Why Finding the Root of a Function Is So Important: The Hidden Connection Between Equations, Equilibrium, Optimization, and Real-World Problems

Introduction

At first glance, finding the root of a function might seem like a relatively narrow mathematical exercise.

Given a function, the objective is simply to determine the value of x for which the function becomes zero.

For example:

Finding its roots means solving:

Therefore:

But why has so much of mathematics, numerical analysis, engineering, economics, physics, and computer science been built around finding roots?

The answer is surprisingly important:

Finding the root of a function is one of the most general ways of expressing the solution to a mathematical problem.

Many problems that initially appear unrelated can be transformed into the same mathematical question:

At what value of x does a particular function become zero?

Understanding this connection reveals why root-finding is such a fundamental concept.

1. Why Zero Is a Special Target

Suppose someone asks:

At what value of x does a function equal 10?

Mathematically, we can write:

Now move 10 to the left-hand side:

Define a new function:

The original question has now become:

The same transformation works for any target value.

Suppose:

We can always rewrite it as:

This means that finding where a function reaches a particular value is mathematically equivalent to finding the root of another function.

Zero is therefore not necessarily the original target. It is a convenient mathematical reference point that allows different problems to be expressed in a common form.

An important geometric interpretation

Imagine a curve representing:

Suppose we want to find where the curve reaches y = 5.

Geometrically, this means finding the intersection of the curve with the horizontal line:

Now subtract 5 from the function:

The intersection occurs precisely where:

In other words, the horizontal line y = 5 has effectively become the horizontal axis of a new function.

This is the fundamental geometric reason why different target values can be converted into root-finding problems.

2. Solving Almost Any Equation Becomes a Root-Finding Problem

Consider a general equation:

We can move everything to one side:

Define:

Now the problem becomes:

This transformation is extremely powerful because it means that we do not need a separate mathematical framework for every type of equation.

We can use a common collection of techniques to solve many different problems.

Example: Finding where two curves intersect

Suppose two functions are:

and:

To find their intersection points, we need to determine where their values are equal.

Therefore:

Move everything to one side:

Thus:

We have found two intersection points by finding the roots of a single function.

This same approach works even when the functions cannot be solved algebraically.

That is where numerical root-finding becomes particularly useful.

3. Equilibrium and Balance: When Different Quantities Become Equal

One of the most important applications of root-finding is the study of equilibrium.

An equilibrium occurs when different influences balance each other.

Rather than asking directly where two quantities are equal, we can calculate their difference and find where that difference becomes zero.

Example A: Supply and demand

In economics, the quantity supplied and the quantity demanded depend on price.

Suppose demand is:

And supply is:

An equilibrium occurs when supply equals demand:

Substituting:

Rearranging:

Therefore:

At a price of 20, supply and demand are equal.

Notice what happened: an economic equilibrium became a root-finding problem.

Example B: Mechanical equilibrium

Imagine an object acted upon by several forces.

If the forces are represented algebraically, the net force is:

An equilibrium position can occur when:

At that position, the forces balance.

There is an important distinction, however: zero net force establishes mechanical equilibrium, but it does not automatically establish stability. An object might return to its equilibrium position after a small disturbance, or move farther away from it.

Example C: Chemical equilibrium

Chemical equilibrium occurs when the forward and reverse reaction rates become equal.

If the forward reaction rate is Rf and the reverse reaction rate is Rr, equilibrium requires:

Again, the difference between two quantities becomes a function whose root identifies the equilibrium condition.

The same mathematical structure appears in economics, mechanics, and chemistry despite the underlying physical processes being very different.

4. Optimization: Finding Maximum and Minimum Values

Root-finding becomes even more interesting when we move from solving equations to optimization.

Suppose we have a function representing the profit of a business:

We want to determine the value of x that produces maximum profit.

How can we find it?

Differentiation provides the answer.

The derivative measures the instantaneous rate of change of a function.

At an interior maximum or minimum of a differentiable function, the derivative is zero, provided the derivative exists there.

Therefore:

The optimization problem has now become a root-finding problem.

Instead of finding the roots of the original profit function, we find the roots of its derivative.

Example: Maximizing a profit function

Suppose profit is represented by:

Differentiate:

For a stationary point:

Therefore:

To determine whether this is a maximum or minimum, examine the second derivative:

Since the second derivative is negative, the function is concave downward.

Consequently, x = 50 gives the maximum profit, assuming this value lies within the permitted range of x.

We have converted an optimization problem into a root-finding problem involving the derivative.

Why this matters in machine learning

Machine learning models often contain adjustable parameters.

The objective is to find parameter values that minimize a loss function, which measures how far the model’s predictions are from the desired results.

A simplified optimization problem can be written as:

Here, w represents the model parameters.

For a differentiable loss function, a necessary condition for an interior minimum is:

With many parameters, the corresponding condition is that the gradient of the loss function becomes zero.

The gradient is a collection of partial derivatives.

Many optimization algorithms, including gradient-based methods, use this information to move towards a minimum.

However, there is an important qualification: finding a zero gradient does not automatically guarantee a minimum. It could also represent a maximum or a saddle point. Practical optimization algorithms must deal with these possibilities.

Nevertheless, this demonstrates a profound connection:

A large part of optimization can be understood as finding where the derivative or gradient becomes zero.

5. Thresholds, Decisions, and Break-Even Points

Another important application is determining when a quantity crosses a particular threshold.

These questions appear throughout engineering, medicine, finance, and everyday decision-making.

Example A: Business break-even analysis

Suppose a business has:

• Fixed costs of 1,000.
• Variable costs of 20 per product.
• Selling price of 30 per product.

Revenue is:

Total cost is:

The business breaks even when revenue equals total cost:

Rearranging:

Simplifying:

Therefore:

The business must sell 100 products to break even.

Geometrically, this is the intersection of the revenue and cost curves.

Example B: A safety threshold

Imagine an engineering system where a measured quantity changes with time.

Suppose the safe operating limit is 80 units.

The system’s output is represented by:

We want to know when the output reaches the safety threshold.

Therefore:

Rearranging:

The output reaches the threshold after 8 time units.

Notice that the original problem was not about finding zero. It was about finding when the output reached 80.

But subtracting the threshold converted the question into a root-finding problem.

This technique is useful for identifying when a system crosses a safety limit, when a medication concentration reaches a specified level, or when an investment reaches its break-even value.

6. Why Changing Signs Helps Us Find Roots

There is another reason why zero is particularly useful.

A root is a point where a function becomes zero.

For many continuous functions, we can identify an interval containing a root simply by checking the signs of the function at its endpoints.

Suppose:

And:

If the function is continuous throughout the interval from a to b, the Intermediate Value Theorem guarantees that there is at least one value c between a and b for which:

This is an extremely powerful mathematical result.

Understanding it geometrically

Imagine a continuous curve.

At the first point, the curve is above the horizontal axis.

At the second point, it is below the horizontal axis.

If the curve is continuous, it cannot move from above the axis to below it without crossing the axis somewhere in between.

That crossing represents a root.

The important point is that we do not need to know the exact root to establish that one exists.

We only need to know the signs at two endpoints.

An important qualification

A change of sign guarantees a root for a continuous function, but the absence of a sign change does not mean that a root does not exist.

For example:

At both x = -1 and x = 1:

Both values are positive, yet:

The function touches the horizontal axis without crossing it.

Therefore, sign-change methods can detect many roots, but they can miss roots where the curve merely touches the axis.

7. How the Bisection Method Uses This Property

The Intermediate Value Theorem provides the mathematical foundation for the bisection method.

Suppose we want to solve:

We know:

And:

Because the function is continuous and its values have opposite signs, at least one root lies between 1 and 2.

The bisection method repeatedly divides the interval into two halves.

First iteration

The midpoint is:

Evaluate the function:

Since f(1.5) is positive and f(1) is negative, the root must lie between 1 and 1.5.

Second iteration

The new midpoint is:

Evaluate:

The root must now lie between 1.25 and 1.5.

Continuing the process

Each iteration reduces the interval containing the root by half.

Eventually, we obtain an increasingly accurate approximation.

The actual positive root is:

The remarkable thing about bisection is that it does not need to know the exact mathematical solution.

It only needs continuity and an interval with opposite endpoint signs.

This makes it one of the most reliable numerical root-finding methods.

8. Newton’s Method: Using Derivatives to Find Roots

The bisection method uses the sign of the function to narrow down an interval.

Newton’s method takes a different approach.

It uses the derivative to estimate where the function crosses the horizontal axis.

Imagine starting at a point on a curve.

Draw a tangent line at that point.

Instead of finding where the original curve crosses the horizontal axis, find where the tangent line crosses it.

That intersection provides a new approximation to the root.

The formula is:

Here:

• xn is the current approximation.
• f(xn) is the function value.
• f'(xn) is the derivative.
• xn+1 is the improved approximation.

Example

Consider:

Its derivative is:

Suppose we start with:

Using Newton’s method:

Therefore:

Repeating the process:

We are already very close to the actual root.

Newton’s method can converge extremely quickly when the starting approximation is suitable and the mathematical conditions are favorable.

However, unlike bisection, it does not always converge. A poor starting point, a zero or very small derivative, or complicated function behaviour can cause difficulties.

This is why different root-finding methods are useful for different situations.

9. Why Numerical Root-Finding Became So Important: The History of Algebra

The importance of finding roots also has a fascinating historical background.

For centuries, mathematicians investigated polynomial equations.

Consider the general linear equation:

Its solution is straightforward:

Quadratic equations introduced another level of complexity:

The quadratic formula provides their solutions:

Mathematicians subsequently developed general solution formulas for cubic and quartic equations.

But the situation changed dramatically with fifth-degree equations.

The mystery of the quintic

Consider a general polynomial:

For a general polynomial of degree five, there is no formula using only a finite combination of arithmetic operations and radicals that expresses all its roots in terms of its coefficients.

This result is associated with the work of mathematicians including Niels Henrik Abel and Évariste Galois.

Galois theory eventually provided a deeper explanation of why general solutions by radicals are impossible for many higher-degree polynomial equations.

But this did not mean that such equations could not be solved.

It meant that their solutions could not always be expressed through familiar algebraic formulas.

Numerical methods offered another way forward.

Instead of trying to express every root in an exact symbolic form, mathematicians could calculate increasingly accurate numerical approximations.

This distinction remains important today:

• An exact solution expresses a root symbolically.
• A numerical solution estimates the root to a chosen degree of accuracy.

For many practical problems, a numerical approximation is entirely sufficient.

10. Why Root-Finding Is a Universal Mathematical Technique

We can now see that root-finding is not limited to polynomial equations.

It provides a common mathematical framework for many different types of problems.

Problem	Original question	Root-finding formulation
Equation solving	When does g(x) equal h(x)?	g(x) – h(x) = 0
Supply and demand	When are supply and demand equal?	S(x) – D(x) = 0
Mechanical equilibrium	When do forces balance?	Net force = 0
Chemical equilibrium	When are reaction rates equal?	Forward rate – reverse rate = 0
Optimization	Where is the stationary point?	f'(x) = 0
Break-even analysis	When does revenue equal cost?	R(x) – C(x) = 0
Safety thresholds	When does output reach a critical value?	Q(x) – critical value = 0
Curve intersections	Where do two curves meet?	g(x) – h(x) = 0

Different disciplines may use different terminology, but the mathematical structure is often remarkably similar.

The central idea is to construct a function representing the difference between two quantities or the deviation from a desired condition.

Then we find where that difference becomes zero.

11. The Deeper Mathematical Insight: Zero Represents the Absence of Difference

There is a philosophical and mathematical reason why zero appears so frequently.

Suppose we have two quantities, A and B.

We want to know when they are equal.

Their difference is:

When the difference is positive, A is greater than B.

When the difference is negative, A is smaller than B.

When the difference is zero, the two quantities are equal.

Thus:

Zero represents the point at which the difference disappears.

This interpretation is more general than simply saying that zero is where a graph crosses the horizontal axis.

It explains why zero naturally represents:

• Balance between opposing forces.
• Equality between supply and demand.
• The difference between revenue and cost disappearing.
• The derivative becoming zero at a stationary point.
• The deviation from a target value disappearing.
• The residual error in an equation becoming zero.

In each case, zero represents the satisfaction of a mathematical condition.

Conclusion: Finding Roots Is About Finding Where a Condition Is Satisfied

The importance of root-finding becomes much clearer when we stop thinking of it merely as solving equations such as:

Instead, we should recognize it as a general mathematical strategy.

Whenever a problem asks:

• When are two quantities equal?
• When does a system reach equilibrium?
• When does a function reach a specified value?
• When does a business break even?
• When does a derivative become zero?
• When does a measured quantity cross a threshold?

We can often transform the question into finding the root of an appropriately constructed function.

This is why methods such as bisection, Newton’s method, and the secant method are so important.

They provide different ways of locating the point where a mathematical condition becomes satisfied.

The deeper lesson is that finding zero is not necessarily the original problem. It is a common mathematical language into which many different problems can be translated.

Once this connection is understood, root-finding stops looking like an isolated numerical technique and becomes one of the central ideas connecting algebra, calculus, geometry, science, engineering, economics, and modern computing.