School of Mathematics

MAT9004 Monash University

Applied Class Week 4 A/Prof Heiko Dietrich
Prof Nick Wormald

Question 4.1
How would you apply the methods given in lectures to establish the minimum value of the function
f (x) = 2x3 9x2 + 12x for x in the domain [2/3, 1)? (Hint: break the domain into two parts.)

S OLUTION : We have f 0 (x) = 6(x 1)(x 2). So stationary points are x = 1 and x = 2 and there
are no singular points. To classify these, use f 00 (x) = 12x 18. So x = 1 is a local max and x = 2 is
a local min. To deal with the unbounded domain, there are a couple of options for breaking the domain
into two intervals and using the results from lectures.

Alternative 1: break it so that f is convex on the unbounded interval:

(1i) [2/3, 3/2]. The only internal stationary point is x = 1 which is a local max. So we only need to
look at endpoints: f (2/3) = 4 1627 , f (3/2) = 9/2. So the min here is 9/2.
(1ii) (3/2, 1). Here f 00 > 0 so we know the local min is the global min on this interval. Since f (2) = 4,
this is also the global min on [2/3, 1).

Alternative 2: break it so that f is increasing on the unbounded interval:

(2i) [2/3, 3]. The only internal stationary point that’s a local min is x = 2. So we only need to look at
this point as well as the endpoints: f (2) = 4, f (2/3) = 4 16
27 , f (3) = 54 81 + 36 = 9. So the min in
this interval is 4.
(2ii) (3, 1). Here f 0 > 0 so we know the function is increasing on this interval. By the note on Slide 8,
Lecture 5, we know that f is increasing on [3, 1). Since f (3) = 9, all values of f on this interval are
bigger than 4.

Using either alternative method, we find the minimum value of f on the two intervals combined is 4,
occurring at x = 2.

Question 4.2
Let f (x) = 4 x2 .
a) Find an antiderivative of f .
b) Find the area of the region bounded by (i) the plot of f that lies above the x axis, and (ii) the x axis.
Produce a plot to visualise this region.

S OLUTION : a) A function F where F 0 = f is needed. By reverse engineering, differentiating 4x

gives 4. Similarly, differentiating x3 gives 3x2 , but we want x2 , so divide x3 by 3. That is, an
antiderivative is F (x) = 4x x3 /3.
b) The area can be calculated as follows:
Z 2
4 x2 dx = F (2) F ( 2) = 16/3 ( 16/3) = 32/3.
2

Your plot should look something like this:

Question 4.3
Rain starts falling on an empty swimming pool one hour after midday. At t hours after midday (for any
real t 1), the rate that rain water enters the pool is 128/t2 litres per hour and the water leaks out of the
pool at a rate of 2t litres per hour (until the pool is empty again).
a) Find the time when the maximum amount of water is in the pool.
b) Find an expression for the amount of water in the pool t hours after midday, that is valid for any
t 1 until the pool is again empty.
c) Find the amount of water in the pool, when it is at its maximum.
d) Show that the pool is not yet empty by 10pm on the same day.

S OLUTION : a) The net rate of increase of the amount of water is 128/t2 2t litres per hour, unless this
quantity is negative and there is no water in the pool. This rate of increase is positive when t = 1, so the
pool starts filling at this time. It is important to note that this formula for rate of increase will be valid
until the pool becomes empty again.
The rate is zero when t = 4. It is positive for t < 4 and negative for t > 4, hence there is a global
maximum at t = 4. (i.e. 4 hours after midday.) The amount in the pool must decrease until it becomes
empty, so this is indeed the maximum amount in the pool.
b) If A(t) is the amount at time t we have A0 (t) = 128/t2 2t. Since A is an antiderivative of this, we
have A(t) = 128/t t2 + C for some constant C. Plugging in A = 0 for t = 1 gives C = 129. So
the formula is 129 128/t t2 .
c) We have A(4) = 129 32 16 = 81 (litres).
d) We have A(10) = 129 12.8 100 = 29 12.8 > 0. We know the pool didn’t get empty before
this, since A0 (t) < 0 from t = 4 until t = 10, so the water level is dropping.

Question 4.4
Let f (x) = x3 3x2 + 2x, and g(x) = x2 2x.
a) Find the area of the bounded region A between the two plots of f and g. (Hint: first find where the
plots meet.)
b) Find the area of the bounded region B between the two plots of f + 100 and g + 100 directly.
c) What is the relation between the answers to (a) and (b)? Why should we have expected this? What
is an easy way to approach such a problem as (a) that does not involve looking where the functions
cross the x axis?
d) If you did not already do so, write the answer to (a) or to (b) as one single integral involving f and g.

S OLUTION : a) Step 1: We have to find where the two plots meet: We observe that f (x) = g(x) if and
only if x3 3x2 + 2x (x2 2x) = 0, that is, if and only if x3 4x2 + 4x = 0. Factorising the left
hand side as x(x 2)2 , we deduce the two plots meet only at x = 0 and x = 2. This means that the
bounded region lies between x = 0 and x = 2.
Step 2: which function is ‘on top’ for 0 < x < 2? Here, the sign of f (x) g(x) = x(x 2)2 is positive.
So f (x) > g(x) for x 2 (0, 2).
Step 3: where do f and g cross the x axis? Well factorising f (x) gives x(x 1)(x 2) so it crosses the
x axis at the points of intersection with g(x), as well as x = 1. On the other hand g(x) = x(x 2) so
does not cross in between.
We can see f (x) > 0 for x 2 (0, 1) and f (x) < 0 for x 2 (1, 2) which means the picture looks like this:

Step 4. Break the area up into pieces that are represented in a simple way by integrals.

The symbols A, B, C, D represent the areas of the regions indicated. The area we want is A+B +C D.
R1 R1
Now A = 0 f (x) dx, B = 0 g(x) dx (since areas below the axis count negatively for integrals), C
R2 R2
= 1 g(x) dx, D = 1 f (x) dx. So the total area between the graphs is
Z 1 Z 1 Z 2 Z 2
f (x) dx g(x) dx g(x) dx + f (x) dx.
0 0 1 1
We can rewrite this as
Z 2 Z 2 Z 2
f (x) dx g(x) dx = (f (x) g(x)) dx
0 0 0
and evaluate it as
Z 2 Z 2
(f (x) g(x)) dx = (x3 4x2 + 4x) dx
0 0
1 4 1 4
= ( 24 ⇥ 23 + 2 ⇥ 22 ) ( 04 ⇥ 03 + 2 ⇥ 02 )
4 3 4 3
4
= .
3
b) Steps 1 and 2 are similar to the above: the plots meet at x = 0 and x = 2, and f (x) + 3 > g(x) + 3
for all x in between 0 and 2.
For step 3, it’s obvious that both f + 100 and g + 100 are positive on [0, 2] and hence the area is the area
under f + 100 minus the area under g + 100, i.e.
Z 2 Z 2
(f (x) + 100) dx (g(x) + 100) dx.
0 0
We can manipulate this by using the rules of differences of integrals to get
Z 2
(f (x) g(x)) dx.
0
So it gives the same answer as (a).
c) They give the same answer.
We could have expected this since the region in (b) is the same as the region in (a) except it has been
shifted by adding 100 to the y coordinate.
The conclusion of this discussion is the following useful fact:

If f (x) g(x) on an interval [a, b], the area between the graphs of f and g on
the interval [a, b] is just
Z b
(f (x) g(x)) dx.
a

R2
d) We did this in our solution to (a): 0 (f (x) g(x)) dx.

Problems for revision at home (Week 4)

Question 4.5
p
Let c be a constant and let f (x) = c x.
a) Find the area between the graph of f and the x axis, for 0  x  c2 .
b) Explain why f has an inverse function (for a suitable codomain), assuming the domain of f is [0, c2 ]
and find the inverse function of f .
Rc
c) Let g = f 1 . Find 0 g(x) dx.
d) Explain why you could tell the answers to (a) and (c) are equal without doing any antidifferentiation
calculations.

S OLUTION : a) For x 2 [0, c2 ], we see f (x) c c = 0. So the area equals

Z c2 Z c2 p
f (x) dx = (c x) dx.
0 0
R
Now (c x1/2 ) dx = cx 23 x3/2 + C where we use “+C” for the constant of integration to distinguish
it from the constant c involved in the question. So, by the Fundamental Theorem of Calculus, the area
equals
2 2 3/2
c ⇥ c2 (c ) = c3 /3.
3
p
b) One can argue that f is injective on that domain, since its derivative is 1/(2 x) which is strictly
negative so the function decreases on (0, c2 ). Then if we define the codomain of f to be the image of f ,
which is [0, c], of course f is surjective.
Another way to get the same conclusion is to observe that the image of f is [0, c] and we can solve
p
c x = y for x and get x = (c y)2 , which exists for all y in the interval [0, c] (which is the image of
f ). This method also finds the inverse function of f : g(y) = (c y)2 . (Or g(x) = (c x)2 .)
c) Z Z
c c
(c x)2 dx = (c2 2cx + x2 ) dx.
0 0
R 1 3
Now (c2 2cx + x2 ) dx = c2 x cx2 + 3x + C, so
Z c
1
g(x) dx = c2 ⇥ c c ⇥ c2 + c3 0 0 0 = c3 /3.
0 3
d) In (c) we computed the area under the graph of g(x) = (c x)2 for 0  x  c, which (if we change
x into y) is the area under the graph of g(y) = (c y)2 for 0  y  c. This is like looking at the graph
of f sideways (see the following picture), so it’s equal to the area under the graph of f for 0  x  c2 ,
as computed in (a).

Another answer that is virtually equivalent is that since f and g are inverse functions, their graphs are the
reflections of each other in the line y = x. But this needs more explanation: it doesn’t work for functions
p
like f (x) = x2 where f 1 (x) = x, since then
Z c Z c2
2 3
x dx = c /3, x1/2 dx = 2c3 /3.
0 0

Why are these not equal?

Question 4.6
A sphere of radius x has surface area 4⇡x2 and volume 43 ⇡x3 . (Notice that here area is the derivative of
volume. Neat eh? There’s a reason for it!) A cube of side q has surface area 6q 2 and volume q 3 . You
want to make a sphere and cube from solid gold with total surface area 4 cm2 . It’s expensive! How big
should the sphere be, to minimise the total volume?

S OLUTION : Set surface area equal to 4:

4⇡x2 + 6q 2 = 4.

Solve for q:
1p
q= 6 6⇡x2 .
3
Plug this into the formula for volume:
4 3 4 3 1
⇡x + q 3 = ⇡x + (6 6⇡x2 )3/2 .
3 3 27
This is what we want to minimise. Differentiate it wrt x:
4 2 2⇡x
x (6 6⇡x2 )1/2 .
⇡ 3
1 p
This derivative is equal to 0 only when x = 0 or x = p⇡+6 . We can set the domain equal to [0, 1/ ⇡]
since we need x 0 but also q must be a non-negative real number (and in particular the square root
p
function in the formula for q needs to be defined) which gives the upper bound 1/ ⇡]. So 0 counts as an
endpoint, and we only have to examine the values of the volume at those two endpoints and the stationary
1
point x = p⇡+6 .
The volume at x = 0 is 63/2 /27 ⇡ 0.54.
p p
At x = 1/ ⇡ the volume is 4/(3 ⇡) ⇡ 0.75.
1 4
At x = p⇡+6 the volume is 3p⇡+6 ⇡ 0.44.
So the sphere should have radius p1 (and the total volume of gold is p4 ).
⇡+6 3 ⇡+6