Skip to content

Mathematics · Ch 2 — Matrices

Method of Reduction

2.3.2

Method of Reduction

As the name suggests, in this method the given equations are reduced -- through row transformations -- to a form from which the solution is read off directly, without ever computing a matrix inverse.

The procedure. Start, as in section 2.3, by converting the given linear equations into the matrix equation AX=BAX=B. Then perform suitable row transformations on AA, reducing it to an upper triangular, lower triangular, or fully diagonal matrix. The same row transformations, in the same order, are performed simultaneously on the constant matrix BB. Once this is done, the transformed matrix equation is rewritten back as a system of linear equations -- but now in a form simple enough to solve directly by the elimination method (each equation, taken in the right order, involves one fewer unknown than the one before it).

Worked Example -- two equations, upper triangular. Solve 2x+3y=92x+3y=9 and y−x=−2y-x=-2 by the method of reduction. Writing the equations as 2x+3y=92x+3y=9 and −x+y=−2-x+y=-2, the matrix equation is [23−11][xy]=[9−2]\begin{bmatrix}2&3\\-1&1\end{bmatrix}\begin{bmatrix}x\\y\end{bmatrix}=\begin{bmatrix}9\\-2\end{bmatrix}. Using R2→2R2+R1R_2\to 2R_2+R_1: [2305][xy]=[95]\begin{bmatrix}2&3\\0&5\end{bmatrix}\begin{bmatrix}x\\y\end{bmatrix}=\begin{bmatrix}9\\5\end{bmatrix}. Rewriting as equations: 2x+3y=92x+3y=9 …(i) and 5y=55y=5 …(ii). From (ii), y=1y=1; substituting into (i) gives x=3x=3. So x=3, y=1x=3,\ y=1.

Worked Example -- three equations, upper triangular. Solve x+3y+3z=12x+3y+3z=12, x+4y+4z=15x+4y+4z=15, x+3y+4z=13x+3y+4z=13 by the method of reduction. The matrix equation is [133144134][xyz]=[121513]\begin{bmatrix}1&3&3\\1&4&4\\1&3&4\end{bmatrix}\begin{bmatrix}x\\y\\z\end{bmatrix}=\begin{bmatrix}12\\15\\13\end{bmatrix}. Using R2→R2−R1R_2\to R_2-R_1 and R3→R3−R1R_3\to R_3-R_1: [133011001][xyz]=[1231]\begin{bmatrix}1&3&3\\0&1&1\\0&0&1\end{bmatrix}\begin{bmatrix}x\\y\\z\end{bmatrix}=\begin{bmatrix}12\\3\\1\end{bmatrix}. This is already upper triangular, so back-substitution gives z=1z=1; then y+z=3⇒y=2y+z=3\Rightarrow y=2; then x+3y+3z=12⇒x=12−6−3=3x+3y+3z=12\Rightarrow x=12-6-3=3. So x=3, y=2, z=1x=3,\ y=2,\ z=1.

Worked Example -- three equations, lower triangular. Solve x+y+z=1x+y+z=1, 2x+3y+2z=22x+3y+2z=2, x+y+2z=4x+y+2z=4. The matrix equation is [111232112][xyz]=[124]\begin{bmatrix}1&1&1\\2&3&2\\1&1&2\end{bmatrix}\begin{bmatrix}x\\y\\z\end{bmatrix}=\begin{bmatrix}1\\2\\4\end{bmatrix}. Using R2→R2−R3R_2\to R_2-R_3 and R1→R1−12R3R_1\to R_1-\tfrac12R_3, then R1→R1−14R2R_1\to R_1-\tfrac14R_2, the coefficient matrix reduces to a lower triangular matrix (this time zeros appear above the diagonal instead of below), and the system rewrites -- solved from the top down -- as x=−2x=-2, then x+2y=−2⇒y=0x+2y=-2\Rightarrow y=0, then x+y+2z=4⇒2z=6⇒z=3x+y+2z=4\Rightarrow 2z=6\Rightarrow z=3. So x=−2, y=0, z=3x=-2,\ y=0,\ z=3. …

Misc 2.3.2aSolved Examples 1-4 -- reduction method for two-variable and three-variable systems, including a word problem

Worked out. Example 1 reduces a 2x2 coefficient matrix to upper triangular form with one row operation and solves by back-substitution; Examples 2 and 3 each reduce a 3x3 coefficient matrix (one to upper triangular, one to lower triangular) using two rounds of row operations and then solve for all three unknowns in sequence; Example 4 is a word problem about the cost of books and notebooks that is translated into two equations, reduced to upper triangular form, and solved for both unit costs. …