MULTI VARIABLE CALCULUS (DIFFERENTIATION)

Introduction:
We know that 𝑦 =𝑓(𝑥) is afunction where „𝑦‟ is dependent variable and „𝑥‟ is independent
variable. We are going to expand the idea of functions to include functions for more than one
independent variable. In day to day life we deal with things which depend on two or more
quantity. For example, the area of the room which is a rectangle consists of two variables: length
(say 𝑎) and breadth (say 𝑏) is given by 𝐴 =𝑎𝑏. Similarly the volume of the rectangular
parallelepiped consists of three variables 𝑎, 𝑏, 𝑐 i.e., length, breadth, height is given by 𝑉 =
𝑎 𝑏 𝑐.
In this chapter we say that 𝑧 is a function of two variables 𝑥 , 𝑦 and write 𝑧 = 𝑓(𝑥 , 𝑦) where „𝑧‟
is dependent variable and „𝑥’ & ‘𝑦’ are independent variables.
Limit of a function of two variables:
A function 𝑓(𝑥, 𝑦) is said to tend to the limit l as (𝑥, 𝑦) tends to (𝑎, 𝑏) i.e.,𝑥 → 𝑎 𝑎𝑛𝑑 𝑦
→ 𝑏 if corresponding to any given positive number ∈ ∃ a positive number 𝛿 𝑠𝑢𝑐𝑕 𝑡𝑕𝑎𝑡 𝑓 𝑥, 𝑦
− 𝑙 <∈ for all points (𝑥, 𝑦) whenever 𝑥 − 𝑎 ≤ 𝛿, 𝑦 − 𝑏 ≤ 𝛿.
In other words the variable value (𝑥 , 𝑦) approaches a finite fixed value
l when the
variable value (𝑥, 𝑦) approaches a fixed value (𝑎, 𝑏) i.e.,
lim𝑥→𝑎 𝑓 𝑥, 𝑦 = 𝑙 𝑜𝑟 lim 𝑥,𝑦 →(𝑎,𝑏) 𝑓 𝑥, 𝑦 = 𝑙
𝑦 →𝑏

Continuity of a function of two variables at a point:

A function 𝑓(𝑥, 𝑦) is continuous at a point (𝑎, 𝑏) if, corresponding to any given
positive number ∈ ∃ a positive number 𝛿 𝑠𝑢𝑐𝑕 𝑡𝑕𝑎𝑡 𝑓 𝑥, 𝑦 − 𝑓(𝑎, 𝑏) <∈ for all points
(𝑥, 𝑦) whenever 0 < (𝑥 − 𝑎)2 + (𝑦 − 𝑏)2 < 𝛿 2
Note: Every differentiable function is always continuous, but converse need not be true.
Solved Problems:
𝟐𝒙𝟐 𝒚
1. Evaluate 𝐥𝐢𝐦𝒙→𝟏
𝒚→𝟐 𝒙𝟐 +𝒚𝟐 +𝟏

2𝑥 2 𝑦 2𝑥 2 𝑦
Sol: lim𝑥→1 = lim𝑥→1 lim𝑦→2
𝑦→2 𝑥 2 +𝑦 2 +1 𝑥 2 +𝑦 2 +1

4𝑥 2
= lim𝑥→1 𝑥 2 +5
4 2
=6 =3

(or)
2𝑥 2 𝑦 2𝑥 2 𝑦
lim𝑥→1 2 2
= lim𝑦 →2 lim𝑥→1
𝑦 →2 𝑥 +𝑦 +1 𝑥 2 +𝑦 2 +1
 2𝑦
= lim𝑥→1
𝑦 2 +2

4
=6
2
=3
𝒙−𝒚
2. If f(x ,y) = 𝟐𝒙+𝒚 show that 𝐥𝐢𝐦𝒙→𝟎 𝐥𝐢𝐦𝒚→𝟎 𝒇(𝒙, 𝒚) ≠ 𝐥𝐢𝐦𝒚→𝟎 𝐥𝐢𝐦𝒙→𝟎 𝒇(𝒙, 𝒚)

𝑥−𝑦
Sol: lim𝑥→0 lim𝑦 →0 𝑓(𝑥, 𝑦) = lim𝑥→0 lim𝑦 →0 2𝑥+𝑦
𝑥
= lim𝑥→0 2𝑥
1
=2
𝑥−𝑦
lim𝑦 →0 lim𝑥→0 𝑓(𝑥, 𝑦) = lim𝑥→0 lim𝑦 →0 2𝑥+𝑦
−𝑦
= lim𝑦 →0 𝑦

= -1
Hence the result follows.
3. Discuss the continuity of the function
𝟐𝒙𝒚
, 𝒙, 𝒚 ≠ (𝟎, 𝟎)
f(x,y) = 𝒙𝟐 +𝒚𝟐
𝟎, 𝒙, 𝒚 = (𝟎, 𝟎)

Sol: Let us consider the limit of the function for testing the continuity along the line 𝑦 =
𝑚𝑥.
2𝑥𝑦
lim 𝑓 𝑥, 𝑦 = lim𝑥→𝑜
Now 𝑥→𝑜
𝑦 →𝑜 𝑥 2 +𝑦 2
𝑦 →0

2𝑚 𝑥 2
= lim𝑥→0 𝑥 2 +𝑚 2 𝑥 2
2𝑚
= 1+𝑚 2

Which is different for the different m selected.

∴ 𝑥→𝑜
lim 𝑓 𝑥, 𝑦 Does not exist.
𝑦 →0

Consider
2𝑥(0)
lim𝑥→0 𝑓 𝑥, 0 = lim𝑥→0 𝑥 2 +0 =lim𝑥→0 0 = 0 = 𝑓(0,0)

2𝑦 (0)
lim𝑦 →0 𝑓 𝑦, 0 = lim𝑦 →0 𝑦 2 +0 =lim𝑦 →0 0 = 0 = 𝑓(0,0)

∴ 𝑓(𝑥, 𝑦) is continuous for given values of x and y but it is not continuous at (0,0)
Partial Differentiation:
𝑓 𝑥+∆𝑥,𝑦 −𝑓(𝑥,𝑦)
Let 𝑧 = 𝑓(𝑥, 𝑦) be a function of two variables 𝑥 and 𝑦. Then lim𝑥→0 , if it
∆𝑥

exists, is said to be partial derivative or partial differential coefficient of 𝑧 or 𝑓(𝑥, 𝑦) , w.r.t.x .

𝜕𝑧 𝜕𝑓
It is denoted by the symbol 𝜕𝑥 or 𝜕𝑥 or 𝑓𝑥 .

i.e, for the partial derivative of 𝑧 = 𝑓 𝑥, 𝑦 w.r.t. „𝑥‟ , „𝑦‟ is kept constant.
Similarly, the partial derivative of 𝑧 = 𝑓 𝑥, 𝑦 w.r.t. „𝑦‟ , „𝑥‟ is kept constant and is defined
f x,y+∆y −f(x,y) 𝜕𝑧 𝜕𝑓
as limy→0 and is denoted by 𝜕𝑦 or 𝜕𝑦 or 𝑓𝑦 .
∆y

Higher order Partial Derivatives:

𝜕𝑓 𝜕𝑓
In general the first order partial derivatives 𝜕𝑥
and are also functions of 𝑥 and 𝑦 and they
𝜕𝑦

can be differentiated repeatedly to get higher order partial derivatives,

∂ ∂f ∂2 f ∂ ∂f ∂2 f ∂ ∂f ∂2 f ∂ ∂f ∂2 f
So ∂x ∂x
= ∂x 2 , = ∂y 2 , = ∂x ∂y , ∂y = ∂y ∂x ,
∂y ∂y ∂x ∂y ∂x

∂ ∂2 f ∂3 f ∂ ∂2 f ∂3 f ∂ ∂2 f ∂3 f
∂x 2
= , = , = , and so on.
∂x ∂x 3 ∂y ∂y 2 ∂y 3 ∂x ∂y 2 ∂x ∂y 2

The chain rule of Partial Differentiation:

Let 𝑧 = 𝑓 𝑢, 𝑣 where 𝑢 = ∅ x, y and 𝑣 = 𝑔(𝑥, 𝑦) . Then
∂z ∂z ∂u ∂z ∂v ∂z ∂z ∂u ∂z ∂v
= + and = +
∂x ∂u ∂x ∂v ∂x ∂y ∂u ∂y ∂v ∂y

Total differential coefficient:

Let z = f x, y where 𝑥 = ∅ t and 𝑦 = 𝑔(𝑡)
Substituting 𝑥 and 𝑦 in 𝑧 = 𝑓(𝑥, 𝑦) , 𝑧 becomes a function of a single variable 𝑡.
𝜕𝑧
Then the derivative of 𝑧 w.r.t. „𝑡‟ i.e, is called the total differential coefficient or total
𝜕𝑡

derivative of 𝑧.
∂z ∂z ∂x ∂z ∂y
∴ = +
∂t ∂x ∂t ∂y ∂t
∂u ∂u
Note: In the differential form, this result can be written as 𝑑𝑢 = . dx + . dy
∂x ∂y
Here, du is called the total differential of 𝑢.

Solved Problems:
𝛛 𝛛 𝛛 𝟐 −𝟗
1. If U = log ( 𝐱 𝟑 + 𝐲 𝟑 + 𝐳 𝟑 − 𝟑𝐱𝐲𝐳) , prove that + 𝛛𝐲 + 𝛛𝐳 𝐔=
𝛛𝐱 𝐱+𝐲+𝐳 𝟐

Sol: Given that U = log( x 3 + y 3 + z 3 − 3xyz)

𝜕𝑈 3𝑥 2 −3𝑦𝑧
∴ 𝜕𝑥
= x 3 +y 3 +z 3 −3xyz
( 𝑦 and 𝑧 are constant)
 𝜕𝑈 3𝑦 2 −3𝑥𝑧
= ( 𝑥 and 𝑧 are constant)
𝜕𝑦 x 3 +y 3 +z 3 −3xyz

𝜕𝑈 3𝑧 2 −3𝑦𝑥
= ( 𝑦 and 𝑥 are constant)
𝜕𝑧 x 3 +y 3 +z 3 −3xyz

𝜕𝑈 𝜕𝑈 𝜕𝑈 3𝑥 2 −3𝑦𝑧 3𝑦 2 −3𝑥𝑧 3𝑧 2 −3𝑦𝑥

∴ + 𝜕𝑦 + 𝜕𝑧 = + +
𝜕𝑥 x 3 +y 3 +z 3 −3xyz x 3 +y 3 +z 3 −3xyz x 3 +y 3 +z 3 −3xyz

3(x 2 + y 2 + z 2 − xy − yz − zx)
=
(x + y + z)(x 2 + y 2 + z 2 − xy − yz − zx)
∂U ∂U ∂U 3
⟹ + + = x+y+z ……..(1)
∂x ∂y ∂z

∂ ∂ ∂ 2 ∂ ∂ ∂ ∂U ∂U ∂U
Now + ∂y + ∂z U = + ∂y + ∂z + ∂y +
∂x ∂x ∂x ∂z
∂ ∂ ∂ 3
= + ∂y + ∂z [from 1 ]
∂x x+y+z
∂ 3 ∂ 3 ∂ 3
= ∂x + ∂y +∂z
x+y+z x+y+z x+y+z
3 3 3
=- - −
x+y+z 2 x+y+z 2 x+y+z 2
9
=-
x+y+z 2

𝛛𝟐 𝐳
2. If 𝐱 𝐱 𝐲 𝐲 𝐳 𝐳 = e show that at 𝒙 = 𝒚 = 𝒛 , 𝛛𝐱𝛛𝐲 = − 𝐱𝐥𝐨𝐠 𝐞𝐱 −𝟏

Sol: Given that x x y y z z = 𝑒

Taking logarithm on both sides, we get
𝑥 𝑙𝑜𝑔 𝑥 + 𝑦 𝑙𝑜𝑔 𝑦 + 𝑧 𝑙𝑜𝑔 𝑧 = 𝑙𝑜𝑔 𝑒
⟹ 𝑧 𝑙𝑜𝑔 𝑧 = 1 − 𝑥 𝑙𝑜𝑔 𝑥 – 𝑦 𝑙𝑜𝑔 𝑦
Differentiating partially w. r. t , ′x ′ , we get
1 ∂z 1
z. z + 1. log z = - x. x + 1. log x
∂x
∂z (1+log x)
⇒ ∂x = - (1+log z) ……..(1)
∂z (1+log y)
Similarly ∂y = - (1+log z) ……..(2)

When 𝑥 = 𝑦 = 𝑧 , we have
𝜕𝑧 𝜕𝑧
= -1 and 𝜕𝑦 = -1
𝜕𝑥

Now differentiating (2) partially w.r.t, „x‟, we get

∂2 z ∂ ∂z ∂ (1+log y)
∂x ∂y
= ∂x = ∂x −
∂y (1+log z)

−2 1 ∂z 1+log y ∂z
= - (1+ log y) − 1 + log z =z …….(3)
z ∂x 1+log z 2 ∂x

When 𝑥 = 𝑦 = 𝑧 from (3) , we have

 ∂2 z 1+log x ∂z
=x (-1) since = −1
∂x ∂y 1+log x 2 ∂x
1 1
= - x(1+log x) = -x(log e+log x) (since 𝑙𝑜𝑔 𝑒 = 1)
1 −1
= - xlog = − x log ex
ex
𝝏𝒖 𝝏𝒖 𝝏𝒖
3. If 𝒖 = 𝒇( 𝒚 − 𝒛 , 𝒛 − 𝒙, 𝒙 − 𝒚) prove that 𝝏𝒙 + 𝝏𝒚 + 𝝏𝒛 = 𝟎

Sol: Let 𝑟 = 𝑦 − 𝑧 , 𝑠 = 𝑧 − 𝑥 , 𝑡 = 𝑥 – 𝑦 .Then 𝑢 = 𝑓 (𝑟, 𝑠, 𝑡)

∂r ∂r ∂r
Now = 0 , ∂y = 1, ∂z = −1
∂x
𝜕𝑠 𝜕𝑠 𝜕𝑠
= −1 , = 0, =1
𝜕𝑥 𝜕𝑥 𝜕𝑥
∂t ∂t ∂t
and = 1, ∂y = −1 , ∂z = 0
∂x

∴ By chain rule of partial differentiation , we have

𝜕𝑢 𝜕𝑢 𝜕𝑟 𝜕𝑢 𝜕𝑠 𝜕𝑢 𝜕𝑡
= + +
𝜕𝑥 𝜕𝑟 𝜕𝑥 𝜕𝑠 𝜕𝑥 𝜕𝑡 𝜕𝑥
𝜕𝑢 𝜕𝑢 𝜕𝑢 𝜕𝑢 𝜕𝑢
= 0 + −1 + (1) = − 𝜕𝑠 + 𝜕𝑡 …(1)
𝜕𝑟 𝜕𝑠 𝜕𝑡

𝜕𝑢 𝜕𝑢 𝜕𝑟 𝜕𝑢 𝜕𝑠 𝜕𝑢 𝜕𝑡
= + +
𝜕𝑦 𝜕𝑟 𝜕𝑦 𝜕𝑠 𝜕𝑦 𝜕𝑡 𝜕𝑦
𝜕𝑢 𝜕𝑢 𝜕𝑢 𝜕𝑢 𝜕𝑢
= 1 + −0 + (−1) = 𝜕𝑟 − 𝜕𝑡 …(2)
𝜕𝑟 𝜕𝑠 𝜕𝑡

and
∂u ∂u ∂r ∂u ∂s ∂u ∂t
= + +
∂z ∂r ∂z ∂s ∂z ∂t ∂z
∂u ∂u ∂u ∂u ∂u
= −1 + 1 + (0) = − ∂r + ∂s …(3)
∂r ∂s ∂t

(1) + (2) + (3) gives

∂u ∂u ∂u ∂u ∂u ∂u ∂u ∂u ∂u
+ ∂y + ∂z = − ∂s + + − + − ∂r + ∂s =0
∂x ∂t ∂r ∂t

Jacobian :
Let 𝑢 = 𝑢 (𝑥 , 𝑦) , 𝑣 = 𝑣(𝑥 , 𝑦) are two functions of the independent variables 𝑥 , 𝑦.
The jacobian of ( 𝑢 , 𝑣 ) w .r .t (𝑥 , 𝑦 ) or the jacobian transformation is given by the
∂u ∂u
∂x ∂y 𝑢𝑥 𝑢𝑦
determinant ∂v ∂v
(or) 𝑣 𝑣𝑦
𝑥
∂x ∂y
𝑢 ,𝑣 𝜕(𝑢,𝑣)
The determinant value is denoted by 𝐽 or 𝜕(𝑥,𝑦)
𝑥,𝑦

Similarly if 𝑢 = 𝑢(𝑥, 𝑦, 𝑧 ) , 𝑣 = 𝑣 (𝑥, 𝑦 , 𝑧) , 𝑤 = 𝑤(𝑥, 𝑦 , 𝑧) ,then the Jacobian of 𝑢 , 𝑣 , 𝑤

w.r.to 𝑥 , 𝑦 , 𝑧 is given by
 𝑢𝑥 𝑢𝑦 𝑢𝑧
𝑢,𝑣,𝑤 𝜕(𝑢,𝑣,𝑤)
𝐽 = = 𝑣𝑥 𝑣𝑦 𝑣𝑧
𝑥,𝑦,𝑧 𝜕(𝑥,𝑦,𝑧)
𝑤𝑥 𝑤𝑦 𝑤𝑧
Properties of Jacobians
(u, v) ( x, y)
1. If J  and J 1  then JJ 1  1
( x, y) (u, v)
𝜕(𝑢,𝑣) 𝜕(𝑢,𝑣) 𝜕(𝑟,𝑠)
2. If 𝑢, 𝑣 are functions of 𝑟, 𝑠 and 𝑟, 𝑠 are functions of 𝑥 , 𝑦 , then 𝜕 = .𝜕
𝑥,𝑦 𝜕 𝑟,𝑠 𝑥,𝑦

Solved Problems:
𝜕(𝑢,𝑣,𝑤 )
1. If 𝒙 + 𝒚𝟐 = 𝒖 , 𝒚 + 𝒛𝟐 = 𝒗 , 𝒛 + 𝒙𝟐 = 𝒘 find 𝜕(𝑥,𝑦,𝑧)

Sol : Given 𝒙 + 𝒚𝟐 = 𝒖 , 𝒚 + 𝒛𝟐 = 𝒗 , 𝒛 + 𝒙𝟐 = 𝒘

We have = =

= 1(1-0) – 2y(0 – 4xz) + 0

= 1 – 2y(-4xz)
= 1 + 8xyz

 = =

2. Show that the functions u = x + y + z , v = x2 + y2 + z2 -2xy – 2yz -2xz and w = x3

+ y3 + z3 -3xyz are functionally related.
Sol: Given u = x + y + z
v = x2 + y2 + z2 -2xy – 2yz -2xz
w = x3 + y3 + z3 -3xyz
we have

=6

c1  c1  c2
c2  c2  c3
 0 0 1
6 2x  2 y 2 y  2z z yx
x  yz  y 2  xz
2
y  xz  z 2  xy
2
z 2  xy

=6[2(x - y) (y2 + xy – xz -z2 )-2(y - z)(x2 + xz – yz - y2)]

=6[2(x - y)( y – z)(x + y + z) – 2(y – z)(x – y)(x + y + z)]
=0
Hence there is a relation between u,v,w.

3. If x + y + z = u , y + z = uv , z = uvw then evaluate

Sol: 𝑥 + 𝑦 + 𝑧 = 𝑢
𝑦 + 𝑧 = 𝑢𝑣
𝑧 = 𝑢𝑣𝑤
𝑦 = 𝑢𝑣 – 𝑢𝑣𝑤 = 𝑢𝑣 (1 – 𝑤)
𝑥 = 𝑢 – 𝑢𝑣 = 𝑢 (1 – 𝑣)

R2  R2  R3

= uv [ u –uv +uv]
= u2 v

4. If u = x2 – y2 , v =2xy where x = r cos , y = r sin S.T = 4r3

Sol: Given u = x2 – y2 , v = 2xy

=r2cos2 – r2sin2 = 2rcos r sin
= r2 (cos2 – sin2 = r2 sin2
= r2 cos2

= =

= (2r)(2r)

= 4r2 [rcos22 + r sin22 ]

 =4r2(r)[ cos22 + sin22 ]
=4r3

5. If u = ,v= ,w= find

Sol: Given u = ,v= ,w=

We have

𝑢𝑥 = 𝑦𝑧(−1/𝑥2) = , 𝑢𝑦 = , 𝑢𝑧 =

= , 𝑥𝑧(−1/𝑦2) = ,

= , = , = xy (-1/z2) =

= . .

= 1[-1(1-1) -1(-1-1) + (1+1) ]

= 0 -1(-2) + (2)
=2 + 2=4

6.𝑰𝒇 𝒙 = 𝒆r 𝒔𝒆𝒄 , 𝒚 = 𝒆r 𝒕𝒂𝒏 𝑷. 𝑻 . = 𝟏

Sol: Given x = er sec , y = er tan

= , =

= er sec =x , = ersec tan

= er tan =y , = er sec2
𝑥2 – 𝑦2 = 𝑒2r (𝑠𝑒𝑐2 − 𝑡𝑎𝑛2 )
 2𝑟 = 𝑙𝑜𝑔 (𝑥2 – 𝑦2 )
 𝑟 = ½ 𝑙𝑜𝑔 (𝑥2 – 𝑦2 )
 1 1 x
rx  ( 2 x)  2
2x y
2 2
(x  y2 )
1 1 y
ry  (2 y)  2
2x y
2 2
(x  y2 )

= = =

 = , = sin-1( )

1  1  y
x  y  2  
y  x  x x2  y2
2
1 2
x

= (1/𝑥) =

= = e2r sec2 - y er sec tan

= e2r sec [sec2 - tan2 ] = e2r sec

x y
 r ,   (x  y2 ) (x  y2 )
2 2
 y
 ( x, y ) 1
x x2  y2 x2  y2

=[ - ]

= = =

. =1
𝝏(𝒙,𝒚) 𝝏(𝒓,𝜽) 𝝏(𝒙,𝒚) 𝝏(𝒓,𝜽)
7.If 𝒙 = 𝒓𝒄𝒐𝒔𝜽, 𝒚 = 𝒓𝒔𝒊𝒏𝜽 then find 𝝏(𝒓,𝜽) and 𝝏(𝒙,𝒚) .Also Show that 𝝏(𝒓,𝜽) . 𝝏(𝒙,𝒚) = 𝟏.

Sol : Given that 𝒙 = 𝒓𝒄𝒐𝒔𝜽, 𝒚 = 𝒓𝒔𝒊𝒏𝜽 -------(1)

𝑦
we have 𝑟 2 = 𝑥 2 + 𝑦 2 ; 𝜃 = 𝑡𝑎𝑛−1 -------(2)
𝑥
𝜕𝑥 𝜕𝑦 𝜕𝑥 𝜕𝑦
= 𝑐𝑜𝑠𝜃, 𝜕𝑟 = 𝑠𝑖𝑛𝜃, 𝜕𝜃 = −𝑟𝑠𝑖𝑛𝜃, 𝜕𝜃 = 𝑟𝑐𝑜𝑠𝜃
𝜕𝑟

𝜕𝜃 1 −𝑦 −𝑦 −𝑠𝑖𝑛𝜃
= 𝑦 2
. = 𝑥 2 +𝑦 2 =
𝜕𝑥 1+ 𝑥2 𝑟
𝑥

𝜕𝜃 1 1 𝑥 𝑐𝑜𝑠𝜃
= 2 . = 2 2
=
𝜕𝑦 𝑦 𝑥 𝑥 +𝑦 𝑟
1+ 𝑥
 𝜕𝑟 𝑥 𝜕𝑟 𝑦 𝜕𝑥
= = 𝑐𝑜𝑠𝜃, = = 𝑠𝑖𝑛𝜃,
𝜕𝑥 𝑟 𝜕𝑦 𝑟 𝜕𝜃
𝜕𝑥 𝜕𝑥
𝜕(𝑥, 𝑦) 𝜕𝜃 = 𝑐𝑜𝑠𝜃 −𝑟𝑠𝑖𝑛𝜃 = 𝑟
= 𝜕𝑟
𝜕 𝑟, 𝜃 𝜕𝑦 𝜕𝑦 𝑠𝑖𝑛𝜃 𝑟𝑐𝑜𝑠𝜃
𝜕𝑟 𝜕𝜃
𝜕𝑟 𝜕𝑟
𝜕(𝑟, 𝜃) 𝜕𝑥 𝜕𝑦 𝑐𝑜𝑠𝜃 𝑠𝑖𝑛𝜃 1
= = − 𝑠𝑖𝑛𝜃 𝑐𝑜𝑠𝜃 =
𝜕 𝑥, 𝑦 𝜕𝜃 𝜕𝜃 𝑟
𝑟 𝑟
𝜕𝑥 𝜕𝑦
𝜕(𝑥,𝑦) 𝜕 (𝑟,𝜃) 1
. = 𝑟. = 1.
𝜕(𝑟,𝜃) 𝜕(𝑥,𝑦) 𝑟

𝒖𝟐 𝒗𝟐 𝝏(𝒖,𝒗)
8.f 𝒙 = 𝒗
,𝒚 = 𝒖
then find 𝝏(𝒙,𝒚).

𝒖𝟐 𝒗𝟐
Sol: Given 𝒙 = ,𝒚 =
𝒗 𝒖

𝜕𝑥 2𝑢 𝜕𝑥 𝑢2
∴ = , =− 2
𝜕𝑢 𝑣 𝜕𝑣 𝑣

𝜕𝑦 𝑣 2 𝜕𝑦 2𝑣
=− 2, =
𝜕𝑢 𝑢 𝜕𝑣 𝑢
𝜕𝑥 𝜕𝑥 2𝑢 𝑢2
𝝏 𝒙, 𝒚 − 2 2 2
= 𝜕𝑢 𝜕𝑣 = 𝑣 𝑣 = 2𝑢 . 2𝑣 − 𝑢 𝑣 = 4 − 1 = 3
𝝏 𝒖, 𝒗 𝜕𝑦 𝜕𝑥 𝑣2 2𝑣 𝑣 𝑢 𝑣 2 𝑢2
− 2
𝜕𝑢 𝜕𝑣 𝑢 𝑢
𝝏 𝒖,𝒗 𝟏
Hence 𝝏 𝒙,𝒚 = 𝝏(𝒙,𝒚) =𝟏 𝟑
𝝏(𝒖,𝒗)

Functional Dependence:
Two functions u and v are functionally dependent if their Jacobian i.e,

J( ) = = =0

If the Jacobian of 𝑢, 𝑣 is not equal to zero then those functions 𝑢, 𝑣 are functionally
independent.
Solved Problems :
𝐱+𝐲 𝝏(𝒖,𝒗)
1. If 𝐮 = , 𝐯 = 𝐭𝐚𝐧−𝟏 𝐱 + 𝐭𝐚𝐧−𝟏 𝐲 . Find . Hence prove that 𝒖 and 𝒗 are
𝟏−𝐱𝐲 𝝏(𝒙,𝒚)

functionally dependent. Find the relation between them.

𝑥 +𝑦
Sol : Given u = 1−𝑥𝑦 and v = tan−1 x + tan−1 y
 ∂u 1+y 2 ∂u 1+x 2 ∂v 1 ∂v 1
∴ = (1−xy )2 , ∂y = (1−xy )2 , ∂x = 1+x 2 and ∂y = 1+y 2
∂x

∂u ∂u 1+y 2 1+x 2
∂(u,v) ∂x ∂y (1−xy )2 (1−xy )2 1 1
∴ = ∂v ∂v
= = (1−xy )2 - =0
∂ x,y 1 1 (1−xy )2
∂x ∂y 1+x 2 1+y 2

∴ 𝑢 and 𝑣 are functionally dependent .

x+y
Now v = tan−1 x + tan−1 y = tan−1 ( 1−xy ) = tan−1 u

∴ v = tan−1 u is the functional relation between 𝑢 and 𝑣.

2. Determine whether the following functions are functionally dependent or not. If
they are functionally dependent , find a relation between them.
𝒙 𝒙+𝒚
i) 𝒖 = 𝐞𝐱 𝐬𝐢𝐧 𝐲 , 𝒗 = 𝐞𝐱 𝐜𝐨𝐬 𝐲 ii) 𝒖 = 𝒚 , 𝒗 = 𝒙−𝒚
∂(u,v) ux uy ex sin y ex cos y
Sol: i) Jacobian = ∂ x,y = v vy = x
x e cos y − ex sin y
= ex ( -sin2 y − cos 2 y ) = −𝑒 𝑥 ≠ 0
∴ 𝑢, 𝑣 are functionally independent .
𝑥 𝑥+𝑦
ii) 𝑢 = 𝑦 , 𝑣 = 𝑥−𝑦
1 −x
u,v ∂(u,v) y y2 2x 2x
∴ J(x,y ) = ∂(x,y) = −2y 2x = y(x−y)2 - y(x−y)2 = 0
(x−y)2 (x−y)2
∴ u and v are functionally dependent ,
x
x+y y ( +1) u+1
y
Now v = x−y
= x = u−1
y ( −1)
y
u+1
∴v= is the functional relation between 𝑢 and 𝑣 .
u−1

3. Show that the functions 𝒖 = 𝒙𝒚 + 𝒚𝒛 + 𝒛𝒙 , 𝒗 = 𝒙𝟐 + 𝒚𝟐 + 𝒛𝟐 𝒂𝒏𝒅 𝒘 = 𝒙 +

𝒚 + 𝒛 are functionally related .find the relation between them.
Sol: We have
u = xy + yz + zx , v = x 2 + y 2 + z 2 , w= x + y + z
∂u ∂u ∂u
∂x ∂y ∂z
y+z z+x x+y
∂(u,v,w) ∂v ∂v ∂v
∴ = ∂x ∂y ∂z
= 2x 2y 2z
∂(x,y,z)
∂w ∂w ∂w 1 1 1
∂x ∂y ∂z

𝑦+𝑧
𝑧+𝑥 𝑥+𝑦
= 2 𝑥
𝑦 𝑧 ( Applying 𝑅1 → 𝑅1 + 𝑅2 )
1
1 1
1 1 1
= 2(𝑥 + 𝑦 + 𝑧) 𝑥 𝑦 𝑧
1 1 1
= 2(𝑥 + 𝑦 + 𝑧) (0) ( since R1 andR1 are identical )
= 0
 Hence 𝑢, 𝑣 and 𝑤 are functionally dependent . That is , the functional relationship
exists between them.
Now w 2 = (x + y + z )2 = x 2 + y 2 + z 2 + 2 xy + yz + zx = v + 2u
∴ w 2 = v + 2u is the functional relation between u ,v and w.
4. Verify if 𝒖 = 𝟐𝒙 – 𝒚 + 𝟑𝒛 , 𝒗 = 𝟐𝒙 – 𝒚 – 𝒛 , 𝒘 = 𝟐𝒙 – 𝒚 + 𝒛 are functionally
dependent and if so , find the relation between them.
Sol: Given 𝑢 = 2𝑥 – 𝑦 + 3𝑧 , 𝑣 = 2𝑥 – 𝑦 – 𝑧 , 𝑤 = 2𝑥 – 𝑦 + 𝑧
𝑢,𝑣,𝑤
The functions u,v,w are functionally dependent if and only if J( )=0
𝑥𝑦,𝑧
∂u ∂u ∂u
∂x ∂y ∂z
∂v ∂v ∂v
2 −1 3 1 1 3
u,v,w
Now J( )= ∂x ∂y ∂z
= 2 −1 −1 = 2(-1) 1 1 −1 = (-2)(0) =0
xy ,z
∂w ∂w ∂w 2 −1 1 1 1 1
∂x ∂y ∂z

∴ u, v, w are functionally dependent

Now 𝑢 + 𝑣 − 2𝑤 = (2𝑥 − 𝑦 + 3𝑧) + (2𝑥 – 𝑦 – 𝑧) – 2 (2𝑥 – 𝑦 + 𝑧 )
= (4𝑥 – 2𝑦 + 2𝑧) – (4𝑥 – 2𝑦 + 2𝑧) = 0
Hence 𝑢 + 𝑣 − 2𝑤 = 0 is the functional relationship between 𝑢, 𝑣 and 𝑤.
5. Show that the functions 𝒖 = 𝒙 + 𝒚 + 𝒛, 𝒗 = 𝒙2+𝒚2+𝒛2−𝟐𝒙𝒚 − 𝟐𝒚𝒛 − 𝟐𝒛𝒙 and
𝒘 = 𝒙3+𝒚3+𝒛3−𝟑𝒙𝒚𝒛 are functionally related.

Sol: Given 𝒖 = 𝒙 + 𝒚 + 𝒛, 𝒗 = 𝒙2+𝒚2+𝒛2−𝟐𝒙𝒚 − 𝟐𝒚𝒛 − 𝟐𝒛𝒙 and 𝒘 =

𝒙3+𝒚3+𝒛3−𝟑𝒙𝒚𝒛
𝜕𝑢 𝜕𝑢 𝜕𝑢
𝜕𝑥 𝜕𝑦 𝜕𝑧 1 1 1
𝜕(𝑢 ,𝑣,𝑤) 𝜕𝑣 𝜕𝑣 𝜕𝑣
Now = 𝜕𝑥 𝜕𝑦 𝜕𝑧
= 2(𝑥 − 𝑦 − 𝑧) 2(𝑦 − 𝑥 − 𝑧) 2(𝑧 − 𝑦 − 𝑥)
𝜕 (𝑥,𝑦,𝑧)
𝜕𝑤 𝜕𝑤 𝜕𝑤 3(𝑥 2 − 𝑦𝑧) 3(𝑦 2 − 𝑥𝑧) 3(𝑧 2 − 𝑥𝑦)
𝜕𝑥 𝜕𝑦 𝜕𝑧

1 1 1
=6 (𝑥 − 𝑦 − 𝑧) (𝑦 − 𝑥 − 𝑧) (𝑧 − 𝑦 − 𝑥)
2
(𝑥 − 𝑦𝑧) (𝑦 2 − 𝑥𝑧) (𝑧 2 − 𝑥𝑦)
= 6
0 0 1
2 𝑥−𝑦 2 𝑦−𝑧 𝑧−𝑦−𝑥
𝑥−𝑦 𝑥+𝑦+𝑧 𝑦−𝑧 𝑥+𝑦+𝑧 𝑧 2 − 𝑥𝑦
C1 → 𝐶1 − 𝐶2 and C2 → 𝐶2 − 𝐶3
𝜕(𝑢,𝑣,𝑤) 𝑥−𝑦 𝑦−𝑧
∴ 𝜕(𝑥,𝑦,𝑧) = 12 𝑥 − 𝑦 (𝑥 + 𝑦 + 𝑧) 𝑦 − 𝑧 (𝑥 + 𝑦 + 𝑧)
 1 1
= 12 (𝑥 − 𝑦)(𝑦 − 𝑧)
(𝑥 + 𝑦 + 𝑧) (𝑥 + 𝑦 + 𝑧)
= 12 (𝑥 − 𝑦)(𝑦 − 𝑧) (0) [C1 and C2 are identical]
= 0
Hence the functional relationship exists between𝑢, 𝑣, 𝑤.
𝒙𝟐 − 𝒚𝟐 𝟐𝒙𝒚
6. Prove that 𝒖 = , 𝒗 = are functionally dependent and find the
𝒙𝟐 + 𝒚𝟐 𝒙𝟐 +𝒚𝟐

relation between them.

𝑥2− 𝑦2 2𝑥𝑦
Sol: We are given u = 𝑥 2 + 𝑦 2 , v = 𝑥 2 +𝑦 2

𝜕𝑢 𝑥 2 +𝑦 2 .2𝑥− 𝑥 2 − 𝑦 2 .2𝑥 2𝑥(𝑥 2 +𝑦 2 −𝑥 2 +𝑦 2 ) 4𝑥𝑦 2

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

𝜕𝑢 𝑥 2 +𝑦 2 .(−2𝑦)− 𝑥 2 − 𝑦 2 .2𝑦 (−2𝑦)(𝑥 2 +𝑦 2 −𝑥 2 +𝑦 2 ) −4𝑦𝑥 2

= = = (𝑥 2 +𝑦 2 )2
𝜕𝑦 (𝑥 2 + 𝑦 2 )2 (𝑥 2 +𝑦 2 )2

𝜕𝑣 𝑥 2 +𝑦 2 .1−𝑥.2𝑥 2𝑦(𝑦 2 −𝑥 2 )
=2 y = and
𝜕𝑥 (𝑥 2 + 𝑦 2 )2 (𝑥 2 + 𝑦 2 )2

𝜕𝑣 𝑥 2 +𝑦 2 .1−𝑦.2𝑦 2𝑥(𝑥 2 −𝑦 2 )
=2 x =
𝜕𝑦 (𝑥 2 + 𝑦 2 )2 (𝑥 2 + 𝑦 2 )2

𝜕𝑢 𝜕𝑢 4𝑥𝑦 2 −4𝑥𝑦 2
𝜕(𝑢,𝑣) 𝜕𝑥 𝜕𝑦 (𝑥 2 +𝑦 2 )2 (𝑥 2 +𝑦 2 )2
Thus 𝜕(𝑥,𝑦) = 𝜕𝑣 𝜕𝑣
= 2𝑦 (𝑦 2 −𝑥 2 ) 2𝑥(𝑥 2 −𝑦 2 )
𝜕𝑥 𝜕𝑦 (𝑥 2 + 𝑦 2 )2 (𝑥 2 + 𝑦 2 )2

8𝑥 2 𝑦 2 (𝑥 2 −𝑦 2 ) 8𝑥 2 𝑦 2 (𝑦 2 −𝑥 2 )
= +
(𝑥 2 +𝑦 2 )4 (𝑥 2 +𝑦 2 )4

8𝑥 2 𝑦 2 𝑥 2 −𝑦 2 − 8𝑥 2 𝑦 2 (𝑦 2 −𝑥 2 )
= =0
(𝑥 2 +𝑦 2 )4

∴ 𝑢, 𝑣 are functionally dependent.

(𝑥 2 −𝑦 2 ) 4𝑥 2 𝑦 2 (𝑥 2 +𝑦 2 )2
𝑢2 + 𝑣 2 = (𝑥 2 + 𝑦 2 )2 + (𝑥 2 +𝑦 2 )2 =(𝑥 2 +𝑦 2 )2 = 1

Hence 𝑢 2 + 𝑣 2 = 1 is the functional relation between 𝑢 and 𝑣.

Maxima & Minima for functions of two Variables:
Definition : Let 𝑓(𝑥, 𝑦) be a function of two variables 𝑥 and 𝑦.
At 𝑥 = 𝑎 ; 𝑦 = 𝑏 , 𝑓(𝑥 , 𝑦) is said to have maximum or minimum value , if 𝑓(𝑎 , 𝑏) >
𝑓( 𝑎 + 𝑕 , 𝑏 + 𝑘) 𝑜𝑟 𝑓 (𝑎 , 𝑏) < 𝑓( 𝑎 + 𝑕 , 𝑏 + 𝑘) respectively where 𝑕 and 𝑘 are
small values.
Extremum: A function which have a maximum or minimum or both is called
„extremum‟
Extreme value: The maximum value or minimum value or both of a function is Extreme
value.
Stationary points: - To get stationary points we solve the equations = 0 and
= 0 i.e the pairs (𝑎1, 𝑏1) (𝑎2, 𝑏2) (𝑎3 , 𝑏3) … … … … … …are called stationary.
Working procedure:
1. Find and Equate each to zero. Solve these equations for 𝑥 & 𝑦 we get the pair
of values (𝑎1, 𝑏1) (𝑎2, 𝑏2) (𝑎3 , 𝑏3) … … … … … …
Find l =  2f ,m   f , n =  f2
2 2 2
2.
x  x y y
2
3 i) If l n –m > 0 and l < 0 at (𝑎1, 𝑏1) then 𝑓(𝑥 , 𝑦) is maximum at (𝑎1, 𝑏1) and
maximum value is 𝑓(𝑎1, 𝑏1)
ii) If l n –m2 > 0 and l > 0 at (𝑎1, 𝑏1) then 𝑓(𝑥 , 𝑦) is minimum at (𝑎1, 𝑏1) and
minimum value is 𝑓(𝑎1, 𝑏1) .
iii) If l n –m2 < 0 and at (𝑎1, 𝑏1) then 𝑓(𝑥, 𝑦) is neither maximum nor minimum at
(𝑎1, 𝑏1).
In this case (𝑎1, 𝑏1) is saddle point.
iv)If l n –m2 = 0 and at (𝑎1, 𝑏1), no conclusion can be drawn about maximum or
minimum and needs further investigation. Similarly we do this for other stationary
points.
Solved Problems:
1. Locate the stationary points & examine their nature of the following functions.
𝒖 = 𝒙4 + 𝒚4 −𝟐𝒙2 +𝟒𝒙𝒚 − 𝟐𝒚2, (𝒙 > 0, 𝑦 > 0)
Sol: Given 𝑢(𝑥 , 𝑦) = 𝑥 4 + 𝑦4 −2𝑥2 +4𝑥𝑦 − 2𝑦2
𝜕𝑢 𝜕𝑢
For maxima & minima 𝑑𝑥 = 0, 𝜕𝑦 = 0

= 4𝑥 3 −4𝑥 + 4𝑦 = 0  𝑥3 – 𝑥 + 𝑦 = 0 -------------------> (1)

= 4𝑦 3 +4𝑥 − 4𝑦 = 0  𝑦3 + 𝑥 – 𝑦 = 0 -------------------> (2)

Adding (1) & (2),

𝑥 3 + 𝑦3 = 0
 𝑥 = – 𝑦 -------------------> (3)

(1)  𝑥3 – 2𝑥  𝑥 = 0, 2,  2

Hence (3)  𝑦 = 0, - 2, 2

 2u 2  2u  2u
𝑙 = = 12x – 4, 𝑚 = = ( ) =4&𝑛 = = 12y2 – 4
x 2 xy y 2

𝑙𝑛 – 𝑚2 = (12x2 – 4 )( 12y2 – 4 ) -16

At ( , ), 𝑙𝑛 – 𝑚2 = (24 – 4)(24 − 4) − 16 = (20) (20) – 16 >

0 𝑎𝑛𝑑 𝑙 = 20 > 0

The function has minimum value at ( , )

At (0,0) , 𝑙𝑛 – 𝑚2 = (0– 4)(0 -4) -16 = 0 ,
(0,0) is not a extreme value.
2. Investigate the maxima & minima, if any, of the function 𝒇(𝒙) = 𝒙3𝒚2 (𝟏 − 𝒙 − 𝒚).
Sol: Given 𝑓(𝑥) = 𝑥3𝑦2 (1 − 𝑥 − 𝑦) = 𝑥3𝑦2− 𝑥4𝑦2 – 𝑥3𝑦3

= 3𝑥 2𝑦2 – 4𝑥3𝑦2 −3𝑥2𝑦3 = 2𝑥3𝑦 – 2𝑥 4𝑦 − 3𝑥3𝑦2

For maxima & minima = 0 and = 0

 3𝑥 2𝑦2 – 4𝑥3𝑦2 −3𝑥2𝑦3 = 0 ⟹ 𝑥2𝑦2(3 – 4𝑥 − 3𝑦) = 0 --------------->

(1)
 2𝑥 3𝑦 – 2𝑥 4𝑦 − 3𝑥3𝑦2 = 0 => 𝑥3𝑦(2 – 2𝑥 − 3𝑦) = 0 ----------------> (2)
From (1) & (2) 4𝑥 + 3𝑦 – 3 = 0
2𝑥 + 3𝑦 − 2 = 0
2𝑥 = 1 => 𝑥 = ½
4 ( ½) + 3𝑦 – 3 = 0 => 3𝑦 = 3 − 2 , 𝑦 = (1/3)

2 f
𝑙 = = 6𝑥𝑦2−12𝑥 2𝑦2 −6𝑥𝑦3
x 2

 2 f 
 2  (1/2,1/3) = 6(1/2)(1/3)2 -12 (1/2)2(1/3)2 -6(1/2)(1/3)3 = 1/3 – 1/3 -1/9 = -1/9
 x 
2 f   f 
m= =   = 6x2y -8 x3y – 9x2y2
 x y x  y 

 2 f 
  (1/2 ,1/3) = 6(1/2)2(1/3) -8 (1/2)3(1/3) -9(1/2)2(1/3)3 = =
 xy 
2 f
𝑛 = = 2𝑥 3 −2𝑥4 −6𝑥3𝑦
y 2

 2 f 
 2  (1/2,1/3) = 2(1/2)3 -2(1/2)4 -6(1/2)3(1/3) = - - = -
 y 

𝑙𝑛 − 𝑚2 = (−1/9)(−1/8) – (−1/12)2 = − = =

1
> 0 𝑎𝑛𝑑 𝑙 = 0
9
The function has a maximum value at (1/2 , 1/3)
 1 1   1 1  1 1  1  1 1  1
Maximum value is f  ,     1        
 2 3   8 9  2 3  72  2 3  432
3. Find three positive numbers whose sum is 100 and whose product is maximum.
Sol: Let 𝑥 , 𝑦 , 𝑧 be three positive numbers.
 Then 𝑥 + 𝑦 + 𝑧 = 100
 𝑧 = 100 – 𝑥 – 𝑦
Let 𝑓 (𝑥, 𝑦) = 𝑥𝑦𝑧 = 𝑥𝑦(100 – 𝑥 – 𝑦) = 100𝑥𝑦 – 𝑥2𝑦 − 𝑥𝑦2

For maxima or minima = 0 and =0

= 100𝑦 – 2𝑥𝑦 − 𝑦2 = 0 ⟹ 𝑦(100 − 2𝑥 – 𝑦) = 0 ----------------> (1)

= 100𝑥 – 𝑥 2 −2𝑥𝑦 = 0 ⟹ 𝑥(100 – 𝑥 − 2𝑦) = 0 ---------------> (2)

100 − 2𝑥 – 𝑦 = 0
200 − 2𝑥 − 4𝑦 = 0
-----------------------------
-100 + 3y = 0 => 3y =100 => y =100/3
100 – x –(200/3) = 0 => x = 100/3
2 f
𝑙= =- 2y
x 2

 2 f 
 2  (100/3 , 100/3 ) = - 200/3
 x 
2 f   f 
𝑚= =   = 100 -2x -2y
 x y x  y 

 2 f 
  (100/3 , 100/3 ) = 100 –(200/3) –(200/3) = -(100/3)
 xy 
2 f
𝑛= = -2x
y 2

 2 f 
 2  (100/3 , 100/3 ) = - 200/3
 y 
𝑙𝑛 − 𝑚2 = (-200/3) (-200/3) - (-100/3)2 = (100)2 /3
The function has a maximum value at (100/3 , 100/3)
100 100 100
i.e. at 𝑥 = 100/3, 𝑦 = 100/3 𝑧 = 100   
3 3 3
The required numbers are 𝑥 = 100/3, 𝑦 = 100/3, 𝑧 = 100/3
4. Find the maxima & minima of the function 𝒇(𝒙) = 𝟐(𝒙2 – 𝒚2) – 𝒙4 +𝒚4
Sol: Given 𝒇(𝒙) = 𝟐(𝒙2 – 𝒚2) – 𝒙4 +𝒚4 = 2𝑥 2 – 2𝑦2 – 𝑥4 +𝑦4
 For maxima & minima = 0 and =0

= 4x - 4x3 = 0 => 4x(1-x2) = 0 => x = 0 , x = ± 1

= -4y + 4y3 = 0 => -4y (1-y2) = 0 =>y = 0, y = ± 1

 2 f 
𝑙 =  2  = 4-12x2
 x 

 2 f 
𝑚 =   . = Error! Bookmark not defined.   f 
 
= 0
 xy  x  y 

 2 f 
𝑛 =  2  Error! Bookmark not defined. = −4 + 12𝑦2
 y 
we have 𝑙𝑛 – 𝑚2 = (4 − 12𝑥 2)( −4 + 12𝑦2 ) – 0
= -16 +48x2 +48y2 -144x2y2 = 48x2 +48y2 -144x2y2 -16
i) At ( 0 , ± 1 )
ln – m2 = 0 + 48 - 0 -16 =32 > 0
l = 4-0 = 4 > 0
f has minimum value at ( 0 , ± 1 )
f (x ,y ) = 2(x2 –y2) –x4 +y4
f ( 0 , ± 1 ) = 0 – 2 – 0 + 1 = -1
The minimum value is „-1 „.
ii) At ( ± 1 ,0 )
𝑙𝑛 – 𝑚2 = 48 + 0 − 0 − 16 = 32 > 0
l = 4-12 = - 8 < 0
f has maximum value at ( ± 1 ,0 )
f (x ,y ) = 2(x2 –y2) –x4 +y4
f ( ± 1 , 0 ) =2 -0 -1 + 0 = 1
The maximum value is „1 „.
iii) At (0,0) , (± 1 , ± 1)
𝑙𝑛 – 𝑚2 < 0
l = 4 -12x2
(0 , 0) & (± 1 , ± 1) are saddle points.
f has no max & min values at (0 , 0) , (± 1 , ± 1).
5. Find the maximum and minimum values of 𝒙𝟑 + 𝒚𝟑 − 𝟑𝒂𝒙𝒚.
Sol : let 𝑧 = 𝑥 3 + 𝑦 3 − 3𝑎𝑥𝑦
𝜕𝑧
Here 𝑑𝑥 = 3𝑥 2 − 3𝑎𝑦 = 0 ------(1)
𝜕𝑧
= 3𝑦 2 − 3𝑎𝑥 = 0 -----------(2)
𝑑𝑦

Solving (1) and (2) then

From (1) 3𝑥 2 = 3𝑎𝑦
𝑥 2 = 𝑎𝑦
𝑥2
𝑦= 𝑎

Substitute the above value in equation (2) then we get

𝑥4
3 2 − 3𝑎𝑥 = 0
𝑎
⇒ 3𝑥 4 − 3𝑎3 𝑥 = 0
⇒ 3𝑥 𝑥 3 − 𝑎3 = 0
⇒ 𝑥 = 0, 𝑥 = 𝑎.
Corresponding values of 𝑦 are 𝑦 = 0, 𝑦 = 𝑎.
𝜕2𝑧 𝜕2𝑧 𝜕2𝑧
Now 𝑙 = 𝜕𝑥 2 = 6𝑥 , 𝑚 = 𝜕𝑥𝑑𝑦 = −3𝑎, 𝑛 = 𝜕𝑦 2 = 6𝑦

(i) At the point 0,0 , 𝑙𝑛 – 𝑚2= 36𝑥𝑦 − 9𝑎2 < 0

Therefore 𝑧 does not have any extreme value.
(ii) At the point 𝑎, 𝑎 , 𝑙𝑛 – 𝑚2= 36𝑎2 − 9𝑎2 = 27𝑎2 > 0 and 𝑙 = 6𝑎 > 0
𝑖𝑓 𝑎 > 0
In this case 𝑧 attains minimum value and minimum value=−𝑎3. 𝑙 = 6𝑎
< 0 𝑖𝑓 𝑎 < 0
In this case 𝑧 attains maximum value and maximum value=−𝑎3.
6. Examine for minimum and maximum values of 𝒔𝒊𝒏𝒙 + 𝒔𝒊𝒏𝒚 + 𝐬𝐢𝐧 𝒙 + 𝒚 .
Sol : let 𝑢 𝑥, 𝑦 = 𝑠𝑖𝑛𝑥 + 𝑠𝑖𝑛𝑦 + sin⁡
(𝑥 + 𝑦) ---------(1)
𝜕𝑢
= 𝑐𝑜𝑠𝑥 + cos 𝑥 + 𝑦 = 0-----------(2)
𝜕𝑥
𝜕𝑢
= 𝑐𝑜𝑠𝑦 + cos 𝑥 + 𝑦 = 0-----------(3)
𝜕𝑦

On solving (2) and (3) we get 𝑐𝑜𝑠𝑥 = 𝑐𝑜𝑠𝑦

⟹𝑥=𝑦
From (2) 𝑐𝑜𝑠𝑥 + 𝑐𝑜𝑠2𝑥 = 0
3𝑥 𝑥
⟹ 2𝑐𝑜𝑠 𝑐𝑜𝑠 = 0
2 2
 3𝑥 𝜋 𝑥 𝜋
⟹ = ± or =±
2 2 2 2
𝜋
⟹𝑥=± 𝑜𝑟 𝑥 = ±𝜋
3
𝜋 𝜋 𝜋 𝜋
When 𝑥 = ± 3 then 𝑦 = ± 3 i.e (± 3 , ± 3 )

When 𝑥 = ±𝜋 then 𝑦 = ±𝜋 i.e (±𝜋, ±𝜋)

𝜕2𝑢 𝜕2𝑢
𝑙 = 𝜕𝑥 2 = −𝑠𝑖𝑛𝑥 − sin⁡
(𝑥 + 𝑦) ; 𝑚 = = −sin⁡
(𝑥 + 𝑦)
𝜕𝑥𝜕𝑦

𝜕2𝑢
𝑛 = 𝜕𝑦 2 = −𝑠𝑖𝑛𝑦 − sin⁡
(𝑥 + 𝑦)

𝜋 𝜋 − 3
(i) At ( , ) then 𝑙 = − 3, 𝑚 = 𝑎𝑛𝑑 𝑛 = − 3
3 3 2
9
𝑙𝑛 – 𝑚2 = > 0 𝑎𝑛𝑑 𝑙 < 0
4
𝜋 𝜋
⟹ 𝑢 has maximum at ( 3 , 3 )
𝜋 𝜋 3 3
At ( 3 , 3 ) , 𝑢 = 2
3 3
∴ Maximum value of 𝑢 = 2
𝜋 𝜋
(ii)At (− 3 , − 3 ) we have 𝑙𝑛 – 𝑚2 > 0 and 𝑙 > 0
𝜋 𝜋
∴ 𝑢 has a minimum at (− 3 , − 3 )
−3 3
∴ Minimum value of 𝑢 = 2
2
At ±𝜋, ±𝜋 , 𝑙𝑛 − 𝑚 = 0
∴ there is a need for further investigation.
7. A rectangular box open at the top is to have volume of 32 cubic ft. Find the
dimensions of the box requiring least material for its construction.
Sol : Let 𝑥 𝑓𝑡 , 𝑦 𝑓𝑡 and 𝑧 𝑓𝑡 be the dimensions of the box and
Let S be the surface of the box. Then we have
𝑆 = 𝑥𝑦 + 2𝑦𝑧 + 2𝑧𝑥 (since open at the top) ------(1)
Given that its volume 𝑥𝑦𝑧 = 32 ------(2)
32
From (2) , 𝑧 = 𝑥𝑦

Substitute the value of 𝑧 in (1) , we get

32 32 64 64
𝑆 = 𝑥𝑦 + 2𝑦 +2 𝑥 = 𝑥𝑦 + +
𝑥𝑦 𝑥𝑦 𝑥 𝑦
𝜕𝑆 64 𝜕𝑆 64
= 𝑦 − 𝑥 2 = 0 , 𝜕𝑦 = 𝑥 − 𝑦 2 = 0
𝜕𝑥

By solving these equations we get 𝑥 = 4 , 𝑦 = 4

 𝜕2𝑆 128 𝜕2𝑆 𝜕2𝑆 128
𝑙 = 𝜕𝑥 2 = , 𝑚 = 𝜕𝑥𝑑𝑦 = 1, 𝑛 = 𝜕𝑦 2 =
𝑥3 𝑦3
128 128 128
At 𝑥, 𝑦 = (4,4) , 𝑙𝑛 – 𝑚2 = . − 1 = 2.2 − 1 = 3 > 0 and 𝑙 = =2>0
𝑥3 𝑦3 𝑥3

Thus, 𝑆 is minimum when 𝑥 = 4, 𝑦 = 4. From (2) we get 𝑧 = 2.

∴ The dimensions of the box for least material for its construction are 4,4,2.
8. Find the volume of the largest rectangular parallelepiped that can be inscribed in
𝒙𝟐 𝒚𝟐 𝒛𝟐
the ellipsoid 𝒂𝟐 + 𝒃𝟐 + 𝒄𝟐 = 𝟏 .

Sol : Let length, breadth and height be 2𝑥, 2𝑦, 2𝑧 and volume is 𝑉
∴ 𝑉 = 2𝑥 2𝑦 2𝑧 = 8𝑥𝑦𝑧
𝑥2 𝑦2
Let 𝑓 = 𝑉 2 = 64𝑥 2 𝑦 2 𝑧 2 = 64𝑥 2 𝑦 2 𝑐 2 (1 − 𝑎 2 − 𝑏 2 ) (from (1) 𝑧 2 value substituted )

𝑥4 𝑦2 𝑥2 𝑦4
2 2 2
⟹ 𝑓 = 64𝑐 (𝑥 𝑦 − 2 − 2
𝑎 𝑏
𝜕𝑓 2 2
4𝑥 3 𝑦 2 2𝑥𝑦 4
= 64𝑐 2𝑥𝑦 − − 2
𝜕𝑥 𝑎2 𝑏
𝜕𝑓 2𝑥 4 𝑦 4𝑥 2 𝑦 3
= 64𝑐 2 2𝑥 2 𝑦 − 2 −
𝜕𝑦 𝑎 𝑏2
𝜕𝑓 2𝑥 2 𝑦2
=0 ⟹ + 𝑏 2 = 1 and
𝜕𝑥 𝑎2
𝜕𝑓 𝑥2 2𝑦 2
= 0 ⟹ 𝑎2 + =1
𝜕𝑦 𝑏2
𝑎 𝑏
Solving these equations , we get 𝑥 = and 𝑦 =
3 3
𝜕2𝑓 2𝑥 3 𝑦 2𝑦 4
Again 𝑙 = 𝜕𝑥 2 = 64𝑐 2 2𝑦 2 − −
𝑎2 𝑏2

𝜕2𝑓 2
8𝑥 3 𝑦 8𝑥𝑦 3
𝑚= = 64𝑐 4𝑥𝑦 − 2 − 2
𝜕𝑥𝜕𝑦 𝑎 𝑏
𝜕2𝑓 2 2
2𝑥 4 12𝑥 2 𝑦 2
𝑛 = 2 = 64𝑐 2𝑥 − 2 −
𝜕𝑦 𝑎 𝑏2
𝑎 𝑏
At ( , ) , 𝑙𝑛 – 𝑚2 > 0 and 𝑙 < 0
3 3
𝑎 𝑏
Therefore 𝑓 is maximum at At ( , )
3 3
𝑎2 𝑏2 𝑎4𝑏2 𝑎2𝑏4 64𝑐 2
Maximum value of 𝑓 = 64𝑐 2 − 9.3.𝑎 2 − 3.9.𝑏 2 = (3𝑎2 𝑏 2 − 2𝑎2 𝑏 2 )
3 3 27
64 8𝑎𝑏𝑐
i.e 𝑉 2 = 27 𝑎2 𝑏 2 𝑐 2 ⟹ 𝑉 = 27
.

9. Find the points on the surface 𝒛𝟐 = 𝒙𝒚 + 𝟏 that are nearest to the origin.
Sol : Let 𝑃 𝑥, 𝑦, 𝑧 be any point on the surface
Let the surface be ∅ 𝑥, 𝑦, 𝑧 = 𝑧 2 − 𝑥𝑦 − 1 = 0 -------(1)
Let 𝑂(0,0,0, ) be the origin then 𝑂𝑃 = 𝑓 = 𝑥 2 + 𝑦 2 +𝑧 2 -------(1)
We have to minimize OP (i.e 𝑓 ) subject to the condition (1)
From (2) 𝑓 2 = 𝒙𝟐 + 𝒚𝟐 +𝒛𝟐 = 𝒙𝟐 + 𝒚𝟐 + 𝒙𝒚 + 𝟏 [ from (1) ]
Now differentiating 𝑓 partially on both sides w.r.to 𝑥 𝑎𝑛𝑑 𝑦 then we get
𝜕𝑓 𝜕𝑓 2𝑥+𝑦
2𝑓 𝜕𝑥 = 2𝑥 + 𝑦 ⟹ 𝜕𝑥 = -------(3)
2𝑓
𝜕𝑓 𝜕𝑓 2𝑦 +𝑥
2𝑓 𝜕𝑦 = 2𝑦 + 𝑥 ⟹ = ---------(4)
𝜕𝑦 2𝑓
𝜕𝑓 𝜕𝑓
Now the critical points are given by = 0, =0
𝜕𝑥 𝜕𝑦

⟹ 2𝑥 + 𝑦 = 0 𝑎𝑛𝑑 2𝑦 + 𝑥 = 0 ⟹ 𝑥 = 0 , 𝑦 = 0
Now from equation (1) we get 𝑧 = ±1
Therefore 𝑃 0,0,1 and 𝑄 0,0, −1 are the critical points of 𝑓
Differentiate (3) partially w.r.to 𝑥 𝑎𝑛𝑑 𝑦 then we get
𝜕𝑓
𝜕 2 𝑓 2 2𝑓 − 2𝑥 + 𝑦 2 𝜕𝑥
𝑙= 2=
𝜕𝑥 2𝑓 2
𝜕𝑓
𝜕2𝑓 2𝑓 − 2𝑥 + 𝑦 2
𝜕𝑦
𝑚= = 2
𝜕𝑥𝜕𝑦 2𝑓
𝜕𝑓
𝜕 2 𝑓 2 2𝑓 − 2𝑦 + 𝑥 2 𝜕𝑦
𝑛= 2=
𝜕𝑦 2𝑓 2
Now at 𝑃 0,0,1 , 𝑙𝑛 – 𝑚2 > 0 and 𝑙 < 0
Therefore 𝑓 has minimum at 𝑃 0,0,1
Now at 𝑄 0,0,1 , 𝑙𝑛 – 𝑚2 > 0 and 𝑙 < 0
Therefore 𝑓 has minimum at 𝑄 0,0, −1
Hence required points are 𝑃 0,0,1 and 𝑄 0,0, −1 .
10. Discuss the maximum and minimum of 𝒙𝟐 + 𝒚𝟐 +6x+12
Sol: Let 𝑓 𝑥, 𝑦 = 𝑥 2 + 𝑦 2 + 6𝑥 + 12
𝜕𝑓 𝜕𝑓
We have 𝜕𝑥 = 2𝑥 + 6 , 𝜕𝑦 = 2𝑦
𝜕 𝜕𝑓 𝜕 𝜕𝑓 𝜕 𝜕𝑓
And 𝑙 = 𝜕𝑥 (𝜕𝑥 ) = 2 , 𝑚 = 𝜕𝑥 = 0 , 𝑛 = 𝜕𝑦 =2.
𝜕𝑦 𝜕𝑦
𝜕𝑓 𝜕𝑓
For maxima and minima the condition is 𝜕𝑥
= 0 𝑎𝑛𝑑 =0.
𝜕𝑦

i.e2𝑥 + 6 = 0 𝑎𝑛𝑑 2𝑦 = 0 => 𝑥 = −3 𝑎𝑛𝑑 𝑦 = 0

 At the point (−3,0) :
𝑙𝑛 − 𝑚2 = 2 ∗ 2 − 0 = 4 > 0 𝑎𝑛𝑑 𝑙 = 2 > 0
Hence 𝑓(𝑥, 𝑦) will be minimum when 𝑥 = −3 𝑎𝑛𝑑 𝑦 = 0
Minimum value =𝑓 −3,0 = 9 + 0 − 18 + 12 = 3.
11.Find the maximum and minimum values of the function 𝒇 𝒙 = 𝒙𝟓 − 𝟑𝒙𝟒 + 𝟓 .
Sol: Given 𝑓 𝑥 = 𝑥 5 − 3𝑥 4 + 5
∴ 𝑓 1 𝑥 = 5𝑥 4 − 12𝑥 3 ,𝑓 11 𝑥 = 20𝑥 3 − 36𝑥 2 = 4𝑥 2 (5𝑥 − 9) .
For 𝑓(𝑥) to be maximum or minimum,𝑓 1 𝑥 = 0 .
12
i.e𝑥 3 5𝑥 − 12 = 0 => 𝑥 = 0 , .
5

when 𝑥 = 0 , 𝑓 11 𝑥 = 0 .
∴ ,𝑓 𝑥 has neither maxmum nor minimum at 𝑥 = 0
12 4 144
when 𝑥 = ,𝑓 11 𝑥 = 12 − 9 > 0
5 25
12
∴ ,𝑓 𝑥 has minimum at 𝑥 = 5
12 12 5 12 4
minimum value =𝑓 = 5
−3 +5 ,
5 5 5 4

=(2.4)5 − 3 (2.4)4 + 5 = −14.91

Lagrange‟s method of undetermined multipliers
Suppose it is required to find the extremum for the function f(x , y , z) subject to condition
( x , y , z) = 0 ------------- (1)
Step 1 : Form lagrangean function 𝐹(𝑥 , 𝑦 , 𝑧) = f(x , y , z) + ( x , y , z) where is called
Lagrange‟s constant, which is determined by the following conditions.
Step 2: Obtain the equations
F
= 0 => + = 0 --------------- (2)
x
F
= 0 => + = 0 --------------- (3)
y

F
= 0 => + = 0 --------------- (4)
z
Step 3: Solving the equations (1) (2) (3) & (4) we get the stationary point (x, y, z).
Step 4 : Substitute the value of 𝑥 , 𝑦 , 𝑧 so obtained in equation (1) we get the extremum.
Solved Problems:
1. Find the minimum value of 𝒙2 +𝒚2 +𝒛2, given 𝒙 + 𝒚 + 𝒛 = 𝟑𝒂
Sol: 𝑢 = 𝑥2 +𝑦2 +𝑧2
 = 𝑥 + 𝑦 + 𝑧 − 3𝑎 = 0
Using Lagrange‟s function
𝐹(𝑥 , 𝑦 , 𝑧) = 𝑢(𝑥 , 𝑦 , 𝑧) + ( 𝑥 , 𝑦 , 𝑧)
For maxima or minima
F
= + = 2𝑥 + = 0 ------------ (1)
x
F
= + = 2𝑦 + = 0 ------------ (2)
y

F
= + = 2z + = 0 ------------ (3)
z
(1) , (2) & (3)
= -2x = -2y = -2z

= x + x + x - 3a = 0
=a
= y =z = a
Minimum value of u = a2 + a2 + a2 =3 a2
2. Find the minimum value of 𝐱 𝟐 + 𝐲 𝟐 + 𝐳 𝟐 , given that xyz = 𝐚𝟑
Sol: Let u = x 2 + y 2 + z 2 …… (1)
And ∅ = xyz - 𝑎3 = 0 …… (2)
Consider the lagrangean function F(x,y,z) = u( x,y,z ) + λ ϕ(x, y, z)
i.e , F(x,y,z) = x 2 + y 2 + z 2 + λ (xyz − a3 ) ……(3)
∂F ∂u ∂∅
Now ∂x = 0 ⇒ ∂x + λ ∂x = 2x + λyz = 0 ……(4)
∂F ∂u ∂∅
= 0 ⇒ ∂y + λ ∂y = 2y + λxz = 0 ……(5)
∂y
∂F ∂u ∂∅
= 0 ⇒ ∂z + λ ∂z = 2z + λyx = 0 ……(6)
∂z
x y z λ
From (4) , (5) and (6) ,we have yz = = = −2 ……(7)
xz xy
x y
From the first two members , we have yz = ⇒ x2 = y2 …(8)
xz
y z
From the last members , we have xz = ⇒ y2 = z2 ……(9)
xy

From (8) and (9) , we have x 2 = y 2 = z 2 ⇒ x = y = z ……(10)

on solving (2) and (10) , we get , x = y = z = a
∴ Minimum value of u = a2 + a2 = 3a2
3. Find the maximum value of u = 𝒙𝟐 𝒚𝟑 𝒛𝟒 if 2x + 3y + 4z = a
Sol: Given u = x 2 y 3 z 4 .….. (1)
Let ϕ(x, y, z) = 2x + 3y + 4z - a = 0 ……(2)
Consider the lagrangean function F(x,y,z) = u( x,y,z ) + λ ϕ(x, y, z)
i.e, , F(x,y,z) = x 2 y 3 z 4 + λ (2x + 3y + 4z – a) ……(3)
∂F ∂F ∂F
for maxima or minima ∂x = 0 , ∂y = 0 , =0
∂z
∂F
Now ∂x = 0 ⇒ 2x𝑦 3 𝑧 4 + 2λ = 0 ⇒ x𝑦 3 𝑧 4 = − λ ……(4)
∂F
= 0 ⇒ 3x 2 y 2 z 4 +3λ = 0 ⇒ x 2 y 2 z 4 = − λ …….(5)
∂y
∂F
and = 0 ⇒ 4x 2 y 3 z 3 + 4λ =0 ⇒ x 2 y 3 z 3 = − λ …….(6)
∂z

From (4) and (5) , we have x = y …….(7)

From (5) and (6) , we have y = z …….(8)
Hence from (7) and (8) , we get x = y = z …….(9)
a
On solving (2) and (9) ,we get x = y = z = 9
a 2 a 3 a 4 a 9
∴ Maximum value of u = =
9 9 9 9
4. Show that the rectangular solid of maximum volume that can be inscribed in a
sphere is a cube.
Sol: Let 2x ,2y, 2z are the length , breadth and height of rectangular solid
Then its volume V = 8 xyz ……..(1)
Let the sphere have a radius of „r‟ so that x 2 + y 2 + z 2 = r 2 ……(2)
Consider the lagrangean function F(x,y,z) = u( x,y,z ) + λ ϕ(x, y, z)
i.e, F(x,y,z) = V + 𝜆 (x 2 + y 2 + z 2 − r 2 )
= 8xyz + 𝜆 (x 2 + y 2 + z 2 − r 2 ) …….(3)
∂F ∂F ∂F
For maxima or minima ∂x = 0 , ∂y = 0 , =0
∂z
∂F
= 0 ⇒ 8yz +2 λx = 0 ……(4)
∂x
∂F
= 0 ⇒ 8zx +2 λy = 0 ……(5)
∂y
∂F
= 0 ⇒ 8xz +2 λz = 0 ……(6)
∂z

From (4) ,(5) and (6) we have 2𝑥 2 𝜆 = - 8xyz = - 2𝑦 2 𝜆 = - 2𝑧 2 𝜆

⇒x=y=z
Thus for a maximum value x = y = z which shows that the rectangular solid is a cube.
 𝟏 𝟏 𝟏
5. Find the maximum and minimum values of x + y + z subject to + + = 𝟏
𝒙 𝒚 𝒛
𝟏 𝟏 𝟏
Sol:Let the function be 𝑓 𝑥, 𝑦, 𝑧 = 𝐱 + 𝐲 + 𝐳 subject to the constraints 𝒙 + 𝒚 + 𝒛 = 𝟏
𝟏 𝟏 𝟏
The Auxiliary function 𝐹 𝑥, 𝑦, 𝑧 = 𝑥 + 𝑦 + 𝑧 + 𝜆(𝒙 + 𝒚 + 𝒛 − 𝟏) -------(1)

Differentiating (1) partially w.r.to 𝑥, 𝑦, 𝑧 and equating to zero, we get

𝜕𝐹 𝜆
= 1 − 𝑥 2 = 0 -----(2)
𝜕𝑥
𝜕𝐹 𝜆
= 1 − 𝑦 2 = 0 ----(3)
𝜕𝑦
𝜕𝐹 𝜆
= 1 − 𝑧 2 = 0 -----(4)
𝜕𝑧

On solving (2) ,(3) and (4) for 𝑥, 𝑦, 𝑧 then we get 𝑥 = ± 𝜆 , 𝑦 = ± 𝜆, 𝑧 = ± 𝜆

1 1 1
Now substitute these 𝑥, 𝑦, 𝑧 in the given constraint then we get + + =1
𝜆 𝜆 𝜆

⇒ 𝜆 = 9 ⇒ x = ±3, 𝑦 = ±3, 𝑧 = ±3
Thus the maximum and minimum values are 9 and -9.
6.Find the point on the plane 𝒙 + 𝟐𝒚 + 𝟑𝒛 = 𝟒 that is closest to the origin .
Sol:
Let 𝑝 𝑥, 𝑦, 𝑧 be a point on the given plane .then 𝑂𝑃 = 𝑥 2 + 𝑦 2 + 𝑧 2
Let 𝑓 𝑥, 𝑦, 𝑧 = 𝑥 2 + 𝑦 2 + 𝑧 2 … … … . > (1)
Now ,we have to minimize (1) subject to the condition
𝑢 𝑥, 𝑦, 𝑧 = 𝑥 + 2𝑦 + 3𝑧 − 4 = 0 … … … … … . > (2)
Consider the Lagrangian function 𝐹 𝑥. 𝑦. 𝑧 = 𝑓 𝑥, 𝑦, 𝑧 + 𝑡 . 𝑢(𝑥, 𝑦, 𝑧)
i.e𝐹 𝑥, 𝑦, 𝑧 = 𝑥 2 + 𝑦 2 + 𝑧 2 +t(x+2y+3z-4)

𝜕𝐹 𝜕𝐹 𝜕𝐹
For 𝐹 to be minima,𝜕𝑥 = 0 , 𝜕𝑦 = 0 , 𝜕𝑧 = 0 .

𝜕𝐹 𝑡
= 0 => 2𝑥 + 𝑡 = 0 => 𝑥 = − … … … (3)
𝜕𝑥 2
𝜕𝐹
𝜕𝑦
= 0 => 2𝑦 + 2𝑡 = 0 => y=-t ………..(4)
𝜕𝐹 3𝑡
= 0 => 2𝑧 + 3𝑡 = 0 => 𝑧 = − … … … (5)
𝜕𝑧 2
𝑡 9𝑡
Substituting (3),(4) and (5) in (2) ,we get − 2 − 2𝑡 − − 4 = 0 => 14𝑡 + 8 = 0
2
4
∴𝑡=−
7
2 4 6
Putting this value of t in (3), (4) and (5), we get 𝑥 = 7 , y=7 ,𝑧 = 7
2 4 6
Hence (7 , 7 , 7 ) is the point on the given plane which is nearest to the origin.
7.Find the volume of the largest rectangular parallelopiped that can be inscribedin ellipsoid
𝑥2 𝑦2 𝑧2
+ 𝑏 2 + 𝑐 2 = 1.
𝑎2
Sol: Let 2𝑥, 2𝑦, 2𝑧 be the length, breadth and height of the rectangular parallelopiped that can
be inscribed in the ellipsoid
𝑥2 𝑦2 𝑧2
+ 𝑏 2 + 𝑐 2 − 1 = 0……………..(1)
𝑎2
Then the centroid of the parallelopiped coincides with the center 𝑂(0,0,0) of the ellipsoid and
the corners of the parallelopiped lie on the surface of the ellipsoid (1).
If 𝑥, 𝑦, 𝑧 is any corner of the parallelopiped then it satisfies condition (1).
Let V be the volume of the parallelopiped
i.e𝑉 = 8𝑥𝑦𝑧.
We have to find the maximum value of V subject to the condition (1).
Consider the Lagrangean function :
𝑥2 𝑦2 𝑧2
𝐹 𝑥, 𝑦, 𝑧 = 𝑉 + 𝑡 ( 2 + 2 + 2 − 1) … … … . (2)
𝑎 𝑏 𝑐
Where “t” is the multiplier to be determined such that
𝜕𝐹 𝜕𝐹 𝜕𝐹
=0, = 0, =0
𝜕𝑥 𝜕𝑦 𝜕𝑧
2𝑥
=>8𝑦𝑧 + 𝑡 2 = 0 … … … … . . (3)
𝑎
2𝑦
=>8𝑥𝑧 + 𝑡 = 0 … … … … . . (4)
𝑏2
2𝑧
=>8𝑦𝑥 + 𝑡 𝑐 2 = 0 … … … … . . (5)
Now (3),(4) ,(5) are combined as :
𝑎2 𝑦𝑧 𝑏 2 𝑧𝑥 𝑐 2 𝑥𝑦 𝑡
= = = − … … … (6)
𝑥 𝑦 𝑧 4

𝑎2𝑦 𝑏2𝑥 𝑥2 𝑦2
From first two fractions,we have = => 𝑎 2 = 𝑏 2 … … … … … . > (7)
𝑥 𝑦

𝑦2 𝑧2
Similarly, 𝑏 2 = 𝑐 2 … … … . . (8)
Substituting (7) and (8) in (1) ,we get in(1) ,we get
𝑎 𝑏 𝑐
𝑥 = 3 ,𝑦 = 3 ,𝑧 = 3
𝑎 𝑏 𝑐
Hence the possible extreme point is 𝑝( , , ),
3 3 3
First ,we note that for fixed x and y,𝑉 = 8𝑥𝑦𝑧 ,is an increasing function of z.

If 𝑧 = 0 ,the parallelepiped reduces into a two dimensional lamina for which 𝑉 = 0 ,as z
increases ,V also increased.
𝑎 𝑏 𝑐
But for 𝑥 = , 𝑦 = 3 ,the maximum value of z satisfying the condition (1) is 𝑧 = 3 .
3
𝑎 𝑏 𝑐 8𝑎𝑏𝑐
Thus V is maximum at 𝑝( , , ) and its maximum value is 𝑉 = 8𝑥𝑦𝑧 = 𝑐. 𝑢.
3 3 3 3 3
Taylor‟s series for a function of two variables:

Consider a function 𝑓(𝑥, 𝑦) defined in a region enclosing (𝑎, 𝑏) and having successive
partial derivatives ,then taylor‟s series gives an expansion of 𝑓(𝑥, 𝑦) in powers of (𝑥 − 𝑎)
and (𝑦 − 𝑏) and partial derivatives of 𝑓 at (𝑎, 𝑏) and is expressed in ascending powers of
(𝑥 − 𝑎) and (𝑦 − 𝑏) .
1
f(x,y) = f(a,b) + (x-a) fx a, b + y − b fy a, b + 2! [ (x − a)2 fxx a, b + 2 x − a
1
y − b fxy a, b + y − b)2 fyy a. b + 3! [ (x − a)3 fxxx a, b + 3 x − a 2
y−

b fxxy a, b + 3(x − a)(y − b)2 fxyy a, b + y − b)3 fyyy a, b + …….

 Note: The above expansion is called the expansion of f(x,y) at (a,b) or in the
neighbouhood of (a,b) or in powers of (x-a) and (y-b) .
Solved Problems:
𝝅
1. Expand 𝐞𝐱 𝐜𝐨𝐬 𝐲 near (1, 𝟒 )
𝜋 𝑒
Sol: Let f(x,y) = ex cos y ⇒ f(1, 4 ) = . Then
2
π e
fx x, y = ex cos y ⇒ f(1, ) =
4 2
π −e
fy x, y = −ex sin y ⇒ f(1, 4 ) = 2
π −e
fxx x, y = −ex cos y ⇒ f(1, 4 ) = 2
π −e
fxy x, y = −ex sin y ⇒ f(1, 4 ) = 2
π −e
fyy x, y = −ex cos y ⇒ f(1, 4 ) = 2

By substituting above values in taylor ′ s series , we get

π 2
e π (x−1)2 π y−
x 4
e cos y = 1+ x−1 − y−4 + − x−1 y−4 − +⋯
2 2! 2!

2. Expand 𝐱 𝟐 𝐲 + 𝟑𝐲 − 𝟐 𝐢𝐧 𝐩𝐨𝐰𝐞𝐫𝐬 𝐨𝐟 𝐱 − 𝟏 𝐚𝐧𝐝 𝐲 + 𝟐 using taylor‟s thorem.

Sol: Let f(x,y) = x 2 y + 3y − 2 , a = 1 , b = -2
Now f(a,b) = f(1,-2) = -10
fx a, b = 2xy ⇒ fx 1, −2 = −10
fy a, b = x 2 + 3 ⇒ fy 1, −2 = 4
fxx a, b = 2y ⇒ fxx 1, −2 = - 4
fxy a, b = 2x ⇒ fxy 1, −2 = 4
fyy a, b = 0 ⇒ fyy 1, −2 = 0
fxxx a, b = 0 ⇒ fxxx 1, −2 = 0
fxxy a, b = 2 ⇒ fxxy 1, −2 = 2
fxyy a, b = 0 ⇒ fxyy 1, −2 = 0
fyyy a, b = 0 ⇒ fyyy 1, −2 = 0
All other partial derivatives of higher order will vanish
By substituting above values in taylor ′ s series , we get
1
x 2 y + 3y − 2 = 10 + (x − 1 (−4) + (y + 2)4] + [ x−1 2
−4 + 2 x − 1 y +
2

22+ y+220+16[x−13(0)+3(x−1 2y+22+3x−1(y+2)20+(y+2)30]

 = 10 - 4(x - 1) + 4( y + 2) - 2(x − 1)2 + 2 (x - 1) (y + 2) + (x − 1)2 (y + 2).
3. Expand the function 𝒇 𝒙, 𝒚 = 𝒆𝒚 𝐥𝐨𝐠⁡
(𝟏 + 𝒚) in terms of 𝒙 and y up to the terms of
third degree using Taylor‟s theorem.
Sol: Given that 𝑓 𝑥, 𝑦 = 𝑒 𝑦 log⁡
(1 + 𝑦); 𝑓 0,0 = 0
𝑓𝑥 𝑥, 𝑦 = 𝑒 𝑥 log⁡(1 + 𝑦) ; 𝑓𝑥 0,0 = 0
1
𝑓𝑦 𝑥, 𝑦 = 𝑒 𝑥 1+𝑦 ; 𝑓𝑦 0,0 = 1
1
𝑓𝑥𝑦 𝑥, 𝑦 = 𝑒 𝑥 1+𝑦 ; 𝑓𝑥𝑦 0,0 = 1

𝑓𝑥𝑥 𝑥, 𝑦 = 𝑒 𝑥 log⁡
(1 + 𝑦) ; 𝑓𝑥𝑥 0,0 = 0
1
𝑓𝑦𝑦 𝑥, 𝑦 = −𝑒 𝑥 ; 𝑓𝑦𝑦 0,0 = −1
1+𝑦 2
1
𝑓𝑥𝑥𝑦 𝑥, 𝑦 = 𝑒 𝑥 . 1+𝑦 ; 𝑓𝑥𝑥𝑦 0,0 = 1
1
𝑓𝑥𝑦𝑦 𝑥, 𝑦 = −𝑒 𝑥 . ; 𝑓𝑥𝑦𝑦 0,0 = −1
1+𝑦 2

𝑓𝑥𝑥𝑥 𝑥, 𝑦 = 𝑒 𝑥 log⁡
(1 + 𝑦) ; 𝑓𝑥𝑥𝑥 0,0 = 0
1
𝑓𝑦𝑦𝑦 𝑥, 𝑦 = 2𝑒 𝑥 . ; 𝑓𝑦𝑦𝑦 0,0 = 2
1+𝑦 3

Therefore by Taylor‟s theorem

𝑓 𝑥, 𝑦 = 𝑓 0,0 + 𝑥𝑓𝑥 0,0 + 𝑦𝑓𝑦 0,0
1 2
+ 𝑥 𝑓𝑥𝑥 0,0 + 2𝑥𝑦 𝑓𝑥𝑦 0,0 + 𝑦 2 𝑓𝑦𝑦 0,0
2!
1 3
+ 𝑥 𝑓𝑥𝑥𝑥 0,0 + 3𝑥 2 𝑦𝑓𝑥𝑥𝑦 0,0 + 3𝑥𝑦 2 𝑓𝑥𝑦𝑦 0,0 + 𝑦 3 𝑓𝑦𝑦𝑦 0,0
3!
+ −−−−−−
1 1
=0+0+𝑦+ 0 + 2𝑥𝑦 − 𝑦 2 + [0 + 3𝑥 2 𝑦 − 3𝑥𝑦 2 + 2𝑦 3 ]
2! 3!
𝑦 2 𝑥 2 𝑦 𝑥𝑦 2 𝑦 3
𝑓(𝑥, 𝑦) = 𝑦 + 𝑥𝑦 − + − +
2 2 2 3
4.Expand 𝒆𝒙 𝒔𝒊𝒏𝒚 in powers of 𝒙 and 𝒚
Sol:
Let 𝑓 𝑥, 𝑦 = 𝑒 𝑥 𝑠𝑖𝑛𝑦 Then Let 𝑓 0,0 = 0
𝑓𝑥 𝑥, 𝑦 = 𝑒 𝑥 𝑠𝑖𝑛𝑦 𝑓𝑥 0,0 = 0
𝑓𝑦 𝑥, 𝑦 = 𝑒 𝑥 𝑐𝑜𝑠𝑦 𝑓𝑦 0,0 = 1
𝑓𝑥𝑥 𝑥, 𝑦 = 𝑒 𝑥 𝑠𝑖𝑛𝑦 𝑓𝑥𝑥 0,0 = 0
𝑓𝑥𝑦 𝑥, 𝑦 = 𝑒 𝑥 𝑐𝑜𝑠𝑦 𝑓𝑥𝑦 0,0 = 1
𝑓𝑦𝑦 𝑥, 𝑦 = −𝑒 𝑥 𝑠𝑖𝑛𝑦 𝑓𝑦𝑦 0,0 = 0
 𝑓𝑥𝑥𝑥 𝑥, 𝑦 = 𝑒 𝑥 𝑠𝑖𝑛𝑦 𝑓𝑥𝑥𝑥 0,0 = 0
𝑓𝑥𝑥𝑦 𝑥, 𝑦 = 𝑒 𝑥 𝑐𝑜𝑠𝑦 𝑓𝑥𝑥𝑦 0,0 = 1
𝑓𝑥𝑦𝑦 𝑥, 𝑦 = −𝑒 𝑥 𝑠𝑖𝑛𝑦 𝑓𝑥𝑦𝑦 0,0 = 0
𝑓𝑦𝑦 𝑦 𝑥, 𝑦 = −𝑒 𝑥 𝑐𝑜𝑠𝑦 𝑓𝑦𝑦𝑦 0,0 = −1
By Taylors theorem,
1
𝑓 𝑥, 𝑦 = 𝑓 0,0 + 𝑥𝑓𝑥 (0,0)+ 𝑦𝑓𝑦 0,0 + 2! 𝑥 2 𝑓𝑥𝑥 0,0 + 2𝑥𝑦𝑓𝑥𝑦 0,0 + 𝑦 2 𝑓𝑦𝑦 0,0 +
1
𝑥 3 𝑓𝑥𝑥𝑥 0,0 + 3𝑥 2 𝑦𝑓𝑥𝑥𝑦 0,0 + 3𝑥𝑦 2 𝑓𝑥𝑦𝑦 0,0 + 𝑦 3 𝑓𝑦𝑦𝑦 0,0 + ⋯
3!
1 1
= 0 + 0 + 𝑦 + 2! 0 + 2𝑥𝑦 + 0 + 3! 0 + 3𝑥 2 𝑦 + 0 − 𝑦 3 + ⋯
𝑥2𝑦 𝑦3
=𝑦 + 𝑥𝑦 + − +⋯
2 6

5. Expand 𝒆𝒙𝒚 in the neighborhood of (𝟏, 𝟏)

Sol:
Let 𝑓 𝑥, 𝑦 = 𝑒 𝑥𝑦 Then Let 𝑓 1,1 = 𝑒
𝑓𝑥 𝑥, 𝑦 = 𝑦𝑒 𝑥𝑦 𝑓𝑥 1,1 = 𝑒
𝑓𝑦 𝑥, 𝑦 = 𝑥𝑒 𝑥𝑦 𝑓𝑦 1,1 = 𝑒
𝑓𝑥𝑥 𝑥, 𝑦 = 𝑦 2 𝑒 𝑥𝑦 𝑓𝑥𝑥 1,1 = 𝑒
𝑓𝑥𝑦 𝑥, 𝑦 = 𝑥𝑦𝑒 𝑥𝑦 + 𝑒 𝑥𝑦 𝑓𝑥𝑦 1,1 = 2𝑒
𝑓𝑦𝑦 𝑥, 𝑦 = 𝑥 2 𝑒 𝑥𝑦 𝑓𝑦𝑦 1,1 = 𝑒
By Taylors theorem
𝑓 𝑥, 𝑦 = 𝑓 1,1 + 𝑥 − 1 𝑓𝑥 1,1 + 𝑦 − 1 𝑓𝑦 1,1 +
1
(𝑥 − 1)2 𝑓𝑥𝑥 1,1 + 2 𝑥 − 1 (𝑦−1)𝑓𝑥𝑦 1,1 + (𝑦 − 1)2 𝑓𝑦𝑦 1,1 +…
2!
2
1 𝑒 𝑥−1 + 4𝑒 𝑥 − 1 𝑦 − 1 +
𝑒 𝑥𝑦 = 𝑒 + 𝑒 𝑥 − 1 + 𝑒 𝑦 − 1 + +…
2! 𝑒(𝑦 − 1)2
1
= 𝑒[1 + 𝑥 − 1 + 𝑦 − 1 + (𝑥 − 1)2 + 4 𝑥 − 1 (𝑦 − 1) + (𝑦 − 1)2 +…]
2!