Course contentsShow
← Mathematics
Lesson 2114 of 3,12144. Linear Algebra — Matrices and SystemsFree lesson

Gaussian Elimination with Multiple Right-Hand Sides

Solving several systems with the same coefficient matrix simultaneously using an extended augmented matrix.

Gaussian Elimination with Multiple Right-Hand Sides

What You'll Learn

You'll learn how to efficiently solve several linear systems that share the same coefficient matrix by processing all right-hand sides simultaneously in one extended augmented matrix.

The Core Idea

Imagine you're solving three different problems at a company, all using the same equations but with different target values. Instead of repeating Gaussian elimination three separate times, you can solve them all at once.

When multiple systems share the same coefficient matrix A but have different right-hand sides b₁, b₂, etc., you can create an extended augmented matrix by placing all the right-hand side vectors as additional columns:

[A | b₁ | b₂ | b₃]

Then perform Gaussian elimination just once. Every row operation you apply affects all the right-hand sides simultaneously. After reaching row echelon form, you can use back substitution independently on each right-hand side column to find all solutions in one pass.

Why This Matters

This technique is far more efficient than solving each system separately. Since the most computationally expensive part of Gaussian elimination is reducing the coefficient matrix, doing it once and carrying along multiple right-hand sides saves enormous time—especially important for large systems in engineering and scientific computing.

Example Structure

If solving Ax = b₁ and Ax = b₂, instead of working with [A | b₁] and then [A | b₂], you work with:

[A | b₁ | b₂]

Apply the same elementary row operations to the entire extended matrix, and you get both solutions from one elimination process.

Key Takeaway: When solving multiple systems with the same coefficient matrix, combine all right-hand sides into an extended augmented matrix and perform Gaussian elimination once to solve all systems simultaneously.