EXERCISES 1

1) (The solution is not given.) Given the following programming problems:

max z = -3x1 + 5x2

s. t. (sen 17) x1 – 6 x2 ≤ 2
x1 ≤ 1.5
x2 ≤ 1.5
x1 , x2 ≥ 0

max z = 3x1 - x2
s. t. x1 - x2  ≤ 2
x1, x2 ≥ 0

say if they are linear.

2) You want to purchase quantities x1, x2, x3 of three goods at prices, 5 2, 7 (in euros) respectively
such that the sum of these quantities is maximised, the total expenditure does not exceed 20 and the
product of the quantities x1 and x2 is greater than 11. Write (without solving it) the problem of
constrained maximisation. Is it a linear programming problem?

3) (The solution is not given.) Given the following problems solved in Lecture 1:

max z = x1 + x2
s. t. 2x1 + x2 ≥ 2
x1 + x2 ≤ 3
x1, x2 ≥ 0

min z = x1 + x2
s. t. 2x1 + x2 ≤ 2
x1 + x2 ≥ 3
x1, x2 ≥ 0

max z = x1 + x2
s. t. x1 + 2x2 ≥ 6
x2 ≤ 2
x1, x2 ≥ 0

min z = x2
s. t. x1 + x2 ≥ 3
x2 ≤ 2
x1, x2 ≥ 0

solve each of them as a minimum problem (if maximum) or as a maximum problem (if minimum),
both with the gradient method and (if possible) with the vertices method.

4) Solve by the geometric method the problem:

opt z = -5x1 + 4x2

1
 s. t. x1 + x2 ≤ 4
2x1 + x2 ≤ 5
- x1 + 4x2 ≥ 1
x1, x2 ≥ 0.

5) Solve by the geometric method the problem:

max z = 13x1 + 23x2

5x1 + 15x2 ≤ 480
4x1 + 4x2 ≤ 160
35x1 + 20x2 ≤ 1&190
x 1, x2 ≥ 0.

2
SOLUTIONS

2) We obtain:

max z = x1 + x2 + x3
s.t. 5x1 + 2x2 + 7x3 ≤ 20
x1 ⋅ x2 > 11
x1, x2, x3 ≥ 0.

It is not a linear programming problem because the second constraint contains the product of two
variables.

4)

Since X is limited, we can also use the vertices method:

A = (0, 1/4) B = (0, 4) C = (1, 3) D = (19/9, 7/9)

z (0, 1/4) = 1 z(0, 4) = 16 z(1, 3) = 7 z(19/9, 7/9) = - 67/9

giving the constrained minimum point x* = (19/9, 7/9) and the constrained minimum z* = - 67/9;
the constrained minimum point is x* = (0, 4) and the constrained minimum is z* = 16.

5) The feasible set is defined by the system of inequalities:

5 x1 + 15 x2 ≤ 480
4 x1 + 4 x2 ≤ 160
35 x1 + 20 x2 ≤ 1&190
x 1, x2 ≥ 0.

Having drawn the set X, in order to find the optimal solution we represent a level line and the
gradient:

3
 z = 0 ⇒ 13 x1 + 23 x2 = ⇒ 0 x2 = -13/23 x1
∇z = (13 23)

Hence the solution:

x* = (12, 28) constrained maximum point
z* = z (x*) = 800 constrained maximum.

Since X is limited we can also use the vertices method:

O = (0, 0) A = (0, 32) B = (12, 28) C = (26, 14) D = (34, 0)

z (0, 0) = 0 z(0, 32) = 736 z(12, 28) = 800 z(26, 14) = 660 z(34, 0) = 442

which gives the same solution.