Access this content offline
Omniguides
Skip to the reading

Optimization

This original textbook reading covers part of the AP topic. Follow the AP course outline for complete topic instruction and practice.

The most practical use of a derivative

Consider that you are an engineer hired by a multi-billion dollar soda company. The company is holding a contest for manufacturing a new design for their cylindrical aluminum cans with a prize of $1 000 000. Since aluminum is a relatively expensive metal, your job is to minimize the use of this metal in manufacturing a can no larger than 20 cm high and no wider than 10 cm in diameter. Assume the can must hold 250 cm3 of soda, and the cost of aluminum is $0.001 per cm2, what should be the height and diameter for your award-winning aluminum can that would minimize the design costs?

People face problems like that every day. While those poor folks who did not learn calculus will most likely give up when they encounter a real-life problem like that, those of us who have mastered derivatives can find an optimum solution that will give us the right answer.

Optimization involves finding conditions under which a variable is either at its minimum or maximum. Hm... minimum... maximum... we already know how to do that! This should be easy. Optimization is used every day in large corporations as well as smaller businesses, as well as by you! It is used by doctors in deciding whether to put patients on pills or give them surgery, and by students to decide how much sleep they should get before studying for their exam.

Example Problem 6.2.1

You really want to win that contest! So you decide to start working on the highly efficient can that holds 250 cm3 of soda using the least amount of aluminum.

a) Find the dimensions of the cylindrical can.​
b) What is the cost to manufacture one can, assuming a sheet of aluminum costs $0.001 per cm
2?​
c) Is this design realistic? If not, what should the dimensions be, given that the radius can’t be wider than 10 cm and the height no longer than 20 cm?

a) You might be thinking that we are not given enough information. Let’s write down the two equations we know related to cylinders.

A=2πrh+2πr2

V=πr2h

We don’t know the area, radius, or height, but we do know the volume, which is 250 cm3 . In all optimization problems, there are always at least two of the following variables:

  1. The optimizing variable - This is the variable to be maximized or minimized. It could be profits, volume, area, perimeter, anything.

2. The constraint - This is our limiting variable. This is something that is fixed that will not change in the problem. Usually we are given the value of constraints. What is our constraint in this problem? Volume? Area? Radius? Height?

If you think Area is the constraint... *buzz* so sorry, but no. Area is the optimizing variable, because we want to use as little aluminum as possible. If you picked volume as the constraint, you would be correct, because we are given how many milliliters of soda each can should hold. This is not going to change. Remember, we want to minimize area, and keep the volume constant. What do you think the derivative of the area equation is going to be?

250=πr2h

We do not want to have to worry about two variables in our area equation, so let’s find h in terms of r by dividing both sides by πr2 .

250πr2=h

Now let’s plug this into the Area equation, then take the derivative.

A=2πr250πr2+2πr2

A=500r+2πr2​
Now we can find the derivative of Area and set it equal to zero because this is the quantity we want to minimize. Notice that by setting h in terms of r, we are only differentiating using 1 variable, and do not have to worry about implicit differentiation.

A′=-500r2+4πr

0=-500r2+4πr​
500r2=4πr​
r3=5004π​
r=39.8 cm33=3.4 cm​

 

Interactive 6.2.1 Aluminum can

This is a 3-D version of our can. I would drink out of that.

Loading the original 3D model…

Original preview and model Interactive 6.2.1 Aluminum can

This viewer rotates the original source mesh. Lighting and controls are a web adaptation.

Download original COLLADA model

The question asks us to find the dimensions of the can, so let’s knock out height and Area.

250π(3.4)2=h

h=6.9 cm​
A=2π(3.4)(6.9)+2π(3.4)2

​
A=220 cm2

  1. The cost of aluminum is given as a tenth of a penny per cm2 . Our can has an area of 220 cm2 . This is simple math.

cost per can=220 cm2⋅$0.001cm2=$0.22

  1. Reality check: See the following image of the proposed can (interactive 6.2.1).

Pop Quiz 6.2.1 Conceptualizing Optimization

Question 1 of 5 · 0 correct

You want to buy 1 acre of land with dimensions such that it will minimize the amount you spend on fencing. What is the optimizing variable, assuming the land is rectangular?

Select one answer.
Glossary and deeper reading · 2 refreshers

These editorial refreshers supplement the reading. Definitions and examples are added guidance; original source wording and recorded answers remain in their original context.

  • Derivative

    A derivative gives the instantaneous rate of change of a function with respect to its input. At a point, it is the finite limit of the difference quotient when that limit exists. For example, the derivative of x² is 2x, so the tangent slope at x=3 is 6. A derivative function collects those rates across the inputs where the function is differentiable.

    Read more: 2.2 · Derivative →

  • Optimization

    Optimization finds the greatest or least value of an objective while satisfying the problem’s constraints. Express the objective in suitable variables, determine the allowed domain, and compare all relevant candidates. For example, among rectangles with a fixed perimeter, writing area in terms of one side reduces the search to a one-variable problem. Endpoint and feasibility checks are part of the argument, not optional final details.

    Read more: 5.11 · Optimization →

Search lessons, terms and problems