Chapter 8

Linear Programming with Matlab

The Matlab function LINPROG can be used to solve a linear programming problem with the
following syntax (help LINPROG):

X=LINPROG(f, A, b) solves the linear programming problem:

min f'*x subject to: A*x <= b

X=LINPROG(f, A, b, Aeq, beq) solves the problem above while additionally satisfying the
equality constraints Aeq*x = beq.

X=LINPROG(f, A, b, Aeq, beq, LB, UB) defines a set of lower and upper bounds on the
design variables, X, so that the solution is in the range LB <= X <= UB. Use empty matrices
for LB and UB if no bounds exist. Set LB(i) = -Inf if X(i) is unbounded below; set UB(i) =
Inf if X(i) is unbounded above.

Since LINPROG minimizes an objective function, the example 3.2-1 is reformulated as

Minimize z = − 150x1 − 175x2

Note: Maximizing (150x1 + 175x2) is equivalent to minimizing (− 150x1 − 175x2)

Subject to

7x1 + 11x2 ≤ 77 (material constraint)

The coefficient vector f for the objective function is

 − 150
f=  
 − 175

The matrix coefficient A for the inequality constraints is

 7 11
A=  
10 8 

8-9
The right hand vector b for the inequality constraints is

77
b=  
80

Since there are no equality constraints in this example, Aeq and beq are zeros.

0 0  0
Aeq =   and beq =  
0 0  0

The lower and upper bounds vectors are given by

 0 9 
LB =   and UB = 6
 0  

The following Matlab statements are used to solve this linear programming problem.

Matlab Example -----------------------------------------------------

% Example 8.3-1
f=[-150;-175];
A=[7 11;10 8];
b=[77;80];
Aeq=[0 0;0 0];
beq=[0;0];
LB=[0;0];UB=[9;6];
x=linprog(f,A,b,Aeq,beq,LB,UB)

>> e3d2d1
Optimization terminated successfully.
x=
4.8889
3.8889
-------------------------------------------------------------------------------------------------------------
Example 8.3-2 ----------------------------------------------------------------------------------
Solve the following linear programming problem using Matlab LINPROG

Maximize y = x1 + 2x2

Subject to

2x1 + x2 ≤ 10
x1 + x2 ≤ 6
− x1 + x2 ≤ 2
− 2x1 + x2 ≤ 1
x1 , x2 ≥ 0

8-10
Solution ------------------------------------------------------------------------------------------

The problem is reformulated using slack variable S1, S2, S3, and S4

Minimize y = − x1 − 2x2

Subject to

2x1 + x2 + S1 = 10
x1 + x2 + S2 = 6
− x1 + x2 + S3 = 2
− 2x1 + x2 + S4 = 1
x1, x2, S1, S2, S3, and S4 ≥ 0

The arguments for the Matlab command X=LINPROG(f, A, b, Aeq, beq, LB, UB) are given
as
 − 1  2 1 1 0 0 0 10  0 ∞
 − 2  1 1 0 1 0 0 6  0 ∞
         
 0  − 1 1 0 0 1 0 2  0 ∞
f =   , A = 0, b = 0, Aeq =   , beq =   , LB =   , UB =  
 0  − 2 1 0 0 0 1 2  0 ∞
 0  0 0 0 0 0 0 0  0 ∞
         
 0  0 0 0 0 0 0 0  0 ∞

The following Matlab statements are used to solve this linear programming problem.

Matlab Example -----------------------------------------------------

% Example 8.3-2
f=[-1;-2;0;0;0;0];
A=zeros(6,6);
b=zeros(6,1);
Aeq=[2 1 1 0 0 0;1 1 0 1 0 0; -1 1 0 0 1 0;-2 1 0 0 0 1;0 0 0 0 0 0;0 0 0 0 0 0];
beq=[10;6;2;1;0;0];
LB=[0;0;0;0;0;0];UB=[inf;inf;inf;inf;inf;inf];
x=linprog(f,A,b,Aeq,beq,LB,UB)

>> e8d3d2
Optimization terminated successfully.
x=
2.0000
4.0000
2.0000
0.0000
0.0000
1.0000
------------------------------------------------------------------------------

8-11
The solution from the Matlab program is

x1 = 2

x2 = 4

S1 = 2

S2 = 0

S3 = 0

S4 = 1

This solution is verified with the graphical solution shown in Figure 8.3-2. The vertex D is
the basic feasible solution obtained graphically.

x2 2x1 + x2 = 10
-2x1 + x2 = 1
8
-x1 + x2 = 2

4
D
C
E
Feasible
Region x 1 + x2 = 6
B
A
4 F 8 x
1
Figure 8.3-2 Geometric representation of the linear programming problem.

The problem can also be reformulated using only one slack variable S1

Minimize y = − x1 − 2x2

Subject to

2x1 + x2 + S1 = 10
x1 + x2 ≤ 6
− x1 + x2 ≤ 2
− 2x1 + x2 ≤ 1
x1, x2, and S1≥ 0

8-12
Consider the Matlab command X=LINPROG(f, A, b, Aeq, beq, LB, UB). The coefficient
vector f for the objective function is

 − 1
f =  − 2
 
 0 

The matrix coefficient A for the inequality constraints is

 1 1 0
A =  − 1 1 0
 
 − 2 1 0

The right hand vector b for the inequality constraints is

 6
b =  2
 
1

The equality constraints Aeq and beq are given as

 2 1 1 10
Aeq = 0 0 0 and beq = 0
   
0 0 0  0 

The lower and upper bounds vectors are given by

 0 ∞
LB = 0 and UB = ∞
   
0 ∞

The following Matlab statements are used to solve this linear programming problem.

% Example 8.3-2b
f=[-1;-2;0];
A=[1 1 0;-1 1 0;-2 1 0];
b=[6;2;1];
Aeq=[2 1 1; 0 0 0;0 0 0];
beq=[10;0;0];
LB=[0;0;0];UB=[inf;inf;inf];
x=linprog(f,A,b,Aeq,beq,LB,UB)

>> e8d3d2b

8-13
Optimization terminated successfully.
x=
2.0000
4.0000
2.0000

The solution from the Matlab program is

x1 = 2

x2 = 4

S1 = 2

------------------------------------------------------------------------------------------------------

8-14