3.2. Illustration with an example
For the sake of clearly describing the concept of proposed method, we consider the following ILP problem
with three constraints and two variables:
maximize z= 7x1 +10x2
subject to
x1 + 3x2 < 6;
7x1 + x2 < 35;
x1 + 2x2 < 2;
x1; x2 >0 and are integers.
Step 1: Solve the relaxed problem to get the optimum point A(4.5, 3.5) as shown in Fig. 1.
Step 2: Select (5, 0) and (0, 2) to calculate z
ð5;0Þ ¼ 35 and z
ð0;2Þ ¼ 20. In this case, we find that (5, 0) is the nearest
extreme point to A and designate (5, 0) as B.
Step 3: Draw a line passing through B with the same slope of objective function. After that we obtain a new
intersection C and the first subspace S1. Then we add new constraint BC and remove irrelative constraints
x1 + 2x2 >2 as shown in Fig. 2. The original problem could be revised as below:
maximize z = 7x1 + 10x2
subject to
x1 +3x2 < 6;
7x1 +x2 35;
x1; x2 > 0 and are integers.
Step 4: Use B&B procedure to obtain optimum solution in subspace S1, we could get the maximum value z = 58 when x1 = 4 and x2 = 3. It is also the optimum integer solution in the original space and we could claim that this is the solution of the original ILP problem.