Applications of Integration

⚡ Key Takeaways

The definite integral, originally conceived as the area under a curve, becomes a universal accumulation tool: it computes areas, volumes, arc lengths, surface areas, average values, work, fluid force, centers of mass, and more. Key result in words: any quantity that can be built up as a sum of infinitely many infinitesimal pieces equals a definite integral of the rate at which those pieces accumulate. Where it applies: geometry, physics, engineering, economics, probability. Who/when: rooted in Archimedes (c. 250 BCE), formalized by Newton and Leibniz (1660s–1680s), made rigorous by Riemann (1854).

💡 The Core Insight

Every application of integration is a single sentence repeated in different costumes:

Total quantity = limit of a sum of slices = an integral of (slice measure) over the relevant variable. The hard part is never the integration — it is correctly identifying the representative slice $dQ$ , the tiny contribution from one thin strip, shell, or interval. Once you can write $dQ = (\text{something})\,dx$ , the answer is $Q = \int_a^b dQ$ . Master the slice, and every formula below becomes a consequence rather than a thing to memorize.

DefinitionThe Definite Integral as Accumulation

For a continuous function $f$ on $[a,b]$ , the definite integral is the limit of Riemann sums:

1

\quad \int_a^b f(x)\,dx = \lim_{n\to\infty}\sum_{i=1}^{n} f(x_i^*)\,\Delta x, \qquad \Delta x = \frac{b-a}{n}

Derivation hint: partition $[a,b]$ into $n$ equal subintervals, sample $f$ at a point $x_i^*$ in each, sum the rectangle areas $f(x_i^*)\Delta x$ , and let the mesh shrink to zero. Here $f(x_i^*)\,\Delta x$ is the representative slice: the area of one thin rectangle. Every application replaces this slice with a different physical or geometric quantity.

TheoremThe Slicing Principle

If a total quantity $Q$ over $[a,b]$ can be approximated by a sum $\sum g(x_i^*)\,\Delta x$ , where $g(x_i^*)\,\Delta x$ is the contribution of the slice at $x_i^*$ with width $\Delta x$ , and the approximation error vanishes as $\Delta x\to 0$ , then

2

\quad Q = \int_a^b g(x)\,dx.

Derivation hint: this is the definition of the integral applied to the slice-contribution function $g$ ; rigor requires that the sum of approximation errors $\to 0$ , which holds when $g$ is continuous.

DefinitionArea Between Two Curves

The area enclosed between $y=f(x)$ (upper) and $y=g(x)$ (lower) over $[a,b]$ is

3

\quad A = \int_a^b \bigl[f(x)-g(x)\bigr]\,dx, \qquad f(x)\ge g(x)\text{ on }[a,b]

Derivation hint: the representative vertical slice is a rectangle of height $f(x)-g(x)$ and width $dx$ ; integrate the slice area. Symbols: $A$ = area (square units); $f-g$ = vertical gap; $dx$ = slice width.

DefinitionVolume by Slicing / Disks / Washers

If a solid has known cross-sectional area $A(x)$ perpendicular to the $x$ -axis, its volume is

4

\quad V = \int_a^b A(x)\,dx.

Derivation hint: the slice is a slab of cross-section $A(x)$ and thickness $dx$ , with volume $A(x)\,dx$ ; sum the slabs. For a region rotated about the $x$ -axis (disk method), $A(x)=\pi[f(x)]^2$ :

5

\quad V = \pi\int_a^b [f(x)]^2\,dx.

Derivation hint: each slice is a disk of radius $f(x)$ , area $\pi r^2$ . For rotation between two radii $R(x)$ (outer) and $r(x)$ (inner) — the washer method:

6

\quad V = \pi\int_a^b \bigl([R(x)]^2-[r(x)]^2\bigr)\,dx.

Derivation hint: subtract the hole; slice area is the difference of two circles.

DefinitionVolume by Cylindrical Shells

Rotating the region under $y=f(x)$ about the $y$ -axis, the shell method gives

7

\quad V = 2\pi\int_a^b x\,f(x)\,dx.

Derivation hint: the slice is a thin cylindrical shell of radius $x$ , height $f(x)$ , thickness $dx$ ; unroll it into a rectangular sheet of area $2\pi x\cdot f(x)$ and thickness $dx$ , volume $2\pi x f(x)\,dx$ . Symbols: $2\pi x$ = circumference of the shell; $f(x)$ = shell height; $dx$ = shell thickness.

DefinitionArc Length

The length of the curve $y=f(x)$ from $x=a$ to $x=b$ is

8

\quad L = \int_a^b \sqrt{1+\bigl[f'(x)\bigr]^2}\,dx.

Derivation hint: a tiny piece of curve is the hypotenuse $ds=\sqrt{dx^2+dy^2}=\sqrt{1+(dy/dx)^2}\,dx$ ; integrate $ds$ . Symbols: $L$ = length; $ds$ = arc element; $f'(x)=dy/dx$ = slope.

DefinitionSurface Area of Revolution

Rotating $y=f(x)$ about the $x$ -axis produces a surface of area

9

\quad S = 2\pi\int_a^b f(x)\sqrt{1+\bigl[f'(x)\bigr]^2}\,dx.

Derivation hint: the slice is a frustum band of radius $f(x)$ and slant width $ds$ ; its area is circumference $\times$ slant width $=2\pi f(x)\,ds$ .

DefinitionAverage Value of a Function

The mean value of $f$ over $[a,b]$ is

10

\quad \bar{f} = \frac{1}{b-a}\int_a^b f(x)\,dx.

Derivation hint: average of $n$ samples is $\frac{1}{n}\sum f(x_i^*)$ ; multiply and divide by $\Delta x=(b-a)/n$ to turn the sum into an integral.

DefinitionWork Done by a Variable Force

For a force $F(x)$ acting along a line from $x=a$ to $x=b$ ,

11

\quad W = \int_a^b F(x)\,dx.

Derivation hint: over a tiny displacement $dx$ the force is essentially constant, so $dW=F(x)\,dx$ ; sum the increments. Symbols: $W$ = work (joules); $F(x)$ = force (newtons); $dx$ = displacement (meters).

DefinitionCenter of Mass / Centroid

For a thin plate of constant density between $y=f(x)\ge 0$ and the $x$ -axis over $[a,b]$ , the centroid $(\bar x,\bar y)$ is

12

\quad \bar x = \frac{\displaystyle\int_a^b x\,f(x)\,dx}{\displaystyle\int_a^b f(x)\,dx},\qquad \bar y = \frac{\displaystyle\int_a^b \tfrac12[f(x)]^2\,dx}{\displaystyle\int_a^b f(x)\,dx}.

Derivation hint: $\bar x = (\text{first moment about } y\text{-axis})/(\text{total area})$ ; the slice has area $f(x)\,dx$ , its own centroid at height $f(x)/2$ , giving moments $x f(x)\,dx$ and $\tfrac12 f^2 dx$ .

ProofWhy the Shell Formula Works

We derive equation [7], the most error-prone of the volume methods, from the Slicing Principle. 1. Set up the region. Take the region under $y=f(x)$ over $[a,b]$ with $a\ge 0$ , revolved about the $y$ -axis. 2. Choose the slice. Partition $[a,b]$ and consider the thin vertical strip at $x_i^*$ of width $\Delta x$ . When revolved about the $y$ -axis it sweeps out a thin cylindrical shell, not a disk. 3. Approximate one shell. The shell has inner radius $\approx x_i^*$ , height $f(x_i^*)$ , and thickness $\Delta x$ . Cut the shell vertically and unroll it: it becomes (approximately) a flat rectangular slab. 4. Compute the slab volume. The slab has length = circumference $2\pi x_i^*$ , height $f(x_i^*)$ , and thickness $\Delta x$ :

\Delta V_i \approx 2\pi x_i^*\,f(x_i^*)\,\Delta x.

5. Sum and take the limit. Total volume $V=\lim_{n\to\infty}\sum_{i=1}^n 2\pi x_i^* f(x_i^*)\Delta x$ . By the definition of the integral [1], this sum converges to

V = 2\pi\int_a^b x\,f(x)\,dx.

6. Error vanishes. The "unrolling" error is second order in $\Delta x$ (the shell is slightly wedge-shaped), so summed errors $\to 0$ . $\blacksquare$

∎

Figure

Three slicing strategies: vertical strip (area), disk, and cylindrical shell

mathematician · Bernhard Riemann | 1826–1866

Riemann gave the definite integral its first fully rigorous definition in his 1854 habilitation lecture, replacing the intuitive "area under a curve" with the precise limit of sums in equation [1]. This made the applications trustworthy: before Riemann, formulas for volume and work rested on geometric hand-waving. After him, each application became a theorem — a limit of sums that provably converges. Though Newton and Leibniz invented the calculus of applications, it was Riemann who guaranteed the slices truly add up to the whole.

🧠 Intuition

Think of integration as paying with infinitely thin coins. To find the total area of a strangely shaped region, you can't measure it directly — but you can slice it into vertical ribbons so thin that each is essentially a rectangle whose area you know. Stack up all the ribbon-areas and you have the total. The deep move is recognizing that "total" almost never means addition of equal things. A car braking from highway speed covers more distance per second early on than late; pumping water from a deep tank takes more work for the bottom layers than the top. In each case the contribution changes as you move through the variable. Integration is the only honest way to sum varying contributions — it lets each slice be weighted by its own local value before summing. A practical reading aid: whenever a problem says total, accumulated, net, or how much altogether, and the rate or density varies, you are almost certainly looking at an integral. Ask "what is the contribution of one thin slice?" — write it as $(\,\cdots\,)dx$ — and integrate.

ExampleVolume of a Paraboloid by Disks

Given: Rotate the region under $y=\sqrt{x}$ from $x=0$ to $x=4$ about the $x$ -axis. Find the volume.

Steps: 1. The slice is a disk of radius $r=y=\sqrt{x}$ , so $A(x)=\pi(\sqrt{x})^2=\pi x$ . 2. Apply the disk formula [5]:

V=\pi\int_0^4 (\sqrt{x})^2\,dx = \pi\int_0^4 x\,dx.

3. Integrate: $\displaystyle\pi\left[\frac{x^2}{2}\right]_0^4 = \pi\cdot\frac{16}{2}=8\pi.$

Answer:

\boxed{V = 8\pi \approx 25.13 \text{ cubic units}}

ExampleWork to Pump Water from a Tank

Given: A cylindrical tank of radius $2$ m and height $5$ m is full of water (density $\rho=1000\ \text{kg/m}^3$ , $g=9.8\ \text{m/s}^2$ ). Find the work to pump all the water out over the top rim.

Steps: 1. Let $x$ = depth below the rim ( $x=0$ at top, $x=5$ at bottom). A horizontal slab at depth $x$ has thickness $dx$ and volume $dV=\pi(2)^2\,dx=4\pi\,dx$ . 2. Its weight is $dF=\rho g\,dV = 1000\cdot 9.8\cdot 4\pi\,dx = 39200\pi\,dx$ newtons. 3. That slab must be lifted a distance $x$ to reach the rim, so $dW=x\,dF=39200\pi\,x\,dx$ . 4. Apply [11] (integrating the work increments):

W=\int_0^5 39200\pi\,x\,dx = 39200\pi\left[\frac{x^2}{2}\right]_0^5 = 39200\pi\cdot\frac{25}{2}.

5. Compute: $W=490000\pi \approx 1.539\times 10^6$ J.

Answer:

\boxed{W \approx 1.54\times 10^{6}\ \text{J} \;(\approx 1.54\ \text{MJ})}

⚠️ Common mistake

The single most common error is integrating with respect to the wrong variable, or forgetting the geometry of the slice.

Disk vs. shell confusion. If you rotate about the $y$ -axis but slice vertically, you get shells (use [7]); if you slice horizontally, you get disks in $y$ (use [5] with roles of $x,y$ swapped). Mixing the slice direction with the wrong formula is the #1 mistake.
Forgetting to square the radius in disk/washer volumes, or writing $(R-r)^2$ instead of $R^2-r^2$ in washers — these are not equal.
Dropping the $\sqrt{1+(f')^2}$ in arc length and surface area; the slice is a slanted hypotenuse, never just $dx$ .
Sign errors in area between curves: always integrate (upper − lower). If the curves cross, you must split the integral at the intersection points and take each region's positive contribution.

CounterexampleWasher ≠ difference of disks squared

Suppose you rotate the region between $R(x)=3$ and $r(x)=1$ about the $x$ -axis over $[0,1]$ . The correct washer volume [6] is

\pi\int_0^1 (3^2-1^2)\,dx = \pi(9-1)=8\pi.

A tempting wrong approach writes the slice as $\pi(R-r)^2=\pi(3-1)^2=4\pi$ , giving $4\pi$ . This treats the washer as a solid disk of radius $R-r=2$ — but a washer is an annulus, not a small disk. The true area is $\pi R^2-\pi r^2=8\pi$ , exactly double the wrong answer. The lesson: subtract the squared radii, never square the difference.

connection

Applications of integration sit at the crossroads of geometry and analysis. They depend entirely on the Fundamental Theorem of Calculus, which turns the limit-of-sums in [1] into an antiderivative evaluation, making every formula here computable. The slicing logic is the discrete-to-continuous bridge formalized by Riemann Sums. Many applications require Techniques of Integration (substitution, parts, partial fractions) just to evaluate the resulting integral — e.g., arc length integrals often demand trigonometric substitution. The center-of-mass and average-value ideas generalize into Multiple Integrals when the region is two- or three-dimensional, and the work/force applications feed directly into Differential Equations when forces depend on position and velocity. The probability density interpretation links [10] to Continuous Probability Distributions, where total probability is an integral equal to 1.

🔬 Where this appears

⏳ History

- c. 250 BCE — Archimedes computes the area of a parabolic segment and the volume of a sphere by the "method of exhaustion," summing inscribed shapes — the conceptual ancestor of slicing.

- 1615 — Johannes Kepler writes Nova Stereometria, calculating volumes of wine barrels by treating them as stacks of thin disks; a practical birth of integration.

- 1635 — Bonaventura Cavalieri publishes his "method of indivisibles," asserting that solids with equal cross-sections have equal volumes (Cavalieri's principle), directly underlying eq. [4].

- 1660s–1680s — Newton and Leibniz independently develop the calculus, linking area (integration) to slope (differentiation) and creating systematic formulas for volume, length, and work.

- 1691 — Jacob Bernoulli and contemporaries derive arc length and surface-area integrals, equations [8] and [9], applying them to curves like the catenary.

- 1750s — Leonhard Euler systematizes integration techniques and applies them broadly to mechanics, including work and center-of-mass computations.

- 1823 — Augustin-Louis Cauchy gives the first careful treatment of the integral as a limit, sharpening when the application formulas are valid.

- 1854 — Bernhard Riemann publishes the rigorous Riemann integral (eq. [1]), guaranteeing convergence of the slicing sums for continuous integrands.

- 1902 — Henri Lebesgue introduces measure-theoretic integration, extending applications to functions too irregular for Riemann's approach and grounding modern probability.

open-problem

The slicing principle [2] silently assumes the approximation error in each slice is higher order than $\Delta x$ , so that errors vanish in the limit. For continuous, smooth integrands this is guaranteed — but the boundary cases are subtle and historically caused real confusion:

Surface area paradoxes. Naively "unrolling" a surface of revolution as a cylinder (using $2\pi f(x)\,dx$ instead of the slant element $2\pi f(x)\,ds$ ) gives the wrong answer, because the slant of the surface contributes at first order. The famous Schwarz lantern shows that inscribing a cylinder with triangles can make the approximate surface area diverge — surface area is genuinely more delicate than volume.
Improper applications. When a region extends to infinity (e.g. Gabriel's Horn, $y=1/x$ for $x\ge 1$ ) the volume integral [5] converges to $\pi$ while the surface-area integral [9] diverges — a finite-volume solid with infinite surface area. Such improper integrals require careful limit definitions.
Non-Riemann-integrable cases. If the cross-section or density function is too discontinuous, the Riemann formulation fails and one needs the Lebesgue integral. Determining exactly which physical accumulation problems require this remains a point of analytic care rather than a settled formula.

Formal Statement

Why It Holds / Derivation

Building Intuition

Worked Examples

Common Misconceptions

Connections

Real-World Applications

Historical Development

Subtleties & Limits

See Also