More optimization (Calc 1; Lecture 2-10; Fall 26)

Overview

The lecture develops a repeatable method for calculus optimization: translate a word problem into a one-variable function, find and test critical points, then interpret the result in the requested units or dimensions. Worked examples maximize a chip index under an $18 budget, find the largest axis-aligned rectangle on a curve, minimize the material cost of a cylinder with volume 192π, and begin finding the point on y = x² nearest to (0, a).

Key takeaways

A reliable optimization workflow is to identify the objective, use a constraint to express it with one variable, apply derivative tests, and then translate the result back into the requested form and units.
With an $18 chip budget and index n²p, the constraint p = 18 - 3n produces the maximum at 4 nacho bags and 6 potato chip bags.
For the rectangle on y = 6/(x² + 2x + 9), comparing the candidates x = ±3 by physical dimensions selects a 3-by-1/2 rectangle.
Weighting the cylinder’s bottom cost by three and using volume 192π yields a least-cost design with radius 4 inches and height 12 inches.
For the point on y = x² nearest to (0, a), the nearest point is the origin when a ≤ 1/2; when a > 1/2, two symmetric nearest points occur at x = ±√(a - 1/2).
When minimizing a nonnegative distance, minimizing its square gives the same minimizing point and can simplify differentiation by removing the square root.

Chapters

0:00 Logarithm Rules Prepare for Logarithmic Differentiation

The quick review expands a logarithmic expression using product, quotient, and power rules.
The example becomes 3x + x ln(1 + x²) - 1/2 ln(x - 1).
Comfort with logarithm properties is identified as preparation for logarithmic differentiation.

2:02 Critical Points, Local Extrema, and an Absolute Maximum

For f(x) = (x² + 4x)/(x² + 9), the derivative has critical points x = -3/2 and x = 6; the denominator is never zero.
A sign chart shows the function decreases, increases, then decreases, so x = 6 is a local maximum with value 4/3.
The horizontal asymptote is y = 1, which helps support that 4/3 is the absolute maximum despite the domain not being closed and bounded.

9:01 Maximizing the Nacho and Potato Chip Index

The party index is I = n²p, with nacho bags costing $3, potato chip bags costing $1, and an $18 budget.
The budget constraint gives p = 18 - 3n, reducing the objective to I(n) = 18n² - 3n³.
Solving I'(n) = 9n(4 - n) gives candidates n = 0 and n = 4; the zero-nacho option has index zero.
The maximizing purchase is 4 nacho bags and 6 potato chip bags.

16:43 Three Stages of Optimization and Ron Graham’s Time Advice

The lecture separates optimization into pre-processing, calculus processing, and post-processing.
Pre-processing translates a word problem into a one-variable objective; processing finds and tests critical points and relevant endpoints.
Post-processing checks whether the answer needs to be a parameter, an optimized value, or a quantity with units.
Ron Graham’s advice is to make good use of the 24 hours available each day without overworking to the point of burnout.

20:29 Modeling the Largest Rectangle on a Rational Curve

The rectangle has one corner at the origin, sides parallel to the axes, and an opposite corner on y = 6/(x² + 2x + 9).
Its signed area model is A(x) = xy = 6x/(x² + 2x + 9), reducing the problem to one variable.
Negative x values are allowed, so the signed expression can be negative even though physical area depends on the magnitude of the base.
As |x| grows, the area approaches zero, supporting the expectation of a finite maximum.

26:03 Finding the Rectangle Dimensions at x = ±3

The quotient rule simplifies the derivative to A'(x) = -6(x² - 9)/(x² + 2x + 9)².
The only critical points are x = 3 and x = -3 because the denominator does not vanish.
At x = 3, the height is 1/4; at x = -3, the height is 1/2.
Comparing physical dimensions gives the largest rectangle as 3 by 1/2.

31:03 Building a Cost Function for an Open-Topped Cylinder

The container has radius r, height h, no top, and required volume 192π cubic inches.
The bottom material costs three times as much as the side material, so the relative cost is C = 3πr² + 2πrh.
The side area is circumference 2πr multiplied by height h; the bottom area is πr².
Using πr²h = 192π gives h = 192/r² and reduces cost to C(r) = 3πr² + 384π/r.

38:02 Minimizing Cylinder Cost Gives a 4-by-12-Inch Design

Differentiating gives C'(r) = 6πr - 384π/r².
Setting C'(r) = 0 and multiplying by r² yields r³ = 64, so the positive radius is 4 inches.
Substituting r = 4 into h = 192/r² gives a height of 12 inches.
The least-cost cylinder under the stated material pricing and volume constraint has radius 4 inches and height 12 inches.

42:01 Converting Nearest-Point Distance to a One-Variable Objective

A point on the parabola y = x² can be written as (x, x²), while the fixed target is (0, a).
The distance is √[x² + (x² - a)²], which is minimized by minimizing its square instead.
The squared-distance objective is f(x) = x² + (x² - a)², avoiding a square root while preserving the minimizing point.
The derivative is f'(x) = 2x[1 + 2x² - 2a], producing candidates x = 0 and, when a ≥ 1/2, x = ±√(a - 1/2).

46:42 Nearest Points Depend on Whether a Exceeds One-Half

When a < 1/2, the nonzero critical-point formula has no real solutions, leaving (0, 0) as the nearest point on the parabola.
At a = 1/2, the nonzero candidates merge with x = 0; for a > 1/2, two symmetric candidates appear.
For a > 1/2, the squared-distance derivative has the correct sign pattern negative, positive, negative, positive across the ordered critical points.
The two minima are at x = ±√(a - 1/2), giving nearest points (±√(a - 1/2), a - 1/2); the lecture cuts off during the sign-chart discussion.

Summary, takeaways, and chapters were generated by AI from the video's transcript and may contain errors. The video belongs to its creator, Beard Meets Calculus.