h = 15. [Note V = 3 π r h ]
1
4.) The area Z of a triangle is given by Z= 2 XY Cosθ, where θ is the
angle between sides X and Y. If X is increasing at 0.4 cm/s, Y is
decreasing at 0.8 cm/s and θ is increasing at 0.2 radians/s, using
chain rule, find the rate of change of the area of the triangle when
X is 3 cm, Y is 4 cm and θ is π/6 radians (≡ 300)

5.)An equation for heat generated H is H = i2Rt. Using total

differential, determine the percentage error in the calculated
value of H if the error in measuring current i is +2%, the error
in measuring resistance R is -3% and the error in measuring
time t is +1%.

6.)The volume V of a liquid of viscosity coefficient η delivered

after time t when passed through a tube of length L and
4
pd t
diameter d by a pressure p is given by V = . If the errors
128 η L
in V, p and L are 1%, 2% and 3% respectively, using total
differential, determine the error in η. Assume there in no
error in t and d.
2
E
7.) The power P dissipated in a resistor is given by P= . If
R
E=200 volts and R=8 ohms , find the change in P resulting from a

drop of 5 volts in E and an increase of 0.2 ohms in R .

8.) The radius, r , of a cylindrical can is reduced by 20 % and its

height, h, increased by 80 %. Using total differential, determine
the change in volume when ( r , h )= (1 , 5 ).

9.) In a right-angled triangle, c denotes the hypotenuse, while a and b

represent its other two sides. If c is increasing at the rate of 2 cm/s
while a is decreasing at the rate of 3 cm/s . Using chain rule,

16
 calculate the rate at which b is changing when c=5 cm and a=3 cm .
Hint: use Pythagoras theorem to relate b to a and c .

−1
10.) If θ=kHL V 2 , where k is a constant, and there are possible
errors of ± 1 percent in measuring H , L∧V , find the maximum
possible error in the calculated value of θ.

cm
11.) The radius of a cylinder increases at the rate of 0.2 s while
cm
the height decreases at the rate of 0.5 s . Find the rate at which
the volume is changing at the instant when r =8 cm and h=12 cm

R R
In a balanced bridge circuit, R1= R . If R2 , R 3 , R 4 have
2 3
12.)
4

known tolerances of ± x percent , ± y percent , ± z percent respectively,

determine the maximum percentage error in R1, expressed in
terms of x , y ∧z.
13.) The deflection y at the centre of a circular plate suspended
4
kw d
at the edge and uniformly loaded is given by y= 3
, where
t
w=total load , d=diameter of plate , t=thickness∧k isa constant . Calculate

the approximate percentage change in y if w is increased by 3

percent, d is decreased by 2 ½ and t is increased by 4 percent .

14.) The coefficient of rigidity ( n ) of a wire of length ( L ) and

AL
uniform diameter ( d ) is given by n= 4 , where A is a constant. If
d
errors of ± 0.25 percent and ± 1 percent are possible in measuring L
and d respectively, determine the maximum percentage error in the
calculated value of n.

17
 n p
15.) If k /k o=( T /T o ) . 760 , show that the change in k due to small

changes of a percent in T and b percent in p is approximately ( na+ b )

percent.

16.) A rectangular box has sides of length x cm, y cm and z

cm. Sides x and z are expanding at rates of 0.3 cm/s and 0.5
cm/s respectively and side y is contracting at a rate of 0.2
cm/s. Using chain rule, determine the rate of change of
volume when x is 3 cm, y is 1.5 cm and z is 6 cm.

17.) The rate of flow of gas in a pipe, v is given by:

1 /2 −5/ 6
v=C d T
where C is a constant, d is the diameter of the pipe and T is the
thermodynamic temperature of the gas. When determining the rate
of flow experimentally, d is measured and subsequently found to
be in error by +1.4% of d , and T has an error of –1.8% of T .
(i) Using total differential, determine the error (change) in the
rate of flow, v , based on the measured values. Leave your
answer in terms of v .
(ii) Hence, determine the percentage error in the rate of flow.

18.) From the ideal gas law PV = nRT , where nR is constant,

a) Using total differential, determine an expression for small
change in pressure, dP. b) Hence, estimate the percentage change
in pressure, P, if the temperature, T , is increased by 3% and the
volume,V , is decreased by 4%.

19.) The total surface area S of a closed cone of base radius r cm

and perpendicular height h cm is given by: S=π r 2 + πr √( r 2+ h2 ). If r
and h are each increasing at the rate of 0.25 cm sec-1, find the rate
at which S is increasing at the instant when r =3 and h = 4.

4
R θ
20.) Modulus of rigidity G= , where R is the radius, θ is the
L
angle of twist and L length. Using Total Differential, determine

18
 the approximate change in G, and hence the approximate
percentage change in, G when R is increased by 2%, θ is reduced
by 5% and L is increased by 4%.

21.) Pressure P and volume V of a gas are connected by the

equation P V 1.4=k .
i) Using total differential, determine the approximate
change in k when the pressure is increased by 4% and the
volume is decreased by 1.5%, (leave your answer in terms of
k).
ii) Hence, determine the percentage error (% change) in k
.

22.) Q factor in a resonant electrical circuit is given by:

Q=
√
1 L
R C

Using total differential, find the percentage change in Q when L

increases by 4%, R decreases by 3% and C decreases by 2%.

∂z ∂z
23.) Using the general implicit formula, find and , given that:
∂x ∂y
sin ( xyz )=x +3 z + y
24.) Given eu cos V – x = 0, using the general implicit formula,
∂ u ∂ u ,∧∂ v
find ∂ x , ∂ v ∂ x

25.) Given eu sin V = y, using the general implicit formula, find

∂u ∂ u ,∧∂ v
,
∂ y ∂v ∂ y

26.) Given u2 – v2 -y = 0, using the general implicit formula, find

∂ u ∂u ∂v
,
∂v ∂ y
and ∂y

∂z ∂z
27.) Using general implicit formula, find ∂ x and ∂ y for the
function:

19
 3 2 5 2 3
x z −5 x y z=x + y

( )
2
an
28.) Given that P+ 2 ( v−nb )−nRT =0 , using the general
v
∂T ∂ P ∂V ∂T ∂P
implicit formula, find ∂ P , ∂ V and ∂T hence show that ∂ P x ∂ V
∂V
x ∂T = -1
∂z ∂z
29.) Using the general implicit formula, find
∂x
and
∂y
, given that:

x y 2+ z 3 +sin ( xyz )=0

§2.0 SERIES REPRESENTATION OF FUNCTIONS

20
2.1 POWER SERIES
Sequence is an ordered list of objects (or events). A series is the result of adding the
infinite sequence of numbers. The sum of the 1 st terms of the series are called partial
sums.
The basic mathematics involved in power series is a natural extension of the series of
∞
∑ an x n=a0 +a1 x +a2 x 2+a 3 x3 +.. .+a n x n+. ..
the type: n=0

Where a 0 , a1 , a 2 ,. . . are independent of x , is called a power series. We refer to the

constants, a k ( k=0 , 1 ,2 , .. .) as the coefficient of the series.

Convergence of power series

Power series will, in general converge for certain values of x and diverge elsewhere.

When the sum Sn of a series of n terms tends to a limit asn→ ∞ , the series is
convergent. Applying d’Alembert’s ratio test to the above series, we see that it is
n+1
lim | an+1 x |< 1
n→∞ a n x n
absolutely convergent when:
Thus the series converges if

lim | a n+1 |
|x| n→∞ a n ⊲ 1
lim | an+1 |
That is, if |x| ⊲ n→∞ an

lim |an |
Denoting n→∞ a n+1 by r, we see that the series is absolutely convergent for
−r ⊲ x ⊲ r and divergent for x ⊲−r and x ⊳ r . The limit r is called the radius of

convergent of the series.

NOTE: D’Alembert’s ratio test.

21
 ∞
lim un+1
∑ uk
Suppose we have a series of positive termsk =o , and also n→∞ un = l exists.

Then the series is convergent if l ⊲ 1 and divergent ifl ⊳ 1 . Ifl=1 , we are not able to
decide, using this test, whether the series converges or diverges.

2.2 REPRESENTATION OF FUNCTIONS USING SPECIAL POWER

SERIES
The primary reason for studying series is that we can use them to represent functions.
This opens up all kinds of possibilities for us, from approximating the values of
transcendental functions to calculating derivatives and integrals of such functions, to
studying differential equations.

Power series may be added, multiplied and divided within their common domains of
convergence (provided the denominator is nonzero within the common domain) to
give power series that are convergent, and the properties are often exploited to express
a given power series in terms of standard series band to obtain power series expansion
of complicated functions.
Four elementary power series that are of wide spread use are:
(a) Geometric series
1
1+ x = 1−x+x 2 −x 3 +. .. .. . .. .. .. . .. .. ...+(−1)n x n +. .. . .. .. . .. .. . .. .. .. ....
(−1< x <1 )
(b) The binomial Series

(1+x ) r
= 1+
(r1) x (r2) x (r1) x
+
2
+
3
+ . .. .. . .. .. +
(rn) x n
+ . .. .. . .. .. .

r
Where n
() =
r(r−1 ). . .. .. .. . .. .. . .. .. . .. .(r−n+1)
1⋅2⋅3 .. .. . .. .. . .. .. . .. .. . .. n
is the binomial coefficient.

r is any real number. When r is a positive integer, say N, the series terminates at the
N
term x . When r is not a positive integer the series does not terminate.
Setting r = -1 gives

22
 −1 1 (−1) (−1) (−2 ) 2 (−1)(−2)(−3 ) 3
(1+x ) = =1+ x+ x+ x +. .. . .. .
1+x 1 1⋅2 1⋅2⋅3
which simplifies to the geometric series
1
=1−x+x 2 −x 3 +. .. . .. .. . .. ..
1+ x
Similarly,
−2 (−2) (−2) (−3 ) 2 (−2 )(−3 )(−4 ) 3
(1+x ) =1+ x+ x+ x +. .. . .. .. . .. .. . .. .
1 1⋅2 1⋅2⋅3
which simplifies to the arithmetic-geometric series
1
2
=1−2 x +3 x 2 −4 x 3 +.. .. .. . .. .. . .. .. .
(1+x )
So the geometric series may be thought of as a special case of the binomial series.

(c) The exponential series

x x 2 x3 xn
e x=1+ + + +. .. .. .. . .. .+ +. . .. ..
1! 2 ! 3 ! n! (all x )
The number e is defined by

( )
n
lim 1+
x
e = n→∞ n
x
Similarly, the function e is defined by

( )
n
lim 1+
1
e = n→∞ n

Using the binomial expansion, we have

e=
lim
x
n→∞
1+
1 n
+
1⋅2 n{ ()
n x n(n-1) x 2 n(n-1)(n-2) x 3
+
1⋅2⋅3 n () ()
+.. .. . .. .. . .. .. . .. .. . .. }
{ ( 1- ) ( 1- )( 1- ( )
}
1 1 2
lim n n n
e x= 1
1+ X + X + 2 3
X +. .. .. . .. .. . .. .. . .. .. . .
n→∞ 1 1⋅2 1⋅2⋅3

x x2 x 3
=1+ + + +. .. . .. .. . ..
1! 2! 3!

So we see the connection between binomial and exponential series.

23
(d) The logarithmic series

x2 x3 x4 x n+1
ln (1+x )=x− + − +. .. . .. .. . .. .. .+(−1 )n +. .. . .. .. .. . .. .
2 3 4 n+1 (-1< x <1)

Example

Exercise

using Maclaurin series obtain the four elementary power series above.

2.3 TAYLOR SERIES

Taylor series is a method of constructing polynomial approximations to functions that

are not polynomials. The higher derivatives of all orders will play a pivotal role. They
are essential means for exploring and computing with transcendental functions (e.g.
sin x , cos x , ln x , e x )

∞
∑ an x n
Suppose that the power series n=0 has radius of convergencer≻0 . This means

that the series converges absolutely to some function f on the interval(−r ,r ) . We

have

∞
∑ an x n=a0 +a1 x +a2 x 2+a 3 x3 +a 4 x 4 .. .. . .. .. . .. ..
f ( x ) = n=0 for each (−r ,r )

Differentiating term by term, we get

∞
f ( x )= ∑ a n nx n−1=a1 +2 a2 x +3 a 3 x 2 +4 a4 x 3 +. .. .. .
'

n=0 , again for each x ∈(-r, r ) .

Likewise, we get

24
 ∞
f ( x )= ∑ an n(n−1)x n−2 =2 a2 +3⋅2 a3 x +4⋅3 a 4 x 2 +. .. . ..
''

n=0 and

∞
f ( x )= ∑ an n (n−1 )(n−2 )x n−3=3⋅2 a3 +4⋅3⋅2 a 4 x +.. . .. .
'' '

n=0

and so on (all valid for-r,< x< r ).

Notice that if we substitute x=0 in each of the above derivatives all the terms of
the series drop out, except one. We get

f (0)=a0 ,

'
f ( 0)=a1 ,

''
f (0)
a 2=
= 2 !a2
''
f ( 0 )=2 a2 ∴ 2!

f ' ' ' ( 0)

f ( 0)=3⋅2 a3 = 3 !a3
'' ' a 3=
∴ 3!

And so on. Observe that in general, we have

f n (0 )=n! an , solving for a n we have

f n ( 0)
a n=
n! for n=0,1,2,3,........

∞
∑ an x n
Thus, we found that if n=0 is a convergence power series with radius of
convergencer >0 , then the series converges to some function f that we can write as

∞
f n (0 ) n
∞
∑
''
an x n
∑ x f ( 0)+ f ' x+ f ( 0 ) x 2 +. .. .. . .
f ( x ) =n=0 , = n=0 n ! for x ∈(-r, +r )
(0 )
= 2!

25
Instead of starting with a series, suppose that you start with an infinitely differentiable

function, f (i.e. f can be differentiated infinitely often). Then we can construct the
series


f n (0) n

n 0 n!
x

called a TAYLOR SERIES expansion of f about x=0 .

We have constructed polynomials to approximate a given function near 0 by taking

the functions and their derivatives at x=0 . This is also called Maclaurin series only
when the expansion is about x=0 .

But we can do the same thing near any number a . Instead of expressing the
approximating polynomial in terms of powers of x , it will be convenient to write it in
terms of x−a . To find the formula for these polynomials, we will imitate for a
what we did at 0. That is for the power series

∞
∑ an (x −a )n 2 3 4
f ( x ) = n=0 =a 0 +a 1 ( x−a )+a 2 ( x−a ) +a3 (x −a ) +a 4 ( x−a ) +.. .. . .. .

for each x ∈( a−r , a+r ) . Obtaining f ( a ) and differentiating term by term, at x=a ,
we get for the function and derivatives

∞
f (a )= ∑ an ( x−a) n=a0
n=0 = 0! at x=a
∞
∑ an n( x−a)n−1
= a 1 = 1! a 1
'
f ( a) = n=0 at x=a

∞
∑ an n( n−1 )(x−a )n−2
= 2 a2 = 2 ! a 2 at x=a
''
f ( a ) =n=0

∞
∑ an n( n−1 )(n−2 )(x−a )n−3
= 3⋅2 a 3 = 3 !a3
'' '
f ( a ) = n=0 at x=a

and so on. Thus in general, we have

26
f n ( a)=n! an

solving for a n , we have

f n ( a)
a n=
n! , for n=0,1,2,........

Hence, in general if the function f has derivatives through order n at a , then the

nth-order Taylor polynomial of f at a is defined as

''
f (a ) f n (a )
pn ( x ;a )=f (a )+f ' (a )( x−a)+ ( x−a)2 +. . .. .. . ..+ (x −a )n
2! n!

A Taylor polynomial for f at a is a partial sum of the Taylor series generally defined
as

n
( a)
∞
f ( x−a )n ''
f (a) f n( a )
∑ n!
'
=f ( a )+f ( a )( x−a )+
2!
( x−a )2 +. . .. .. . ..+
n!
( x −a )n +. .. ..
n=0 .

This series is called the Taylor series at x=a associated with the function f ( x )

when a=0 , the series is also called Maclaurin series associated with f ( x )

Example:

Find the Taylor polynomial expanded about x=0 (i.e. Maclaurin series) to the fifth
order [ P5 ( x ; 0 ) ] associated with the function f ( x )=sin x .

Solution

f (x)=sin x
For Taylor series expansion:

For the 5th order Maclaurin series,

2 3 4 5
' x '' x ' '' x iv x v
f ( x ) ≅ f ( 0 ) + xf ( 0 )+ f ( 0 )+ f ( 0 ) + f ( 0 ) + f ( 0 ) … … …
2! 3! 4! 5!

f ( x )=sin x f ( 0 )=sin 0=0

fˈ ( x )=cos x fˈ ( 0 ) =cos 0=1
fˈˈ ( x )=−sin x fˈˈ ( 0 )=−sin 0=0

27
fˈˈˈ ( x )=−cos x fˈˈˈ ( 0 )=−cos 0=−1
iv iv
f ( x )=sin x f ( 0 ) =sin 0=0
v v
f ( x ) =cos x f ( 0 )=cos 0=1
Substituting
−0 2 −1 3 0 4 1 5
sin x ≅ 0+1 x+ x+ x + x + x
2! 3! 4! 5!
−0 2 −1 3 0 4 1 5
≅ 0+1 x + x+ x+ x + x
2x 1 3 x2 x1 4 x 3 x2 x1 5 x 4 x3 x 2x 1

1 1 5
≅ x− x 3+ x or
6 120

∴ sin x ≅ x−0.16667 x3 +0.00833 x 5

Compare this with sin of any number about or close to 0 (in radians)

e.g for sin 0.5, x=0.5 in radians

∴ sin 0.5 ≅ 0.5−0.16667 ( 0.5 )3 +0.00833 ( 0.5 )5

≅ 0.5−0.020834 +0.000521

≅ 0.4797

Example 2

1
Compute the fourth-order Taylor polynomial about a=1 for the function f ( x )= .
x

Solution
For the Taylor series expansion:
'
f ( x ) ≅ f ( a ) + ( x−a ) f ( a ) +f (a)} over {2!} {left (x-a right )} ^ {2} + {fˈˈˈ(a)} over {3!} {left (x-a right )}

Where a=1, then

∴ P 4 ( x ; 1 )=f ( 1 ) + ( x−1 ) f ' (1 )+ f (1)} over {2!} {left (x-1 right )} ^ {2} + {fˈˈˈ left (1 right )} over {3!}

1 −1 1
f ( x )= ≡ x f ( 1 )= =1
x 1
−2 −1 −1
fˈ ( x )=−x ≡ 2 fˈ ( 1 )= 3 =−1
x 1
−3 2 2
fˈˈ ( x )=2 x ≡ 3 fˈˈ ( 1 )= 3 =2
x 1

28
 −4 −6 ' ' ' −6
fˈˈˈ ( x )=−6 x ≡ 4
f ( 1 )= 4 =−6
x 1
iv −5 24 iv 24
f ( x )=24 x ≡ 5 f ( 1 ) = 5 =24
x 1
Substituting
2 2 6 3 24 4
f ( x ) ≅ P4 ( x ;1 )=1−( x−1 ) + ( x−1 ) − ( x−1 ) + ( x−1 )
2! 3! 4!
2 6 24
≅ P 4 ( x ; 1 )=1− ( x −1 )+ ( x −1 )2− ( x−1 )3 + ( x−1 )4
2 6 24
2 3 4
¿ 1−( x−1 )+ ( x −1 ) −( x−1 ) + ( x−1 )

1
This is valid to evaluate the value of f ( x )= for all values of x about or close to 1
x
1 1 1 1
such as , , , etc.
1.1 1.2 0.9 0.98

ASSIGNMENT
MACLAURIN & TAYLOR SERIES ASSIGNMENT
π
1. Compute the 4th order Taylor polynomial of cos x about x= . Use your
3
π
answer to evaluate cos 61 ° . Hint: 1 °= .
180
2. Use Maclaurin’s series to produce a power series for cos 4x as far as the 6th
order.
Use your answer to obtain the value of cos 0.8
3. Compute the Maclaurin series associated with the function ln ( 1+ x ) (truncate
at the 4th order), hence, use your result to estimate ln ( 1.1 )
4. Find the first three non-zero terms in the Maclaurin series of
the function e sinx .
5. Compute the Maclaurin series associated with the function ( 2x + 4 )-1
(truncate at the 3rd order), hence, use your result to estimate ( 0.2 + 4 )-1
6. Compute, up to the 4th order, the Maclaurin series associated with the function:
ln (2 + x), hence, use your result to estimate ln (2.1)
7. Compute the Taylor polynomial of the function ln x (truncate at the 4th order),
hence, use your result to estimate ln ( 1.1 )
8. i) Compute the Maclaurin series, up to the 4th order, of the function:
1
.
( 1+ x )2
1
ii) Use your answer to estimate the value of .
( 1+ 0.2 )2

29
 9. Compute the given Taylor polynomial Pn ( x ; a )
1
a) P4 ( x ;1 ) for
1+ x
b) P4 ( x ; 2 ) for e
x

c) P4 ( x ; π /4 ) for cos x
10. Compute the Maclaurin Series associated with the given functions up to 3rd
order:
a) ln 1+ x
b) e− x
1
c)
√1+ x 2
d) sin x
e) e x sin x
11. Use a Taylor polynomial of degree 4 to approximate the given numbers:
a) ln ( 1.05 )
b) √ 1.2
c) e−0.1

2.4 THE ERROR IN TAYLOR SERIES (TAYLOR THEOREM)

The Remainder Rn ( x ;a )

Definition: The remainder (error) Rn ( x ;a )

Let f be a function and let Pn ( x ;a ) be the associated n − order Taylor polynomial

th

ata . The number Rn ( x ;a ) defined by the equation

f ( x ) = Pn ( x ;a ) + Rn ( x ;a ) That is,
Rn ( x ;a ) = f ( x ) - Pn ( x ;a )

30
is called the remainder or error in using the Taylor polynomial Pn ( x ;a ) to

approximate f ( x )

If the remainder Rn ( x ;a ) does approach 0 asn→ ∞ , then the function f ( x ) is

represented by its Taylor series. The following theorem expresses Rn ( x ;a ) in terms

of a derivative (this is called “derivative” or “Lagrange” form of the remainder, it is

also known as Taylor’s theorem). This formula enables us to show that for f ( x ) =e
x

,sin x , or cos x , Rn ( x ;a ) indeed approaches 0 as n→ ∞ . It also enables us to obtain

an estimate of the error in using a Taylor polynomial to approximate a given function.

Theorem: Lagrange’s form of remainder (or Taylor’s theorem)

Assume that a function f ( x ) has continuous derivatives of orders through n+1 in an

interval that includes the numbers a and x . Let Pn ( x ;a ) be the n Taylor polynomial
th

associated with f ( x ) in powers of x−a . Then there is a number C n between a and

x such that

n+1
(C )
f n
n +1
Rn ( x ;a ) = ( n+1 )! ( x−a)

§3.0 ETREMUM PROBLEMS

In many industrial situations the role of management is to make decisions that will
lead to the most effective use of the resources available. Effective management seeks
to optimize the constituents parts of the whole operation. A wide variety of
mathematical techniques is used to solve such optimization problems, including some
based on the use of calculus, which we shall consider here.

3.1 Optimal values:

31
32
Figure 3.1: Representations of Optimal Points

A basic idea is that the optimal value of a differentiable function f ( x ) (that is, its
maximum or minimum value) generally occurs where its derivative is zero; that is,
'
where f ( x ) = 0.
As can be seen from figure 1, this is a necessary condition since at a maximum or
minimum value of the function its graph has a horizontal tangent. Figure 1 does,
however, show that these extremal values are generally only local maximum or
minimum values, corresponding to turning points on the graph. Hence, in seeking the
extremal values of a function it is also necessary to check the end points, (if any) of
the domain of the function.
'
Having determined the critical or stationary point where f ( x ) = 0, we need to be
able to determine their character or nature; that is , whether they correspond to a local

minimum, a local maximum or a point of inflection of the function f ( x ) . We can do

'
this by examining values of f ( x ) close to and on either side of the critical point. We
see that from figure 2.

33
34
35
Figure 3.2: Slopes (gradients) of extrema points

36
 '
 If the value of f ( x ) , the slope of the tangent, changes from positive to
negative as we pass from left to right through a stationary point then the later
correspond to a local maximum
'
 If the value of f ( x ) changes from negative to positive as we pass from left to
right through a stationary point then the later correspond to a local minimum.
'
 If f ( x ) does not change sign as we pass through a stationary point then the
later corresponds to a point of inflection.

An alternative approach to determining the nature of a stationary point is to

'' ''
calculate the value of the second derivative f ( x ) at the point. Note that f ( x )

determines the rate of change of f ( x ) . Suppose that f ( x ) has a stationary point

'

at x=a , so that f ( a) =0 . Then provided f ( a ) is defined, either f ( a ) ¿ 0 ,

' '' ''

f ( a )=0 or f ( a ) ¿ 0 .
'' ''

If f ( a ) ¿ 0 then f ( x ) is decreasing at x=a , and since f ( a) =0 , it follows that

'' ' '

f ( x ) ¿ 0 for values of x just greater thana . We therefore conclude that x=a

'

corresponds to a local maximum. Since the graph is concave down at a local

maximum, f ( a ) ¿ 0 .
''

Similarly, we can argue that if f ( a ) ¿ 0 , then the stationary point x=a

''

corresponds to a local minimum.

Summarizing, we have

The function f ( x ) has a local maximum at x=a provided f ( a)=0 and

'

f ( a )¿ 0
''

The function f ( x ) has a local minimum at x=a provided f ( a)=0 and

'

f ( a )¿ 0
''

''
If f ( a )=0 , we cannot assume that x=a correspond to a point of inflection, and we
''
must revert to considering the sign of x=f ( x ) on either side of the stationary point.

37
Example.
Determine the stationary points of the function
3 2
f ( x )=x −3 x +6
and using second derivative, confirm the nature of the stationary points of the
function.
Solution
The derivative is:
' 2
f x =3 x −6 x
Stationary points occur when f 'x =0 i.e,
' 2
f x =3 x −6 x=0
∴ 3 x ( x−2 )=0
3 x=0 or ( x−2 ) =0
Thus,
x=0 or x=2
The corresponding y values of the function are obtained by substituting x into y :
At x=0
3 2
f ( 0 )=0 −3 ( 0 ) +6=6 and
At x=2
3 2
f ( 2 )=2 −3 ( 2 ) +6=2
So that the stationary points of f 'x are:
( x , y ) =( 0 , 6 ) and ( 2 , 2 )
In order to investigate their nature, we use 2nd derivative test:
' 2
f x =3 x −6 x
∴ f 'x' =6 x−6
At the stationary point ( x , y ) =( 0 , 6 )
''
f (0 )=6 ( 0 ) −6=−6<0(negative)
Thus, the point ( x , y ) =( 0 , 6 ) corresponds to a local maximum
At the stationary point ( x , y ) =( 2 ,2 )
''
f (2 )=6 ( 2 )−6=6> 0( positive)
Thus, the point ( x , y ) =( 2 ,2 ) corresponds to a local minimum
The figure below shows the sketch of the function.

38
3.2 EXTREMA OF FUNCTIONS OF SEVERAL VARIABLES
Definition:

For a function of two variables, we call f (a , b ) a local maximum of f if there is

an open disk R centred at(a , b ) , for which f (a , b )≥ f ( x , y ) for all( x , y )∈ R .

Similarly, f ( a , b ) is called a local minimum of f if there is an open disk R

centred at(a , b ) , for which f ( a , b )≤f ( x , y ) for all ( x , y )∈ R . In either case

f ( a , b ) is called a local extremum of f sometimes referred to as relative

extremum of f

The idea here is the same as it was in function of single variable. That is, if
f (a , b )≥ f ( x , y ) for all ( x , y ) “near”(a , b ) , we call f ( a , b ) a local maximum.
Similarly if f (a , b) ≤ f (x , y ) for all (x , y ) near (a ,b), we call f (a , b) a local
minimum.
Much as with function of one variable, it turns out that local extrema can occur
where the first (partial) derivatives are zero or do not exist. However, having zero
first partial derivatives at a point does not by itself guarantee a local extremum.

39
Definition:

The point (a , b ) is a critical point of the function f ( x , y ) if (a , b ) is in the

∂f ∂f ∂f ∂f
(a , b) (a, b )
domain of f and either ∂ x =∂ y =0 or one or both of ∂ x and ∂ y do

not exist at(a , b ) .

If (a , b ) is a local extremum (local maximum or local minimum), then (a , b )

must be a critical point of f . Although local extrema can only occur at critical
points, every critical point need not correspond to a local extremum.

Proof:

Suppose that f ( x , y ) has a local maximum at(a , b ) . Holding y constant at y=b ,

notice that the function g( x )=f ( x , b) has a local maximum at x=a . This
∂f
' ' (a , b)
'
implies either g (a )=0 or g (a ) doesn’t exist. Note that g (a ) = ∂ x .
'
Likewise, holding x constant at x=a , observe that the function h ( y )=f (a , y )
' '
has a local maximum at y=b . It follows that h (b )=0 or h (b ) doesn’t exist.
∂f
' (a , b )
Note that h (b ) = ∂ y . Combining these two observations, we have that each
∂f ∂f
(a , b) (a , b )
of ∂ x and ∂ y equals zero or doesn’t exist. We can then conclude that
(a , b ) must be a critical point of f . A nearly identical argument shows that if
f ( x , y ) has a local minimum at (a , b ) then (a , b ) must be a critical point of f .

When looking for local extrema, you must first find all the critical points. Then
you determine the critical location is a local maximum, local minimum or neither,
since local extrema can only occur at critical points. To determine the nature of

40
the critical point of function of several variables, we use a generalisation of the
second derivative test for function of a single variable.

Theorem: (Second derivative test)

Suppose that has continuous second-order partial derivatives in some

open disk containing the point and that . We define

the discriminate for the point by:

(i) If and or , then has a local or

relative minimum at .

(ii) If and or , then has a local or

relative maximum at

(iii) If then has a saddle point at

(iv) If , then no conclusion can be drawn. The point may be a

minimum, maximum or saddle point. Further investigation is required.

When a function along some path through has a maximum value while
along other paths through the same point has a minimum value. Such a point is
called saddle point.

41
Figure3.3: Representations of saddle point

42
 Definition:

The point is a saddle point of if is a critical

point of and if every open disk centred at contains points in the

domain of for which and points in the domain of

for which .

Note that to have , we must have both and

or both and .
In summary we have:

Example
1) Locate and classify all the critical points for the function:
f ( x , y ) ¿ 2 x 2− y 3−2 xy

Solution:
f ( x , y ) ¿ 2 x 2− y 3−2 xy

43
We first compute the 1st partial derivatives and then equate them to zero and solve the
simultaneous equations to obtain the critical points:
At local extrema ,
f x =f y =0 ,
f x =4 x−2 y=0 --------I
2
f y =−3 y −2 x=0 ---------II
¿I
y=2 x ------- III
Substitute y=2 x into II
2
−3 ( 2 x ) −2 x=0
2
−12 x −2 x=0
∴−2 x ( 6 x +1 ) =0
−1
x=0∨x=
6
−1
(Note common mistake from students: x= alone does not give complete answer)
6

From III y=2 x

Therefore, the corresponding values of y are:
At x=0 , y=0
−1 −1
At x= , y=
6 3
Hence the stationary points are,
−1 1
∴ ( x , y )=( 0 ,0 )∧( ,− ) ,
6 3

To classify the local extrema, we use 2nd determinant test, D

2
D=f xx (a , b)∙ f yy (a ,b) – ( f xy (a , b) )
From I
f x =4 x−2 y
f xx =4
f xy =−2
From II
2
f y =−3 y −2 x
f yy =−6 y
2
D=(4)(−6 y )−(−2 )
¿−24 y−4

At point ( x , y )=(0 ,0)

D ( 0 , 0 )=0−4
¿−4
D=−ve .
Thus,point(x , y )=(0 , 0) is a saddle point

−1 1
At point(x , y )=( ,− )
6 3

44
D ( −16 ,− 31 )=−24 ( −13 )−4
¿4
D ( −16 ,− 31 )=+ ve .
f ( ,− )=4
−1 1
xx
6 3
f xx =+ ve .

Hence for point(x , y )= ( −16 ,− 13 )

D is+ve∧f xx is+ ve

The point(x , y )=
6 (
−1 1
)
,− is therefore a local minimum point
3

GEC 220 EXTREMUM PROBLEMS ASSIGNMENT

1. Locate and classify the stationary points of the following functions:
3 2 4 2
a) f ( x , y )=x −2 y −2 y +3 x y
3 2 3
b) f ( x , y )=2 x +6 x y −3 y −150 x
c) f ( x , y )=xy (12−4 x−3 y )
3 3
d) f ( x , y )=2 x +2 y −6 x −24 y +16
3 3
e) f ( x , y )=x + y −3 x−12 y+ 20
f) f ( x , y ) ¿ 2 x 2−4 xy + y 4 +2
3 2 2
g) f ( x , y )=x + x y− y −4 y
4 4
h) f ( x , y )=−x − y + 4 xy
2 2
i) f ( x , y )=x + 3 y −2 xy −8 x
3 3 2
j) f ( x , y )=x −12 x+ y +3 y −9 y
3 3
k) f ( x , y )=4 + x + y −3 xy
2 3 2 2
l) f ( x , y )=3 x y + y −3 x −3 y +2
3 2 2
m) f ( x , y )=x + xy −12 x − y

2) The gravitational attraction at the point (x , y ) in the (x , y ) plane due to point

masses in the plane is:
1 4 9
G ( x , y )= + +
x y 4−x − y
Show that G ( x , y ) has a maximum value of 9

3) What are the values and types of the critical points, if any, of
3 3
f (x , y )=x + y −3 xy−7 ?

3.3 Absolute (or Global) Extrema

45
Theorem:

Suppose that is continuous on the closed and bounded region .

Then has both an absolute (global) maximum and an absolute (global)

minimum on . Further the absolute extrema must occur at either a critical point
in or on the boundary of

Now, we shall explain what is meant by a bounded and closed set of points in the
-plane. A set of points (i.e, a collection of points) is said to be bounded if it is
entirely contained within some square in the plane. For example, the set of all

points such that is bounded. The set of points such

that , with no restriction on , is not bounded. A set of points is said

to be closed if contains all points such that points of can be found as

close as one pleases to . For example, the set of points such that

and is not closed, whereas the set of points ** such that ** is

closed. Again the set of points such that is bounded but not
closed.

Example

3.4 Optimization of constrained function (Lagrange multiplier)

There are frequent situations in engineering applications when we wish to obtain

the stationary values of functions of more than one variable and for which the
variables themselves are subject to one or more constraint conditions. Here we
will show the technique for solving such problems.

Example: Obtain the extremum value of the function

subject to the constraint

46
Solution
In this particular example it is easy to eliminate one of the two variables and .

Eliminating , we can write as

we can now apply the techniques used for functions of one variable to obtain the
extremum value. Differentiating gives:

and

An extremal value occurs when ; that is, , and since

, this correspond to a minimum value. Thus the extremum is a

minimum, at , .

In example above we were fortunate in being able to use the constraint equation to
eliminate one of the variables. In practise, however, it is often difficult, or even
impossible to do this, and we have to retain all the original variables.

Let us consider the general problems of obtaining the stationary points of

subject to the constraint . We shall refer to such point as conditional

stationary points.
This method of solution is a more versatile technique called the method of
Langrange multiplier, in which the introduction of a third variable (multiplier)
enables one to solve constrained optimization problems without first solving the
constraint equation for one of the variables.

More specifically, the method of Langrange multiplier uses the fact that any

relative extremum of the function subject to the constraint

must occur at a critical point of the function:

47
where is a new variable called Langrange multiplier. To find the critical point
of , compute the partial derivatives:

And solve the equations , and simultaneously as follows

or

or

or

Finally evaluate at each critical point of . Then comparing

at the critical points, if has a constrained maximum (minimum), it will

be given by the largest (smallest) of the critical values .

Proof
Although a rigorous explanation of the proof of Lagrange multiplier is beyond the
scope of this course, there is a rather simple geometric argument that is quite
convincing. This argument depends on the fact that for the level curve

, the slope at each point (using implicit function of several variable

method) is given by

Now, considering the constrained optimization problem:

Maximize subject to . Geometrically, this means you must

find the highest level curve of that intersects the constraint curve .
This critical intersection will occur at a point where the constraint curve is tangent

to a level curve; that is where the slope of the constraint curve is

equal to the slope of a level curve . This implies:

Slope of constraint curve = Slope of level curve

48
or equivalently,

If you let denote the common ratio, then

and

From which you get the first two Lagrange equations

and

The third Lagrange equation is simply a statement of the fact that the
point of tangency actually lies on the constraint curve

49