Math 407 Linear Optimization
Graphical solutions of two dimensional LPs
Solution Procedure
Step 1: Graph each of the linear constraints indication on which side of the constraint the
feasible region must lie. Don’t forget the implicit constraints!
Step 2: Shade in the feasible region.
Step 3: Draw the gradient vector of the objective function.
Step 4: Place a straightedge perpendicular to the gradient vector and move the straightedge
either in the direction of the gradient vector for maximization, or in the opposite direc-
tion of the gradient vector for minimization to the last point for which the straightedge
intersects the feasible region. The set of points of intersection between the straightedge
and the feasible region is the set of solutions to the LP.
Step 5: Compute the exact optimal vertex solutions to the LP as the points of intersection
of the lines on the boundary of the feasible region indicated in Step 4. Then compute
the resulting optimal value associated with these points.
1. Sketch the graph of the constraint region for the following LP’s. Then solve them by
sketching the optimal level set of the objective function.
(a)
maximize 2x + 3y
subject to −3x + y ≤ 2
4x + 2y ≤ 44
4x − y ≤ 20
−x + 2y ≤ 14
0 ≤ x, y
(Hint: optimal value = 42)
(b)
minimize x + y
subject to −x + y ≤ 3
2x + y ≤ 18
y ≥ 6
0 ≤ x, y
(Hint: optimal value = 9)
(c)
maximize 3x + 2y
subject to x + 4y ≤ 16
x − 2y ≤ −1
1 ≤ y ≤ 4
0 ≤ x ≤ 4
(Hint: optimal value = 18)
1
2. Graph the following function of α by graphically solving the necessary LPs:
v(α) := maximize x1 + αx2
subject to x1 − x2 ≤ 4
x1 + x2 ≤ 6
−3x1 + x2 ≤ −6
0 ≤ x1 , x 2 .
(Hint: v(α) = 4 for all α ≤ −1)
