Typically m < n, so the system has many solutions; each such solution is called a feasible solution of the LP. Let B be a subset of m indices from. Denote by the square m-by-m matrix made of the m columns of indexed by B. If is nonsingular, the columns indexed by B are a basis of the column space of. In this case, we call B a basis of the LP. Since the rank of is m, it has at least one basis; since has n columns, it has at most bases. Given a basis B, we say that a feasible solution is a basic feasible solution with basis B if all its non-zero variables are indexed by B, i.e., for all.
Properties
1. A BFS is determined only by the constraints of the LP ; it does not depend on the optimization objective. 2. By definition, a BFS has at most m non-zero variables and at least n-m zero variables. A BFS can have less than m non-zero variables; in that case, it can have many different bases, all of which contain the indices of its non-zero variables. 3. A feasible solution is basic if-and-only-if the columns of the matrix are linearly independent, where K is the set of indices of the non-zero elements of. 4. A BFS is uniquely determined by the basis B: for each basis B of m indices, there is at most one BFS with basis B. This is because must satisfy the constraint, and by definition of basis the matrix is non-singular, so the constraint has a unique solution. The opposite is not true: each BFS can come from many different bases. If the unique solution of satisfies the non-negativity constraints, then the basis is called a feasible basis. 5. If a linear program has an optimal solution, then it has an optimal BFS. This is a consequence of the Bauer maximum principle: the objective of a linear program is convex; the set of feasible solutions is convex ; therefore the objective attains its maximum in an extreme point of the set of feasible solutions. Since the number of BFS-s is finite and bounded by, an optimal solution to any LP can be found in finite time by just evaluating the objective function in all BFS-s. This is not the most efficient way to solve an LP; the simplex algorithm examines the BFS-s in a much more efficient way.
Examples
Consider a linear program with the following constraints: The matrix A is: Here, m=2 and there are 10 subsets of 2 indices, however, not all of them are bases: the set is not a basis since columns 3 and 5 are linearly dependent. The set B= is a basis, since the matrix is non-singular. The unique BFS corresponding to this basis is.
Geometric interpretation
The set of all feasible solutions is an intersection of hyperspaces. Therefore, it is a convex polyhedron. If it is bounded, then it is a convex polytope. Each BFS corresponds to a vertex of this polytope.
The simplex algorithm keeps, at each point of its execution, a "current basis" B, a "current BFS", and a "current tableau". The tableau is a representation of the linear program where the basic variables are expressed in terms of the non-basic ones: where is the vector of m basic variables, is the vector of n non-basic variables, and is the maximization objective. Since non-basic variables equal 0, the current BFS is, and the current maximization objective is. If all coefficients in are negative, then is an optimal solution, since all variables must be at least 0, so the second line implies. If some coefficients in are positive, then it may be possible to increase the maximization target. For example, if is non-basic and its coefficient in is positive, then increasing it above 0 may make larger. If it is possible to do so without violating other constraints, then the increased variable becomes basic, while another non-basic variable is decreased to 0 to keep the equality constraints and thus becomes non-basic. If this process is done carefully, then it is possible to guarantee that increases until it reaches the optimal BFS.
