Calculus I: Optimization with Constraints

In engineering, resources are never infinite. You have a fixed budget, a limited amount of material, or specific physical boundaries. Constrained optimization is the mathematical engine that allows you to find the best possible outcome—maximum strength, minimum cost, optimal efficiency—within these real-world limits. Mastering this technique transforms abstract calculus into a powerful design tool for everything from minimizing material use in a can to maximizing the load on a beam.

From Word Problem to Mathematical Model

The first and most critical step is translating an engineering description into a precise mathematical framework. Every problem has two core components: the objective function and the constraint equation.

The objective function is the quantity you want to optimize, expressed as a function of one or more variables. You aim to either maximize it (e.g., profit, volume, strength) or minimize it (e.g., cost, surface area, energy loss). The constraint equation defines the limitation that ties your variables together, such as a fixed amount of material (total surface area), a fixed budget (total cost), or a physical relationship (the Pythagorean theorem for a ladder problem).

For example, consider designing a cylindrical storage tank with a fixed volume of $V_0$ cubic meters. Your goal is to minimize the cost, which is proportional to the surface area of the metal used. Here, the objective function is the surface area $S=2\pi r^2+2\pi rh$ (where $r$ is radius and $h$ is height). The constraint is the fixed volume: $V_0=\pi r^{2}h$. The constraint is what makes this a constrained optimization problem; without it, you could make the surface area arbitrarily small by shrinking the tank.

The Substitution Method for a Single Variable

Once you have identified your objective $f(x,y)$ and constraint $g(x,y)=c$, the most straightforward technique is solving the constraint for one variable and substituting into the objective function. This reduces the problem to a single-variable optimization, which you can solve using standard Calculus I techniques.

Using the tank example, solve the volume constraint for $h$: $h=\frac{V_0}{\pi r^2}$. Substitute this expression for $h$ into the surface area formula: $$S(r)=2\pi r^2+2\pi r\left(\frac{V_0}{\pi r^2}\right)=2\pi r^2+\frac{2V_0}{r}.$$ The problem is now simplified: find the radius $r>0$ that minimizes the single-variable function $S(r)$. You have successfully used the constraint to eliminate a variable, embedding the limitation directly into your objective.

Critical Points and the Derivative Tests

With your objective now as a function of a single variable, $S(r)$, you proceed with standard optimization. Find the critical points by taking the derivative, setting it to zero, and solving: $$S^{\prime }(r)=4\pi r-\frac{2V_0}{r^2}=0.$$ Solving $4\pi r=\frac{2V_0}{r^2}$ gives $4\pi r^3=2V_0$, so a critical point occurs at $r=\sqrt[3]{\frac{V_0}{2\pi }}$.

To classify this critical point as a minimum or maximum, apply the second derivative test. Compute the second derivative: $$S^{\prime \prime }(r)=4\pi +\frac{4V_0}{r^3}.$$ Since $r>0$ and $V_0>0$, $S^{\prime \prime }(r)>0$ for all valid $r$. A positive second derivative indicates the graph is concave up, confirming that this critical point is indeed a local (and in this context, global) minimum. Finally, substitute this optimal $r$ back into the constraint equation $h=V_0/(\pi r^2)$ to find the corresponding optimal height, completing the design.

The Closed Interval Method for Bounded Domains

In many applied problems, the variables have inherent physical or practical bounds. The radius of a tank cannot be negative, and the length of a fence segment cannot exceed your total lumber. When your variable exists on a closed, bounded interval $[a,b]$, the Closed Interval Method is your definitive tool for finding absolute (global) extrema.

The procedure is systematic:

1. Find all critical points of the objective function within the open interval $(a,b)$.
2. Evaluate the objective function at each critical point.
3. Evaluate the objective function at the endpoints $a$ and $b$.
4. The largest of these values is the absolute maximum; the smallest is the absolute minimum.

For instance, if you are fabricating a rectangular beam from a log of fixed diameter $D$, the beam's strength $S$ is proportional to its width $w$ and the square of its depth $d$ (i.e., $S=k\cdot w\cdot d^2$). The constraint from the circular cross-section is $w^2+d^2=D^2$. By substituting $w=\sqrt{D^2-d^2}$, you get $S(d)=k{d}^2\sqrt{D^2-d^2}$. Here, the domain for $d$ is the closed interval $[0,D]$ due to physical reality. You must check the critical points found by setting $S^{\prime }(d)=0$ and the endpoints $d=0$ and $d=D$ to find the dimensions that yield absolute maximum strength.

Applied Optimization: An Engineering Workflow

Let's synthesize these concepts into a reliable five-step engineering workflow for solving constrained optimization problems.

1. Define Variables and Draw: Assign symbols to all relevant quantities. A sketch is almost always invaluable.
2. Formulate Equations: Write the objective function (what to max/min) and the constraint equation (the limiting condition).
3. Reduce Variables: Use the constraint to express the objective function in terms of a single independent variable. State the domain of this variable (e.g., $r>0$, $0\le x\le L$).
4. Optimize: Find the critical points ($f^{\prime }(x)=0$ or $f^{\prime }(x)$ undefined) within the domain. Use the first or second derivative test to classify them, or employ the Closed Interval Method if the domain is closed and bounded.
5. Interpret and Verify: Answer the original question with units. Check if your result makes physical sense. Is it a plausible maximum strength or a reasonable minimum cost?

Common Pitfalls

Even with a strong procedural framework, it's easy to make subtle errors that derail a solution. Being aware of these common mistakes will sharpen your problem-solving.

Misidentifying the Objective and Constraint. This is a foundational error. If you mistakenly try to maximize your constraint (like fixed volume) instead of your objective (like surface area), the problem becomes nonsensical. Always ask: "What am I being asked to maximize or minimize?" That's the objective. "What is the fixed, limiting condition?" That's the constraint.

Neglecting the Domain of the Variable. After substitution, you must consider the realistic domain. For a geometric length, it must be positive ($x>0$). In a closed interval problem, forgetting to check the endpoints means you might miss the absolute extremum. Always explicitly state the domain before taking derivatives.

Failing to Fully Answer the Question. Finding the optimal critical point (e.g., $r=5$ cm) is only half the solution. The problem likely asks for all optimal dimensions or the optimal value of the objective. You must use your constraint to find the other dimension (e.g., $h$) and often compute the final optimized quantity (e.g., minimum surface area = $150\pi$ cm²). Always circle back to the original question's wording.

Algebraic Errors in Substitution and Differentiation. The calculus step is often straightforward; the algebra is where mistakes happen. Carefully perform the substitution and simplify the objective function before differentiating. Double-check your derivative, especially for negative exponents and quotient rules, as an incorrect derivative will lead to incorrect critical points.

Summary

• Constrained optimization problems are defined by an objective function to maximize/minimize and a constraint equation that links the variables.
• The primary solution method is substitution: solve the constraint for one variable, substitute into the objective to create a single-variable function, and then apply standard calculus optimization techniques.
• Always determine and use the practical domain of your variable. On a closed interval $[a,b]$, the Closed Interval Method guarantees you find the absolute extrema by evaluating the function at all critical points and the endpoints.
• Use the first or second derivative test to classify critical points found in the interior of an open domain as local maxima or minima.
• A disciplined, step-by-step workflow—from defining variables and formulating equations to interpreting the final answer—is essential for reliably solving these applied problems and avoiding common setup and algebraic errors.