Determine The Centroid Of The Area Shown By Direct Integration

Staring at a shaded region bounded by curves, knowing you need to find its geometric center, but freezing on the very first step — that’s the universal centroid experience. Here's the thing — you memorized the formulas. On top of that, you know $\bar{x}$ and $\bar{y}$ exist. But when the problem says "by direct integration," the blank page stares back harder than the final exam.

Here’s the thing most textbooks won’t tell you: the calculus isn’t the hard part. Even so, get those right, and the math usually behaves. The hard part is the geometry decisions you make before* you write a single integral sign. Get them wrong, and you’re integrating nonsense for forty minutes.

What Is a Centroid (and Why "Direct Integration"?)

The centroid is the geometric center of a plane area. Also, if you cut the shape out of uniform cardboard, the centroid is the point where it balances perfectly on a pin. For simple shapes — rectangles, triangles, circles — you already know the answers. They’re in the back of every statics textbook.

Direct integration is what you reach for when the shape isn’t* in that table. Now, when the boundary is defined by $y = x^3$, or $y = \sin x$, or some messy combination of curves that doesn’t match a standard formula. It’s the general method. The one that works on anything*, provided you can set up the integrals.

The phrase "direct integration" just means you’re deriving the coordinates from the definition, using calculus, rather than decomposing the shape into known pieces (the composite method). On top of that, both are valid. Direct integration is the one that scales to arbitrary complexity.

The Core Formulas You'll Actually Use

You’ll see these written a dozen ways. Strip away the notation variations and they boil down to this:

$ \bar{x} = \frac{\int x , dA}{\int dA} \qquad \bar{y} = \frac{\int y , dA}{\int dA} $

The denominator is just the total area $A$. The numerators are the first moments of area about the y-axis and x-axis respectively.

That’s it. Plus, two ratios. The entire game is figuring out what $dA$ looks like and how to express $x$ and $y$ (the coordinates of the centroid of that* differential element) in terms of a single integration variable.

Setting Up the Problem: The Step-by-Step That Textbooks Skip

This is where the points live or die. Most worked examples jump straight to the integral. They don’t show the five minutes of staring at the sketch, deciding: vertical strips or horizontal? $dx$ or $dy$? Function of $x$ or function of $y$?

Picking Your Differential Element (dA)

You have two main choices: a vertical strip of width $dx$, or a horizontal strip of width $dy$.

Vertical strips ($dA = y , dx$) work when the top and bottom boundaries are single-valued functions of $x$. The top curve is $y_{top}(x)$, the bottom is $y_{bottom}(x)$. The height of the strip is $y_{top} - y_{bottom}$. The centroid of that thin rectangle sits at its midpoint: $x_{el} = x$, $y_{el} = \frac{y_{top} + y_{bottom}}{2}$.

Horizontal strips ($dA = x , dy$) work when the left and right boundaries are single-valued functions of $y$. The right curve is $x_{right}(y)$, the left is $x_{left}(y)$. The width is $x_{right} - x_{left}$. The element centroid: $x_{el} = \frac{x_{right} + x_{left}}{2}$, $y_{el} = y$.

How do you choose? Here's the thing — if the curves are given as $y = f(x)$ and the region is vertically simple (any vertical line crosses the boundary at most twice), vertical strips are usually cleaner. Look at the boundaries. If the curves are $x = g(y)$ or the region is horizontally simple, go horizontal.

Sometimes both work. One leads to a polynomial integral. Try both mentally. **Sketch it. Because of that, the other leads to a trig substitution you’d rather avoid. Pick the one that makes the algebra hurt less.

Expressing Everything in One Variable

This is the mechanical part that trips people up. Once you pick $dx$ or $dy$, every* quantity in the numerator integrals must be expressed in that same variable.

If you chose vertical strips ($dx$):

• $dA = (y_{top} - y_{bottom}) , dx$
• $x_{el} = x$ (already in terms of $x$)
• $y_{el} = \frac{y_{top}(x) + y_{bottom}(x)}{2}$

If you chose horizontal strips ($dy$):

• $dA = (x_{right} - x_{left}) , dy$
• $x_{el} = \frac{x_{right}(y) + x_{left}(y)}{2}$
• $y_{el} = y$ (already in terms of $y$)

Write these out explicitly before* you set up the integrals. That said, don’t do it in your head. The moment you substitute $y_{el}$ into $\int y_{el} , dA$, you’re multiplying two expressions in $x$. That multiplication is where sign errors and algebra mistakes hide.

For more on this topic, read our article on who is the cute person in the world or check out recent improvements in have increased the pace of globalization..

Determining the Limits

The limits of integration are the projection of the region onto your integration axis.

Vertical strips $\rightarrow$ limits are the minimum and maximum $x$-values of the region. Often these are intersection points of the bounding curves. Solve $y_{top}(x) = y_{bottom}(x)$ for $x$.

Horizontal strips $\rightarrow$ limits are the minimum and maximum $y$-values. Solve $x_{right}(y) = x_{left}(y)$ for $y$.

If the region has a corner where the boundary definition changes (e.This is normal. g.Still, , bounded by $y=x^2$ on the left and $y=2-x$ on the right, meeting at some point), you may need to split the integral into two parts with different expressions for $x_{right}$ or $y_{top}$. Don’t force a single integral where the geometry says there are two.

Walking Through a Classic Example (The Parabolic Spandrel)

Let’s make this concrete. Find the centroid of the area bounded by the x-axis, the line $x = b$, and the parabola $y = kx^2$. Still, (Assume $k>0$, $b>0$). This is the standard "spandrel" shape.

Step 1: Sketch and decide. The region is vertically simple. Top curve: $y = kx^2$. Bottom curve: $y = 0$. Vertical strips are natural. Integration variable: $x$.

Step 2: Limits. $x$ runs from $0$ to $b$.

Step 3: Element geometry.

• Height of strip: $y_{top} - y_{bottom} = kx^2 - 0 = kx

Step 4 – Write the element quantities explicitly

For a vertical strip we have

[ dA=(y_{\text{top}}-y_{\text{bottom}}),dx=kx^{2},dx, \qquad x_{\text{el}}=x, \qquad y_{\text{el}}=\frac{y_{\text{top}}+y_{\text{bottom}}}{2} =\frac{kx^{2}+0}{2} =\frac{k}{2},x^{2}. ]

Notice that every symbol now depends only on the integration variable $x$, as required.

Step 5 – Set up the integrals for the three needed quantities

Area*

[ A=\int_{0}^{b} dA =\int_{0}^{b} kx^{2},dx . ]

[ A = \int_{0}^{b} kx^{2},dx = k\left[\frac{x^{3}}{3}\right]_{0}^{b} = \frac{k b^{3}}{3}. ]

Now the first moments.

Moment about the (y)-axis (for the (x)-coordinate of the centroid):

[ Q_{y} = \int x_{\text{el}},dA = \int_{0}^{b} x \cdot kx^{2},dx = k\int_{0}^{b} x^{3},dx = k\left[\frac{x^{4}}{4}\right]_{0}^{b} = \frac{k b^{4}}{4}. ]

Moment about the (x)-axis (for the (y)-coordinate of the centroid):

[ Q_{x} = \int y_{\text{el}},dA = \int_{0}^{b} \frac{k}{2}x^{2} \cdot kx^{2},dx = \frac{k^{2}}{2}\int_{0}^{b} x^{4},dx = \frac{k^{2}}{2}\left[\frac{x^{5}}{5}\right]_{0}^{b} = \frac{k^{2} b^{5}}{10}. ]

The centroid ((\bar{x}, \bar{y})) is then

[ \bar{x} = \

[ \bar{x}= \frac{Q_{y}}{A} =\frac{\displaystyle\frac{k,b^{4}}{4}}{\displaystyle\frac{k,b^{3}}{3}} =\frac{3}{4},b . ]

[ \bar{y}= \frac{Q_{x}}{A} =\frac{\displaystyle\frac{k^{2},b^{5}}{10}}{\displaystyle\frac{k,b^{3}}{3}} =\frac{3k}{10},b^{2}. ]

Hence the centroid of the parabolic spandrel bounded by the (x)-axis, the vertical line (x=b) and the curve (y=kx^{2}) is

[ \boxed{\bigl(\bar{x},\bar{y}\bigr)=\left(\frac{3b}{4},;\frac{3k,b^{2}}{10}\right)} . ]

The point (\bigl(\frac{3b}{4},\frac{3k b^{2}}{10}\bigr)) lies a quarter‑way from the origin to the right‑hand side of the region and a little more than one‑third of the maximum height of the parabola, which matches the intuitive picture of the area’s “center of mass.”

This example illustrates the systematic approach: (1) describe the region with appropriate strips, (2) write the elemental area and its first moments, (3) integrate over the limits determined by the bounding curves, and (4) divide the resulting moments by the total area to obtain the centroid coordinates. Plus, the same workflow applies to any planar region, whether the strips are vertical, horizontal, or a combination of both, and even when the region must be split into sub‑regions. With practice, the method becomes a reliable tool for locating centroids in engineering, physics, and geometry.