← Back to Linear Algebra

Pivots, free variables, and solution sets

When a system has infinitely many solutions, pivots and free variables describe all of them neatly. The structure of solution sets reveals deep facts about matrices.

Basic and free variables

After row reduction, variables corresponding to pivot columns are basic variables; the rest are free variables. Free variables can take any value, and each choice determines the basic variables. A consistent system with at least one free variable therefore has infinitely many solutions; with no free variables, its solution is unique.

Parametric form

Solutions are written by solving for basic variables in terms of free ones. If RREF gives \(x_1 + 2x_3 = 4\) and \(x_2 - x_3 = 1\) with \(x_3\) free, then \(x_1 = 4 - 2t\), \(x_2 = 1 + t\), \(x_3 = t\) for any real t. In vector form, \(\mathbf{x} = (4, 1, 0) + t(-2, 1, 1)\): a line in \(\mathbb{R}^3\).

Homogeneous systems

A homogeneous system \(A\mathbf{x} = \mathbf{0}\) always has the trivial solution \(\mathbf{x} = \mathbf{0}\). It has non-trivial solutions exactly when there is a free variable. A homogeneous system with more unknowns than equations always has a free variable, and so always has non-trivial solutions.

Structure of solution sets

If \(\mathbf{p}\) is any particular solution of \(A\mathbf{x} = \mathbf{b}\), then every solution has the form \(\mathbf{p} + \mathbf{h}\), where \(\mathbf{h}\) solves the homogeneous system \(A\mathbf{x} = \mathbf{0}\). Geometrically, the solution set of a consistent non-homogeneous system is the homogeneous solution set, a subspace, shifted away from the origin by \(\mathbf{p}\).

Rank

The rank of a matrix is its number of pivots. For an m x n coefficient matrix A, the number of free variables in a consistent system is n minus the rank. If the rank equals n, solutions are unique when they exist; if the rank equals m, the system is consistent for every right-hand side, since no zero row can appear in the coefficient part.

Existence and uniqueness

Existence asks whether a system is consistent: is b a linear combination of A's columns? Uniqueness asks whether there are free variables. These two questions, about pivots in every row and pivots in every column, recur throughout linear algebra, and for square matrices they turn out to be equivalent.

Applications

Balancing a chemical equation leads to a homogeneous system whose free variable sets the overall scale. Network flow problems, such as traffic through junctions, give systems where free variables represent choices of routing. Underdetermined systems in data science, with more unknowns than equations, have infinitely many solutions, so extra criteria, such as smallest length, select one.

Worked example

Balance \(a\,\mathrm{C_3H_8} + b\,\mathrm{O_2} \to c\,\mathrm{CO_2} + d\,\mathrm{H_2O}\). Carbon: \(3a = c\); hydrogen: \(8a = 2d\); oxygen: \(2b = 2c + d\). With a as the free variable, \(c = 3a\), \(d = 4a\), and \(b = \frac{2(3a) + 4a}{2} = 5a\). Choosing a = 1 gives the balanced equation \(\mathrm{C_3H_8} + 5\mathrm{O_2} \to 3\mathrm{CO_2} + 4\mathrm{H_2O}\).

Key takeaways

Pivot columns give basic variables; others are free and can take any value. Consistent systems with free variables have infinitely many solutions, written in parametric or vector form. Homogeneous systems always have the trivial solution and have non-trivial ones exactly when a free variable exists, guaranteed when unknowns outnumber equations. Every solution is a particular solution plus a homogeneous solution. Rank counts pivots; free variables number n minus rank; rank n gives uniqueness and rank m guarantees existence. Applications include balancing reactions and network flows.

Flashcard

Basic variable

Corresponds to a pivot column

Determined by free ones.

Check your understanding

A consistent system has 5 unknowns and a coefficient matrix of rank 3. How many free variables does it have?

← Previous lesson Next lesson →