Optimization Week 8

Content Standards

The learners demonstrate understanding of basic concepts of derivatives.

Performance Standards

The learner shall be able to formulate and solve accurately situational problems involving extreme
values.

Most Essential Learning Competencies

At the end of the lesson, you should be able to:

• Solve optimization problems.

Lesson Presentation
Terminologies

• Mathematical modeling - is the art of translating problems from an application area into tractable
mathematical formulations whose theoretical and numerical analysis provides insight, answers, and
guidance useful for the originating application.
• Critical point - is an interior point in the domain of a function at which 𝑓 ′ (𝑥) = 0 or f ' does not exist.
So, the only possible candidates for the X-coordinate of an extreme point are the critical points and the
endpoints.
• Optimization - consists of maximizing or minimizing a real function by systematically choosing input
values from within an allowed set and computing the value of the function.

Basic Calculus Page 1 of 20

Optimization Week 8

Many problems consist of “optimizing” a function that is, finding its maximum values. For example, you may
want to maximize your profit or minimize the time required to do a task. If you could express your happiness
as a function, you would want to maximize it. One of the principal uses of calculus is that it provides a very
general technique for optimization.

We will concentrate on applications of optimization. Accordingly, we will optimize continuous functions that
are defined on closed intervals (that is, intervals containing both endpoints,𝑎 ≤ 𝑥 ≤ 𝑏), or that have only
one critical value in an interval. Most applications fall into these two categories and the wide range of
examples and exercises in these sections will demonstrate the power of these techniques.

Absolute Extreme Values

The absolute maximum value of a function is the largest value of the function on its domain. Similarly, the
absolute minimum value of a function is the smallest value of the function on its domain. On a graph we set
𝑦 = 𝑓(𝑥), so the maximum and minimum values of 𝑓(𝑥) are the y-coordinates of the highest and lowest

To find the absolute extreme values of a continuous function 𝑓(𝑥)on𝑎 ≤ 𝑥 ≤ 𝑏:

a. Find all critical values of 𝑓(𝑥) inthe domain[𝑎, 𝑏].

b. Evaluate the function at these critical values and at the endpoints 𝑥 = 𝑎 and𝑥 = 𝑏

The largest and smallest of the values found in step (𝑏)are the absolute maximum and absolute minimum
values of 𝑓(𝑥)on𝑎 ≤ 𝑥 ≤ 𝑏.
points. The term absolute extreme values will mean the absolute maximum and absolute minimum values.

Example 1: Find the absolute extreme values of 𝑓(𝑥) = 𝑥 4 + 4𝑥 3 − 20𝑥 2 𝑜𝑛[−1, 4] Solution: The function is
continuous (since it is a polynomial), and the interval is closed, so both extreme values exist.

Basic Calculus Page 2 of 20

Optimization Week 8

Differentiating to find critical values factoring

from the 4𝑥

𝑓′ (𝑥) = 4𝑥 3 + 12𝑥 2 − 40𝑥 from (𝑥 + 5)

= 4𝑥(𝑥 2 + 3𝑥 − 10) from (𝑥 − 2)

= 4𝑥(𝑥 + 5)(𝑥 − 2)

𝑥=0
Critical Value {𝑥 = −5
𝑥=2

The problem states the domain as [−1, 4], so we eliminate the critical value 𝑥 = −5 since it is not in the
domain. We evaluate 𝑓(𝑥) = 𝑥 4 + 4𝑥 3 − 20𝑥 2 at the remaining critical values (CV) and at the endpoints
(EP).

𝑥 = 0 𝑓(0) = 0
C𝑉{
𝑥 = 2 𝑓(2) = 16 + 4 ∙ 8 − 20 ∙ 4 = −32 ← 𝑠𝑚𝑎𝑙𝑙𝑒𝑠𝑡

𝑥 = −1 𝑓(−1) = 1 − 4 − 20 = −23
EP{
𝑥 = 4 𝑓(4) = 256 + 4 ∙ 64 − 20 ∙ 16 = 192 ← 𝑙𝑎𝑟𝑔𝑒𝑠𝑡

The largest (192) and the smallest (−32) of these resulting values are the maximum and minimum of the
function.

Answer: The maximum value of 𝑓(𝑥) is 192 (which occurs at 𝑥 = 4). The minimum value of 𝑓(𝑥) is −32
(which occurs at 𝑥 = 2)

Maximizing Profit

Many management problems involve maximizing profit. Such problems have only three economic ingredients.
The first is that profit equals revenue minus cost.

Basic Calculus Page 2 of 20

Optimization Week 8

Profit = Revenue - Cost

The second ingredient is that revenue equals price times quantity. For example, if a company sell 100 toasters
for $25 each, the revenue will obviously be 25∙ 100 = $2500.

Revenue = (unit price) ∙ (quantity)

The third economic ingredient reflects the fact that, in general, price and quantity are inversely related:
increasing the price decreases sales, while decreasing the price increases sales.

The equation relating the price p and the quantity x that consumers will demand at that price is called the
demand equation. If this equation is solved for price, the resulting function 𝑝(𝑥) is called the price function,
since it gives the price at which consumers will buy exactly x units of the product.

Price Functions

The price function 𝑝(𝑥) gives the price at which consumers will buy exactly x units of a product. In actual
practice, price functions are very difficult to determine, requiring extensive (and expensive) market research.
In this section we will be given the price function. In this lesson we will see how to construct them, at least in
simple cases.

Example 2: It cost the American Automobile Company $5000 to produce each automobile, and fixed costs
(rent and other costs that do not depend on the amount of production) are $20,000 per week. The company’s
price function is 𝑝 = 19,000 − 70𝑥, where p is the price at which exactly x cars will be sold.

a. How many cars should be produced each week to maximize profit?

b. For what price should they be sold?
c. What is the company’s maximum profit?

Basic Calculus Page 3 of 20

Optimization Week 8

Solution: Revenue is price times quantity, 𝑅 = 𝑝 ∙ 𝑥. To express the revenue function as a function of just
one variable, we use the price function 𝑝 = 19,000 − 70𝑥 to replace p by 19,000 − 70𝑥, giving.

𝑹 = 𝒑 ∙ 𝒙 = (𝟏𝟗, 𝟎𝟎𝟎 − 𝟕𝟎𝒙)𝒙 = 𝟏𝟗, 𝟎𝟎𝟎 − 𝟕𝟎𝒙𝟐 revenue function

(19,000 − 70𝑥)𝑥 → 𝑝(𝑥)

19,000 − 70𝑥 2 → 𝑅(𝑥)

Cost is the cost per car ($5000) times the number of cars (x) plus the fixed cost ($20,000).

𝐶(𝑥) = 5000𝑥 + 20, 000 (𝑢𝑛𝑖𝑡𝑐𝑜𝑠𝑡)(𝑞𝑢𝑎𝑛𝑡𝑖𝑡𝑦) + (𝑓𝑖𝑥𝑒𝑑𝑐𝑜𝑠𝑡𝑠)

Profit is revenue minus cost.

𝑃(𝑥) = (19,000 − 70𝑥 2 ) − (5000𝑥 + 20,000)

(19,000 − 70𝑥 2 ) → 𝑅(𝑥)

(5000𝑥 + 20,000) → 𝐶(𝑥)

= −70𝑥 2 + 14,000𝑥 − 20,000 𝑝𝑟𝑜𝑓𝑖𝑡𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛

a. We maximize the profit by setting its derivative equal to zero.

𝑃 ′ (𝑥) = −140𝑥 + 14,000 = 0 𝑑𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑡𝑖𝑎𝑡𝑖𝑛𝑔𝑃 = −= −70𝑥 2 + 14,000𝑥 − 20,000

−140𝑥 = −14,000 𝑠𝑜𝑙𝑣𝑖𝑛𝑔

−14,000
𝑥 = = 100 𝑜𝑛𝑙𝑦𝑜𝑛𝑒𝑐𝑟𝑖𝑡𝑖𝑐𝑎𝑙𝑣𝑎𝑙𝑢𝑒
−140

𝑃 ′′(𝑥) = −140 𝑡ℎ𝑒𝑠𝑒𝑐𝑜𝑛𝑑𝑑𝑒𝑟𝑖𝑣𝑎𝑡𝑖𝑣𝑒𝑖𝑠𝑛𝑒𝑔𝑎𝑡𝑖𝑣𝑒, 𝑠𝑜𝑝𝑟𝑜𝑓𝑖𝑡𝑖𝑠𝑚𝑎𝑥. 𝑎𝑡𝑡ℎ𝑒𝐶𝑉.

Basic Calculus Page 4 of 20

Optimization Week 8

(If the second derivative had involved x, we would have substituted the critical value 𝑥 = 100.) Since x is the
number of cars, the company should produce 100 cars per week (the time period stated in the problem).

b. The selling price p is found from the price function.

𝑝 = 19,000 − 70 ∙ 100 = $12,000 𝑝 = 19,000 − 70𝑥 𝑒𝑣𝑎𝑙𝑢𝑎𝑡𝑒𝑑 𝑎𝑡 𝑥 = 100

c. The maximum profit is found from the profit function.

𝑃(100) = −70(100) 2 + 14,000(100) − 20,000 = $680,000

Finally, state the answer clearly in words.

Answer: The company should make 100 cars per week and sell them for $12,000 each. The maximum profit
will be $680,000.

Lesson 7.2 Optimization Involving Area and Volume

Maximizing or minimizing some function relative to some set, often representing a range of choices available
in a certain situation. The function allows comparison of the different choice-s for determining which might be
“best.” Common applications: Minimal cost, maximal profit, minimal error, optimal design, optimal
management, variational principles.

A Classic Problem

1. You have 40 feet of fence to enclose a rectangular garden along the side of a barn. What is the
maximum area that you can enclose?
𝑤 = 𝑥 𝐴 = 10(40 − 2 ∙ 10)
𝑙 = 40 − 2𝑥 𝐴 = 10(20)
𝐴 = 200 𝑓𝑡 2
𝐴 = 𝑥(40 − 2𝑥)
𝐴 = 40𝑥 − 2𝑥2
𝐴′ = 40 − 40𝑥 𝑡ℎ𝑒𝑟𝑒𝑓𝑜𝑟𝑒,
0 = 40 − 4𝑥 𝑤 = 10𝑓𝑡𝑎𝑛𝑑𝑙 = 20𝑓𝑡.
4𝑥 = 40
𝑥 = 10
Basic Calculus Page 5 of 20
Optimization Week 8

40 – 2𝑥

To find the maximum (or minimum) value of a function:

1. Write it in terms of one variable.

2. Find the first derivative and set it equal to zero.
3. Check the end points if necessary.

We can minimize the material by minimizing the area.

𝐴 = 2𝜋𝑟 2 + 2𝜋𝑟ℎ

2𝜋𝑟 2 ℎ → 𝑎𝑟𝑒𝑎𝑜𝑓𝑒𝑛𝑑𝑠

2𝜋𝑟ℎ → 𝑙𝑎𝑡𝑒𝑟𝑎𝑙𝑎𝑟𝑒𝑎

Example 1: What dimensions for a one liter cylindrical can will use the least amount of material?
therefore,

1000
=ℎ
We need another equation that relates r and h: 𝜋𝑟 2

𝑉 = 𝜋𝑟 2 ℎ 1000
=ℎ
𝜋(5.42)2
(1 𝐿 = 1000 𝑐𝑚3 )
ℎ = 10.38

Basic Calculus Page 6 of 20

Optimization Week 8

1000 = 𝜋𝑟 2 ℎ

1000
=ℎ
𝜋𝑟 2

1000
𝐴 = 2𝜋𝑟 2 + 2𝜋𝑟 ∙
𝜋𝑟 2

2000
𝐴 = 2𝜋𝑟 2 +
𝑟

2000
𝐴′ = 4𝜋𝑟 −
𝑟2

2000
0 = 4𝜋𝑟 −
𝑟2

2000
= 4𝜋𝑟
𝑟2

2000 = 4𝜋𝑟 3

3 500
𝑟=√
𝜋

𝑟 = 5.42 𝑐𝑚

Example 2: A manufacturer wants to design an open box having a square base and a surface area of 108 in2.

Since the box has a square base, its volume is 𝑉=𝑥2ℎ

Note: We call this the primary equation because it gives a formula

for the quantity we wish to optimize.

The surface area=the area of the base + the area of the 4 sides.

Basic Calculus Page 7 of 20

Optimization Week 8

What dimensions will produce a box with maximum volume?

We want to maximize the volume, so express it as a function of just one

variable.

To do this, solve

𝑥 2 + 4𝑥ℎ = 108𝑓𝑜𝑟 ℎ.

Solutions:

𝑆. 𝐴. = 𝑥 2 + 4𝑥ℎ = 108

108−𝑥 2
ℎ= → Substitute this into the Volume equation.
4𝑥

108 − 𝑥 2 𝑥3
𝑉 = 𝑥2ℎ = 𝑥2 ( ) = 27𝑥 −
4𝑥 4

To maximize V, we find the derivative and it’s C.N.’s (critical numbers)

𝑑𝑣 3𝑥 2
= 27 − =0 3𝑥 2 = 108 Critical numbers: ±6
𝑑𝑥 4

We can conclude that V is a maximum when 𝑥 = 6 and the dimensions of the box are 6 𝑖𝑛 × 6 𝑖𝑛 × 3 𝑖𝑛

Example 3: A rectangular box is to be made from a piece of cardboard 24 cm long and 9 cm wide by cutting
out identical squares from the four corners and turning up the sides. Find the volume of the largest
rectangular box that can be formed.

Solution:

Let s be the length of the side of the squares to be cut out, and imagine the “flaps” being turned up to form
the box. The length, width and height of the box would then be 24 − 2𝑠, 9 − 2𝑠, and s, respectively.
Therefore, the volume of the box is 𝑉 (𝑠) = (24 − 2𝑠) (9 − 2𝑠) 𝑠 = 2(108𝑠 − 33𝑠 2 + 2𝑠 3 ).

Basic Calculus Page 8 of 20

Optimization Week 8

We wish to maximize 𝑉 (𝑠) but note that s should be

nonnegative and should not be more than half the width of
the cardboard. That is, 𝑠 ∈ [0, 4.5]. (The case 𝑠 = 0 or
𝑠 = 4.5 does not produce any box because one of the
dimensions would become zero; but to make the interval
closed and bounded, we can think of those cases as
degenerate boxes with zero volume). Since V is just a
polynomial, it is continuous on the closed and bounded
interval [0, 4.5]. Thus, the Extreme Value Theorem applies. Now 𝑉’(𝑠) = 216 − 132𝑠 + 12𝑠2 = 4(54 −
33𝑠 + 3𝑠2) = 4(3𝑠 − 6)(𝑠 − 9) and hence the only critical number of V in (0, 4.5) is 2 (s = 9 is outside
the interval).

s 0 2 4.5
V(s) 0 200 0
We now compare the functional values at the endpoints and at the critical points:

Therefore, from the table, we see that V attains its maximum at 𝑠 = 2, and the maximum volume is equal to
𝑉 (2) = 200 𝑐𝑚3 .

Basic Calculus Page 9 of 20

Optimization Week 8

Terminologies

✓ Chain rule - is a formula for computing the derivative of the composition of two or more functions.
✓ Higher Order Derivatives - The derivative is “the rate of change of function at a specified point”. The
derivative of the function 𝑓 (𝑥) with respect to x at the point 𝑥0 is the function 𝑓 ′ (𝑥0 ), which is given
below: The derivatives other than the first derivative are called the higher order derivatives.
✓ Composite function – The composition of two functions f and g is the function 𝑓(𝑔(𝑥)).
✓ Differentiation - process of finding the derivative, or rate of change, of a function.

Repeated Differentiation

We have seen that from one function we can calculate a new function, the derivative of the original function.
This new function, however, can itself be differentiated, giving what is called the second derivative of the
original function. Differentiating again gives the third derivative of the original function, and so on.

Higher –Order Derivatives

Consider the polynomial function 𝑓(𝑥) = 𝑥 4 + 3𝑥 3 − 4𝑥 + 2

𝑓 ′(𝑥) = 4𝑥3 + 9𝑥2 – 4

is again differentiable. Therefore, we may still differentiate 𝑓 ′ to get

(𝑓′ ) ′ (𝑥) = 𝑓"(𝑥) = 12𝑥 2 + 18𝑥.

This is again another differentiable function. We call f” and its subsequent derivatives the higher-order
derivatives off. Like f’, which we interpret as the slope of the tangent line, the function f” will have an
interpretation on the graph of f (This will be a lesson in college). It also has an important interpretation in
physics if f denotes the displacement of a particle. For now, this lesson will reinforce our skills in
differentiating functions.

Basic Calculus Page 10 of 20

Optimization Week 8

Example 1: From 𝑓(𝑥) = 𝑥 3 − 6𝑥 2 + 2𝑥 − 7 we calculate 𝑓’(𝑥):

𝑓 ′(𝑥) = 3𝑥 2 − 12𝑥 + 2 first derivative

Now this function can itself be differentiated, giving the second derivative of the original function, indicated by
two primes on the f,

𝑓"(𝑥) = 6𝑥 − 12 f double prime

Differentiating again gives the third derivative of the original,

𝑓′′′(𝑥) = 6 f triple prime

Differentiating again gives the fourth derivative,

𝑓 ′′′′(𝑥) = 0

All further derivatives of this function will, of course, be zero.

The prime notation rapidly becomes awkward for highest derivatives. For example, the
tenth derivative would require 10 primes.

Instead, we may replace the primes with the number of derivatives in parentheses. For
example, the tenth derivatives are written 𝑓 (10) (𝑥). The parentheses around the 10 are
important, for without the would mean an the tenth power instead of the tenth derivative.

1
Example 2: Calculate the first five derivatives of the rational function 𝑓(𝑥) = 𝑥.

Solution:

𝑓(𝑥) = 𝑥 −1 𝑓(𝑥) in exponential form

𝑓′ (𝑥) = −𝑥 −2 first derivative

𝑓′′(𝑥) = 2𝑥 −3 second derivative

Basic Calculus Page 11 of 20

Optimization Week 8

𝑓′′′(𝑥) = −6𝑥 −4 third derivative

𝑓 (4) (𝑥) = 24𝑥 −5 fourth derivative

𝑓 (5) (𝑥) = −120𝑥 −6 fifth derivative

Higher-Order Derivatives in Leibniz’s Notation

𝑑 𝑑 𝑑 𝑑2
In the 𝑑𝑥notation, the second derivative ( 𝑑𝑥 𝑓(𝑥)) is written 𝑓(𝑥).
𝑑𝑥 𝑑𝑥 2

The following table shows equivalent statements in the two notations:

Prime Notation d/dx Notation Derivatives

𝑑2
𝑓(𝑥)
𝑓′′(𝑥) 𝑑𝑥 2

𝑦′′ 𝑑2 Second derivative

𝑑𝑥 2

𝑑3
𝑓′′′(𝑥) 𝑓(𝑥)
𝑑𝑥 3

𝑦′′′ 𝑑3 Third derivatives

𝑑𝑥 3

𝑑𝑛
𝑓(𝑥)
𝑓 (𝑛) (𝑥) 𝑑𝑥 𝑛

𝑑𝑛 𝑛𝑡ℎ𝑑𝑒𝑟𝑖𝑣𝑎𝑡𝑖𝑣𝑒
𝑦 (𝑛) 𝑑𝑥 𝑛

Basic Calculus Page 12 of 20

Optimization Week 8

Note that in Leibniz’s notation the superscript comes after the d on the top and after the x on the bottom.
Calculating Higher derivatives merely require repeated use of the differentiation rules.

𝑑2 𝑥 2 +1
Example 3: Find 𝑑𝑥 2 ( )
𝑥

𝑑 𝑥 2 +1 𝑥(2𝑥)−(𝑥 2 +1)
Solution: ( )= the 1st derivative, using the quotient rule
𝑑𝑥 𝑥 𝑥2

2𝑥 2 −𝑥 2 −1 𝑥 2 +1
= 𝑥2
= 𝑥
simplifying

Differentiating this answer gives the second derivative of the original,

𝑑 𝑥 2 −1 𝑥 2 (2𝑥)−2𝑥(𝑥 2 +1)
( )= differentiating the first derivative
𝑑𝑥 𝑥 𝑥4

2𝑥 3 − 2𝑥 3 + 2𝑥 2𝑥 2
= = =
𝑥4 𝑥4 𝑥3

The function in the previous example was a quotient, so it was perhaps natural to use the quotient
rule. It is easier, however, to simply the original function first,

𝑥2 + 1 𝑥2 1
= + = 𝑥 + 𝑥 −1
𝑥 𝑥 𝑥

And then differentiate by the power rule. The first derivative of𝑥 + 𝑥 −1 is 1 − 𝑥 −2, and differentiating
again gives2𝑥 −3 , agreeing with the answer found by the quotient rule. (Always simplify before
differentiating)

1 1
Example 4: Find 𝑓 ′′ (4) if 𝑓(𝑥) =
√𝑥

Solution:

1
𝑓(𝑥) = 𝑥 2 𝑓(𝑥)in exponential form

Basic Calculus Page 13 of 20

Optimization Week 8
1
𝑓 ′ (𝑥) = − 2 𝑥 −3⁄2 differentiating once

3
𝑓 ′′′ (𝑥) = 4 𝑥 −5⁄2 differentiating again

5
1 3 1 −2 3 5
3 5 3 3 1
𝑓 ′′ (4) = ( ) = 4 (4)2 = 4 (√4) = 4 (2)5 = 4 (32) = 24 evaluating 𝑓′′ at 4
4 4

The Chain Rule and the Generalized Power Rule

In this lesson, we define composite functions (functions of functions) and learn the last of the general rules of
differentiation, the chain rule and the generalized power.

Composite Functions

Given a function 𝑓(𝑥) we found 𝑓(𝑥 + ℎ) simply by replacing each occurrence of 𝑥 by 𝑥 + ℎ. similarly, given
two functions 𝑓(𝑥) and 𝑔(𝑥), to find 𝑓 (𝑔(𝑥)) we take 𝑓(𝑥) and replace each occurrence of 𝑥 by 𝑔(𝑥). The
resulting function is called the composition * of 𝑓(𝑥) and 𝑔(𝑥).

Example 1: If𝑓(𝑥) = 𝑥 7 and𝑔(𝑥) = (𝑥 3 − 2), find the composition𝑓(𝑔(𝑥))

Solution:

𝑓(𝑔(𝑥)) = [𝑔(𝑥)]7 𝑓(𝑥) = 𝑥 7 with 𝑥 replaced by𝑔(𝑥)

= [𝑥 3 − 2𝑥]𝟕 using 𝑔(𝑥) = 𝑥 5

𝑥+1
Example 2: If (𝑥) = 𝑥−1and𝑔(𝑥) = 𝑥 5 , find 𝑔(𝑓)(𝑥)).

Solution:

𝑔(𝑥)+1 𝑥+1
𝑓(𝑔(𝑥)) = 𝑔(𝑥)−1 𝑓(𝑥) = 𝑥−1with 𝑥 replaced by 𝑔(𝑥)

𝑥 5 +1
= 𝑥 5 −1 using𝑔(𝑥) = 𝑥 5

Basic Calculus Page 14 of 20

Optimization Week 8

Example 3: I𝑓(𝑥) = √𝑥 and 𝑔(𝑥) = 2𝑥 + 1, find 𝑓(𝑔)(𝑥))and𝑔(𝑓)(𝑥)).

Solution:

𝑓(𝑔)(𝑥)) = √𝑔(𝑥) = √2𝑥 + 1 𝑓(𝑥) = √𝑥 with 𝑥 replaced by 𝑔(𝑥) = 2𝑥 + 1 𝑔(𝑓)(𝑥)) =

2𝑓(𝑥) + 1 = 2√𝑥 + 1

𝑔(𝑥) = 2𝑥 + 1with 𝑥 replaced by 𝑓(𝑥) = √𝑥

To emphasize that 𝑓(𝑔)(𝑥)) and𝑔(𝑓)(𝑥)) are not the same, we evaluate each at 𝑥 = 4:

𝑓(𝑔)(4)) = √2(4) + 1 = √9 = 3

𝑓(𝑔)(4)) = 2√4 + 1 = 2(2) + 1 = 5

Example 4: Find 𝑓(𝑥) and 𝑔(𝑥) such that (𝑥 2 − 5𝑥 − 1)10 is the composition 𝑓(𝑔)(𝑥)).

Solution: Think of(𝑥 2 − 5𝑥 − 1)10 as an inside function 𝑥 2 − 5𝑥 − 1and an outside function (𝑥)10.

Therefore, for (𝑥 2 − 5𝑥 − 1)10 = 𝑓(𝑔)(𝑥)), we take

𝑔(𝑥) = 𝑥 2 − 5𝑥 − 1

𝑓(𝑥) = (𝑥)10

Notice that expressing a function as a composition involves thinking of the function in terms of “blocks,” an
inside block on which a further operation is performed.

The Chain Rule

The chain rule shows how to differentiate a composite function 𝑓(𝑔)(𝑥)).

Basic Calculus Page 15 of 20

Optimization Week 8

Chain Rule

𝑑
𝑓(𝑔)(𝑥)) = 𝑓′(𝑔)(𝑥)) ∙ 𝑔′(𝑥)
𝑑𝑥
To differentiate 𝑓(𝑔)(𝑥)),differentiate 𝑓(𝑥),then replace each 𝑥 by 𝑔(𝑥),and finally, multiply
by the derivative of 𝑔(𝑥).

𝑑
Example 5: Use the chain rule to find 𝑑𝑥 (𝑥 2 − 5𝑥 − 1)10

Solution: In Example 4, we expressed (𝑥 2 − 5𝑥 − 1)10as 𝑓(𝑔)(𝑥)) with

𝑔(𝑥) = 𝑥 2 − 5𝑥 − 1

𝑓(𝑥) = (𝑥)10

From these,

𝑔′(𝑥) = 2𝑥 − 5

𝑓 ′ (𝑥) = 10𝑥 9

so

9 𝑑 2
𝑓′(𝑔)(𝑥)) = 10(𝑔(𝑥)) = 10 (𝑥 − 5𝑥 − 1)9
𝑑𝑥

Substituting these into the chain rule

𝑑
𝑓(𝑔(𝑥)) = 𝑓 ′ (𝑔(𝑥))𝑔′(𝑥)
𝑑𝑥

gives

𝑑 2
(𝑥 − 5𝑥 − 1)10 = 10((𝑥 2 − 5𝑥 − 1)9 (2𝑥 − 5)
𝑑𝑥

Basic Calculus Page 16 of 20

Optimization Week 8

This result says that to differentiate(𝑥 2 − 5𝑥 − 1)10, we bring down the exponent 10, reduce the exponent to
9 (steps familiar from the power rule), and finally multiply by the derivative of the inside function

Generalized Power Rule

Example 5: leads to a general rule for differentiating a function to a power.

Generalized Power Rule

𝑑
[𝑔(𝑥)]𝑛 = 𝑛 ∙ [𝑔(𝑥)]𝑛−1 ∙ 𝑔′(𝑥)
𝑑𝑥

To differentiate a function to a power, bring down the power as a multiplier, reduce the
exponent by 1, and then multiply by the derivative of the inside function.

[provided, of course, that the derivative 𝑔 ′ (𝑥) exists].

The generalized power rule follows from the chain rule by reasoning like that of Example 5: The derivative of
𝑓(𝑥) = 𝑥 𝑛 𝑖𝑠𝑓 ′(𝑥) = 𝑛𝑥 𝑛−1, so

𝑑
𝑓(𝑔)(𝑥)) = 𝑓′(𝑔)(𝑥)) ∙ 𝑔′(𝑥) Chain rule
𝑑𝑥

𝑑
[𝑔(𝑥)]𝑛 = 𝑛 ∙ [𝑔(𝑥)]𝑛−1 . 𝑔′(𝑥) Generalized power rule.
𝑑𝑥

𝑑
Example 6: Find𝑑𝑥 √𝑥 4 − 3𝑥 3 − 4

Solution:

𝑑 4 1 1 1
(𝑥 − 3𝑥 3 − 4)2 = (𝑥 4 − 3𝑥 3 − 4)2 (4𝑥 3 − 9𝑥 2 )
𝑑𝑥 2

Think of the generalized power rule “from the outside in.” That is, first bring down the outer exponent and
then reduce it by 1, and only then multiply by the derivative of the inside function. Be careful – it is the original

Basic Calculus Page 17 of 20

Optimization Week 8

function (not differentiated) that is raised to the power n - 1. Only at the end do you multiply by the derivative
of the inside function.

𝑑 1 3
Example 7: Find𝑑𝑥 (𝑥 2 +1)

Solution: Writing the function as (𝑥 2 + 1)−3 gives

𝑑 2 −3 6𝑥
(𝑥 + 1) = −3(𝑥2 + 1)−4 (2𝑥) = −6𝑥(𝑥2 + 1)−4 = − 2
𝑑𝑥 (𝑥 + 1)−4

Conclusion/ Summary

There is no single, all-purpose procedure for solving word problems. You must think about the problem, draw a
picture if possible, and express the quantity to be maximized or minimized in terms of some appropriate variable.
With practice you can become good at it.

We have two procedures for optimizing continuous functions on intervals.

• If the function has only one critical value in the interval, we may use the second derivative test.
• If the interval is closed, we evaluate the function at all critical values and endpoints.
• The price function p(x) gives the price at which consumers will buy exactly x units of a product.
• Many management problems involve maximizing profit. Such problems have only three economic
ingredients. The first is that profit equals revenue minus cost.
✓ By simply repeating the process of differentiation we can calculate second, third and even higher
derivatives.
✓ To differentiate 𝑓(𝑔)(𝑥)), differentiate 𝑓(𝑥), then replace each 𝑥 by 𝑔(𝑥), and finally, multiply by the
derivative of 𝑔(𝑥).
✓ To differentiate a function to a power, bring down the power as a multiplier, reduce the exponent by 1,
and then multiply by the derivative of the inside function.

Basic Calculus Page 18 of 20

Optimization Week 8

Reference Link

✓ Calculus with applications to the Management, Social, Behavioral, and Biomedical Sciences,
Geoffrey C. Beresford

Basic Calculus Page 19 of 20