Basic solutions
Basic Solutions
A basic solution is obtained by selecting a basis, setting nonbasic variables to zero, and solving the remaining square system.
Setup
Consider
\[Ax=b,\]where $A\in\mathbb{R}^{m\times n}$ and $\operatorname{rank}(A)=m$.
Choose a Basis
Choose $m$ linearly independent columns of $A$:
\[B=[A_{B(1)}\cdots A_{B(m)}].\]The corresponding variables are basic variables $x_B$. The others are nonbasic variables $x_N$.
Construct the Basic Solution
Set
\[x_N=0\]and solve
\[Bx_B=b.\]Thus
\[x_B=B^{-1}b.\]Feasibility Is Separate
A basic solution satisfies $Ax=b$, but it may not satisfy $x\ge 0$.
If any component of $x_B$ is negative, it is not feasible.
At Most m Nonzero Variables
A basic solution has all nonbasic variables equal to zero, so it has at most $m$ nonzero variables.
Checklist
You should be able to choose a valid basis, compute $B^{-1}b$, build the full vector $x$, and test its signs.
See Also
Exam checkpoint
For BFS questions, convert to standard form before checking the point. A candidate in original variables must be lifted with slack and surplus variables before testing basis columns.