Concept understanding — Solution of a System of Linear Equations by Method of Inversion
A system of n linear equations in n unknowns, such as a1x+b1y+c1z=d1, a2x+b2y+c2z=d2, a3x+b3y+c3z=d3, can be written as a single matrix equation AX=B, where A is the n×n matrix of coefficients, X is the n×1 column of unknowns, and B is the n×1 column of constants. Provided a unique solution exists, A must be non-singular, so A−1 exists. Pre-multiplying both sides of AX=B by A−1 gives A−1(AX)=A−1B, i.e. (A−1A)X=A−1B, i.e. IX=A−1B, so X=A−1B — this single matrix product delivers every unknown at once, read off as the corresponding entries of the column X. A−1 itself can be found by either the elementary-transformation method or the adjoint method; the adjoint formula A−1=∣A∣1(adjA) is usually the more direct route once the system is set up. Because the method genuinely requires A−1, it silently fails whenever ∣A∣=0 — a coefficient matrix that turns out singular means the method of inversion cannot produce a unique answer, and the equations must instead be checked directly for consistency (do they describe parallel/coincident lines or planes, giving no solution or infinitely many).
The system is only solvable by inversion when a=0; solve symbolically in terms of a.
Step 2: ∣A∣, expanding along row 1: 1(3⋅2a−2⋅a)−1(2⋅2a−2a)+1(2a−3a)=1(4a)−1(2a)+1(−a)=4a−2a−a=a.
Step 3: So ∣A∣=a; the method of inversion needs ∣A∣=0, i.e. a=0 -- a necessary condition stated up front, since dividing by ∣A∣=a is only valid then.
Step 4: Solving (by row-reduction, equivalent to the adjoint route but cleaner symbolically): from row 3, ax+ay+2az=4⇒a(x+y)+2az=4; using row 1, x+y=1−z, so a(1−z)+2az=4⇒a+az=4⇒z=a4−a (using a=0).
Step 5: From row 2 minus 2×row 1: (2x+3y+2z)−2(x+y+z)=2−2(1)⇒y=0.
Step 6: From row 1: x=1−y−z=1−0−a4−a=1−a4−a=aa−4+a=a2a−4=2−a4.
✓Final answer
x=2−a4,y=0,z=a4−a, valid whenever a=0 (the coefficient matrix is singular and the method of inversion fails when a=0).
Write as AX=B with A containing the parameter a; find A−1 symbolically (valid only when a=0), then compute X=A−1B.
Not stating the necessary condition a=0 before dividing by ∣A∣=a
Treating a as a fixed number instead of carrying it symbolically through the solution
Sign/arithmetic slip in the row-3 expansion involving 2a
Q.If three numbers are added, their sum is 2. If two times the second number is subtracted from the sum of first and third numbers we get 8 and if three times the first number is added to the sum of second and third numbers we get 4. Find the numbers using matrices.
›Reveal solutionSolution
Translate the word problem into 3 linear equations, write as AX=B, and solve by elimination (equivalent to matrix reduction).
Let the numbers be x,y,z.
"If three numbers are added, their sum is 2": x+y+z=2 ... (i)
"If two times the second number is subtracted from the sum of first and third we get 8": (x+z)−2y=8⟹x−2y+z=8 ... (ii)
"If three times the first number is added to the sum of second and third we get 4": 3x+(y+z)=4⟹3x+y+z=4 ... (iii)
In matrix form AX=B: A=1131−21111, X=xyz, B=284
Q.The cost of 4 dozen pencils, 3 dozen pens and 2 dozen erasers is ₹60. The cost of 2 dozen pencils, 4 dozen pens and 6 dozen erasers is ₹90 whereas the cost of 6 dozen pencils, 2 dozen pens and 3 dozen erasers is ₹70. Find the cost of each item per dozen by using matrices.
›Reveal solutionSolution
Set up AX=B from the three cost equations and solve using Cramer's rule (determinants).
Let x,y,z = cost per dozen of pencils, pens, erasers respectively.
4x+3y+2z=60
2x+4y+6z=90
6x+2y+3z=70
In matrix form AX=B with A=426342263, X=xyz, B=609070.