Feasible Region — Definition & Example
A feasible region is the part of a graph where every constraint in a problem holds at the same time. It is the overlap of all the shaded inequality regions.
For a system of linear inequalities in and , the feasible region is the solution set of the system. Each inequality defines a half-plane, and the region is the intersection of those half-planes. The half-planes are closed for or and open for or . The region is always convex. It may be empty, which makes the system infeasible. It may also be unbounded. If it is nonempty and bounded, it is a convex polygon, possibly a degenerate one (a single point or a line segment). Each of its vertices lies where two boundary lines intersect.
Worked Example
Problem: Maximize subject to , , , and .
Step 1: Graph each boundary line and shade the side that satisfies its inequality. The overlap is a four-sided region. Find the corner where the two slanted lines cross by solving them as a system. Substitute into the second equation.
Step 2: Find the corners on the axes. Each one must satisfy every constraint. works because . works because . The axis intercepts and are not corners, because each one breaks the other constraint.
Step 3: Evaluate the objective function at each vertex. On a closed, bounded, nonempty feasible region, a linear objective always has a maximum and a minimum, and each occurs at a vertex.
Answer: The maximum value is , at the vertex .
Why It Matters
Feasible regions are the core of linear programming. You meet linear programming in Algebra 2, precalculus, and finite math courses through profit-maximization and cost-minimization word problems. The same idea runs operations research. Airlines use it to schedule crews, factories to plan production mixes, and dietitians to meet nutrient requirements at minimum cost.
Common Mistakes
Mistake: Treating every intersection of boundary lines as a corner of the feasible region. In the example, a student might test or .
Correction: Check each candidate point against all of the constraints. In the example, gives , so it lies outside the region. Only points that satisfy every inequality count as vertices.
