Week 12

Topic 1. Maximum and Minimum Values (Stewart, pp. 1046 – 1057)

1 Local extreme values

For one variable functions we already saw, how to determine the maximum and minimum values
with the help of derivatives. Here we will see, how partial derivatives are used to determine the
extreme values of a two variable function.

Definition 1. Consider the two variable function f .

• Point (a, b) is a local maximum and f (a, b) is an local maximum value, if

f (a, b) ≥ f (x, y), for all (x, y) around (a, b)

• Point (a, b) is a local minimum and f (a, b) is a local minimum value, if

f (a, b) ≤ f (x, y), for all (x, y) around (a, b)

Theorem 1. If f (x, y) has a local minimum or maximum at (a, b) and its (first) derivatives
exists at (a, b), then

fx (a, b) = 0 and fy (a, b) = 0

Definition 2. Point (a, b) is a critical point of f if the (first) partial derivatives of f at (a, b)
equals 0 or does not exist.

Notice that only at the critical points can f have a local minimum or maximum.
p
Example 1. Consider the cone z = x2 + y 2

From its graph, the cone has a local minimum at (0, 0). Further
x y
fx = p and fy = p which does not exist at (0, 0).
x2 + y 2 x2 + y 2
The cone is thus an example of how extreme values can occur at points where the partial derivatives
does not exist.
Example 2. Find the local minimum and maximum values of
f (x, y) = x2 − 2xy + 2y 2 + y + 3

1
Solution: First determine its critical points.
fx = 2x − 2y and fy = −2x + 4y + 1
Since these derivatives are defined for all (x, y), the critical points of f are those points, which
makes both fx and fy 0.
fx = 2x − 2y = 0 =⇒ x=y
fy = −2x + 4y + 1 = 0 =⇒ 0 = −2x + 4x + 1 = 2x + 1 =⇒ x = − 21 = y
So f ’s only critical point is − 12 , − 12 .

Further, notice that

f (x, y) = x2 − 2xy + 2y 2 + y + 3 = (x − y)2 − y 2 + 2y 2 = (x − y)2 + y 2 + y + 3

2
2
= (x − y)2 + y + 12 − 14 + 3 = (x − y)2 + y + 12 + 2 + 34

2
At − 12 , − 21 , both (x − y)2 and y + 12 equal 0. And also, since both these terms are never

negative, − 21 , − 21 is then a local minimum of f , and 2 + 34 is a local minimum value of f .

Remark. The second derivative test will be a direct way, to check whether a critical point is a
local extreme value. But sometimes the second derivative test is not applicable. In such cases, we
have to be creative as was done above.
Example 3. Find the local maximum and minimum values of
z = f (x, y) = y 2 − x2

Solution: Set
fx = −2x = 0 and fy = 2y = 0
Thus the only critical point is (0, 0). Since
f (x, 0) = −x2
(0, 0) is a local maximum of f in the x-direction. But since
f (0, y) = y 2
(0, 0) is a local minimum of f in the y-direction. Consequently (0, 0) is neither a local minimum
nor a local maximum of f . We can also observe this from the sketch:

y
x

2
We call such critical points “saddle points”.

Definition 3. A differentiable function f has an saddle point at the critical point (a, b), if
around (a, b) there are points (x, y) where f (x, y) > f (a, b) and for other points (x, y) around
(a, b), f (x, y) < f (a, b).
Stated differently, a saddle point is a critical point which is neither a local minimum nor a local
maximum.

2 Second derivative test for two variable functions

Theorem 2 (Second derivative test). Suppose that the second partial derivatives of f are
continuous around and at (a, b), and suppose that fx (a, b) = 0 and fy (a, b) = 0. Let

2
D = D(a, b) = fxx (a, b)fyy (a, b) − fxy (a, b)

(a) If D > 0 and fxx (a, b) > 0, then f (a, b) is a local minimum.
(b) If D > 0 and fxx (a, b) < 0, then f (a, b) is a local maximum.
(c) If D(a, b) < 0, then (a, b) is a saddle point of f .

Remark.
• If D = 0, the test is inconclusive. The point (a, b) can be a local minimum/maximum or a
saddle point.
• We can remember the formula of D as
fxx fxy
D(x, y) =
fyx fyy

• If D > 0, we can also use the sign of fyy instead of fxx . That is, if D > 0, then

fxx > 0 ⇐⇒ fyy > 0 of fxx < 0 ⇐⇒ fyy < 0

Example 4. Consider f (x, y) = x2 − 2xy + 2y 2 + y + 3. In Example 2 we argued that − 12 , − 12

is a local minimum. Alternatively,

fx = 2x − 2y fxx = 2 fxy = −2
fy = −2x + 4y + 1 fyx = −2 fyy = 4

Thus, − 12 , − 12 is a local minimum of f by the

So D = 8 − (−2)(−2) = 4 > 0 and fxx > 0.
second derivative test.

3
Also, consider f (x, y) = y 2 − x2 . In Example 3 we argued that (0, 0) is a saddle point. Alter-
natively,

fx = −2 fxx = −2 fxy = 0
fy = 2y fyx = 0 fyy = 2

So D = −4 < 0. So by the second derivative test, (0, 0) is a saddle point.

Example 5. Determine all critical points of the function

z = f (x, y) = 6xy − x3 + y 3

and determine which of them are local maximums, local minimums or saddle points.

Solution: From the derivatives

fx = 6y − 3x2 fxx = −6x fxy = 6

2
fy = 6x + 3y fyy = 6y fyx = 6

follows that
−6x 6
D(x, y) = = −36xy − 36
6 6y

For the critical points, set fx = 0 and fy = 0:

6y − 3x2 = 0 =⇒ x2 = 2y
6x + 3y 2 = 0 =⇒ y 2 = −2x
x4 = 4y 2 = −8x x x3 + 8 = 0


=⇒ =⇒

So x = 0 of x = −2. We have

x=0 =⇒ y= 1
2 · 02 = 0
x = −2 =⇒ y= 1
2 · (−2)2 = 2

Thus the critical points are (0, 0) and (−2, 2). We have

Critical point D fxx Classification

(0, 0) − n.v.t. Saddle point
(−2, 2) + + Local minimum

Example 6. Consider the function

f (x, y) = 3x − x3 − 2y 2 + y 4

Determine all critical points and determine which of them are local maximums, local minimums or
saddle points.

4
Solution: The derivatives are

fx = 3 − 3x2 fxx = −6x fxy = 0

fy = −4y + 4y 3 fyy = −4 + 12y 2 fyx = 0

D(x, y) = −6x −4 + 12y 2

Thus
Critical points:

fx = 0 =⇒ x2 = 1 =⇒ x = ±1
−4y 1 − y 2 = 0


fy = 0 =⇒ =⇒ y = 0, ±1

There is thus 6 critical points:

(1, 0), (1, 1), (1, −1), (−1, 0), (−1, 1), (−1, −1)

Classification of the critical points

Critical point D fxx Classification

(−1, −1) 48 6 Local minimum
(−1, 0) −24 n.v.t. Saddle point
(−1, 1) 48 6 Local minimum
(1, −1) −48 n.v.t. Saddle point
(1, 0) 24 −6 Local maximum
(1, 1) −48 n.v.t. Saddle point

Example 7. Determine and classify all critical points of

f (x, y) = y sin x

Solution: The derivatives are

fx = y cos x fxx = −y sin x fxy = cos x

fy = sin x fyy = 0 fyx = cos x

Thus D(x, y) = − cos2 x

fx = 0 =⇒ y cos x = 0 =⇒ y = 0 of cos x = 0
fy = 0 =⇒ sin x = 0

The critical points are thus (nπ, 0), n ∈ Z (Why?)

Further, D(nπ, 0) = − cos2 (nπ) = −(±1)2 = −1 < 0
Thus all critical points are saddle points.

Exercise 1. Determine and classify the critical points of

5
 2 −y 2
(a) f (x, y) = x2 ye−x
2 −y 2
(b) f (x, y) = e4y−x

Answer:
(a) All the critical points and their classifications:
   
• 1, √12 , √12 e−3/2 and −1, √12 , √12 e−3/2 are local maximum points
   
• 1, − √12 , √12 e−3/2 and −1, − √12 , √12 e−3/2 are local minimum points

• (0, 0, 0) is a saddle point

• (0, y, 0) are local maximum points for y < 0
• (0, y, 0) are local minimum points for y > 0
(b) 0, 2, e4 is a local maximum.

3 Optimization
Example 8. A delivery firm only accepts rectangular containers whose length plus the circumfer-
ence of the cross section does not exceed 108 cm. Determine the measurements of an acceptable
container with the largest volume.

Solution:

y
x

Volume = V = xyz
Restrictions on x, y and z?

180 = length + circumference of the cross section = x + (2y + 2z) =⇒ x = 108 − 2y − 2z

Thus

V = yz(108 − 2y − 2z) = 180yz − 2y 2 z − 2yz 2

Further

Vy = 108z − 4yz − 2z 2 Vyy = −4z Vyz = 108 − 4y − 4z

2
Vz = 108y − 2y − 4yz Vzz = −4y Vzy = 108 − 4y − 4z

Thus D(y, z) = 16yz − (108 − 4y − 4z)2

Critical points:

6
 Vy = 0 Vz = 0
=⇒ 108z − 4yz − 2z 2 = 0 =⇒ 108y − 2y 2 − 4yz = 0
=⇒ −2z(z + 2y − 54) = 0 =⇒ −2y(z + 2z − 54) = 0
=⇒ y + 2z = 54 =⇒ z + 2y = 54
The solution of the system of equations

y + 2z = 54 and z + 2y = 54

is (y, z) = (18, 18), and thus (18, 18) is the only critical point of V (y, z).

D(18, 18) = 3(36)2 > 0 and Vyy (18, 18) = −4(18) − 72 < 0

Thus (18, 18) is a local maximum, and consequently the absolute maximum. So the measurements
for the largest volume are

y = 18 cm, z = 18 cm, x = 108 − 2(18) − 2(18) = 36 cm

4 Absolute maximum and minimum values

Definition 4. Consider a two variable function f .
• Point (a, b) is an absolute maximum and f (a, b) is an absolute maximum value, if

f (a, b) ≥ f (x, y), for all (x, y) in the domain of f

• Point (a, b) is an absolute minimum and f (a, b) is an absolute minimum value, if

f (a, b) ≤ f (x, y), for all (x, y) in the domain of f

Definition 5. A closed set S ⊆ R2 is a set with the property that if (a, b) is a point such that
any circle around (a, b) contains a point in S, then (a, b) ∈ S.
Stated differently, a closed set S is a set which contains its boundary. And boundary is defined as
all points (a, b) where any circle around (a, b) has a point in S and a point not in S. For example
closed sets non-closed sets

Definition 6. A set S ⊆ R2 is bounded if it can be contained within a circle.

Theorem 3 (Extreme Value Theorem for Functions of Two Variables). If f is continuous on

a closed, bounded set D ⊆ R2 , then f has an absolute maximum and minimum value on D.

7
According to this theorem, the expansion of the Closed Interval Method for two variable function
is as follows.

To determine the absolute maximum and minimum values of a continuous function f on a

closed, bounded set D:
(a) Find the values of f at the critical points of f in D.
(b) Find the extreme values of f on the boundary of D
(c) The largest amongst these values is the absolute maximum value and the smallest is the
absolute minimum value.

Example 9. Determine the absolute maximum and minimum values of

f (x, y) = x + y − xy

on the closed triangular region with vertices (0, 0), (0, 2) and (4, 0).

Solution: The triangular region is

(0, 2)

L2
L3
(1, 1)

(0, 0) L1 (4, 0)

The critical points of f :

fx = 1 − y = 0 and fy = 1 − x = 0

Thus (x, y) = (1, 1) is the only critical point of F . As indicated on the sketch, this critical point is
in the triangular region. We could also argue from the equation of L2 , that (1, 1) is in this region.
We have

f (1, 1) = 1 + 1 − 1 = 1

To determine the maximum and minimum values on the boundaries, consider the three lines sepa-
rately.
The equation of L1 is y = 0, where 0 ≤ x ≤ 4. So on L1

f (x, y) = f (x, 0) = x, 0≤x≤4

So the minimum and maximum values of f on L1 is

f (0, 0) = 0 and f (4, 0) = 4

8
The equation of L3 is x = 0, where 0 ≤ y ≤ 2. So on L3

f (x, y) = f (0, y) = y, 0≤y≤2

So the minimum and maximum values of f on L3 is

f (0, 0) = 0 and f (0, 2) = 2

The equation of L2 is y = − 21 x + 2, where 0 ≤ x ≤ 4. So on L2

f (x, y) = f x, − 12 x + 2 = x − 12 x + 2 − x − 12 x + 2 = 12 x2 − 3x + 4 = g(x),




0≤x≤4

The critical point of this parabola: g ′ (x) = x − 3

2 =0 =⇒ x= 3
2

An the critical points and the boundaries, the vaules of g:

g(4) = 4g − 23 = 21 49 − 92 + 4 = 9−18+16 7



g(0) = 2, 8 = 8

By comparing all these values, we get

• f (0, 0) = 0 is the absolute minimum of f
• f (4, 0) = 4 is the absolute maximum of f

Example 10. Determine the absolute maximum and minimum values of

f (x, y) = x2 + xy + y 2 − 6y op D = {(x, y) ∈ R2 | |x| ≤ 3, 0 ≤ y ≤ 5}

Solution: The region is

L3
(−3, 5) (3, 5)

(−2, 4)
L4 L2

(−3, 0) (3, 0)
L1

Critical points:

fx = 2x + y = 0 and fy = x + 2y − 6 = 0 =⇒ x + 2y = 6

The solution of this system of linear equations is (x, y) = (−2, 4), which is indeed in the region. At
this point f (−2, 4) = 4 − 8 + 16 − 24 = −12
Further, on L1 , y = 0 and −3 ≤ x ≤ 3,

f (x, y) = f (x, 0) = x2 , −3 ≤ x ≤ 3

So the maximum value on L1 is f (3, 0) = f (−3, 0) = 9 and the minimum value f (0, 0) = 0.

9
Further on L2 , x = 3 and 0 ≤ y ≤ 5,

3 2 27
f (x, y) = f (3, y) = 9 + 3y + y 2 − 6y = y 2 − 3y + 9 = y −


2 + ,0 ≤ y ≤ 5
4
So the minimum value on L2 is f 3, 32 = 27


4 . And the maximum value is then f (3, 5) = 19 or
f (3, 0) = 9.
Exercise: Confirm the following table:

Line Minimum Maximum

L1 f (0, 0) = 0 f (3, 0) = f (−3, 0) = 9
L2 f (3, 1.5) = 6.75 f (3, 5) = 19
L3 f (−2.5, 0) = −11.25 f (3, 5) = 19
L4 f (−3, 4.5) = −11.25 f (−3, 0) = 9

Consequently

f (−2, 4) = −12 absolute minimum

f (3, 5) = 19 absolute maximum

Example 11. Determine the absolute minimum and maximum of

f (x, y) = xy 2 op D = {(x, y) ∈ R2 | x2 + y 2 ≤ 3, x ≥ 0, y ≥ 0


Solution: Here the region is

√

0, 3
L2

L3

√

(0, 0) 3, 0
L1

The critical points:

fx = y 2 = 0 and fy = 2xy = 0

So (0, 0) is the only critical point, and f (0, 0) = 0.

On both L1 and L3

f (x, y) = (0, y) = 0 and f (x, y) = (x, 0) = 0

We can represent L2 as
√ √ π
x = 3 cos θ and y = 3 sin θ, where 0≤θ≤ 2

10
So on L2
√ √  √
f (x, y) = f 3 cos θ, 3 sin θ = 3 3 cos(θ) sin2 (θ) = g(θ), 0≤θ≤ π
2

We have g(0) = g(π/2) = 0. Further, for θ ̸= 0 and θ ̸= π2 ,

√ √ √
g ′ (θ) = −3 3 sin3 θ + 2 · 3 3 cos2 (θ) sin θ = 3 3 sin θ − sin2 θ + 2 cos2 θ = 0

=⇒ 0 = − sin2 θ + 2 cos2 θ (Why?)

2
= −1 + 3 cos θ
√
1 p 2
=⇒ cos θ = √ and sin θ = 1 − cos2 θ = √
3 3
So at the critical points of g,
√ √ 1 2
g(θ) = 3 3 cos(θ) sin2 (θ) = 3 3 · √ · = 2
3 3
Thus 0 is the absolute minimum value and 2 the absolute maximum value.

11