Scribd
Upload
4 views

Optimization I Problem Set 2

Problemas de Optimización lineal

Uploaded by

julio_rocha_1
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
4 views

Optimization I Problem Set 2

Problemas de Optimización lineal

Uploaded by

julio_rocha_1
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Optimization I

Problem set 2

1. Solve the following problems graphically.

maximize 3x1 ` 2x2 minimize ´ 4x1 ` 7x2

subject to 2x1 ` x2 ď 10, subject to x1 ` x2 ě 3,


(a) (d)
x1 ` x2 ď 8, ´ x1 ` x2 ď 3,
x1 ď 4, 2x1 ` x2 ď 8,
x1 , x2 ě 0. x1 , x2 ě 0.

maximize 4x1 ` x2 maximize 2x1 ´ x2

(b) subject to 4x1 ` x2 ď 8, (e) subject to x1 ´ x2 ď 1,


5x1 ` 2x2 ď 12, 2x1 ` x2 ě 6,
x1 , x2 ě 0. x1 , x2 ě 0.

maximize 3x1 ` 2x2 maximize 5x1 ` 6x2

(c) subject to 3x1 ` 2x2 ď 12, (f) subject to x1 ´ 2x2 ě 1,


x1 ` x2 ď 7, ´ 2x1 ` x2 ě 1,
x1 ě 3, x2 ě 2. x1 ď 0.

2. Determine the extreme points and extreme directions of the following subset of R4 :

S “ tpx1 , x2 , x3 , x4 q P R4 : ´x1 ` x2 ` x3 “ 2, x1 ´ 2x2 ` x4 “ 2, xj ě 0, j “ 1, 2, 3, 4u

3. Solve the following problems using the simplex algorithm:

minimize ´ 5x1 ´ 4x2 ´ 3x3 maximize 3x1 ` 2x2 ` x3

subject to 2x1 ` 3x2 ` x3 ď 5, subject to 2x1 ` x2 ` x3 ď 12,


(a) (b)
4x1 ` x2 ` 2x3 ď 11, 3x1 ` 4x2 ď 11,
3x1 ` 4x2 ` 2x3 ď 8, x1 ` x2 ě 4,
x1 , x2 , x3 ě 0. x1 , x2 , x3 ě 0.

Bachelor in Applied Mathematics, IE University Page 1 of 3


[email protected]
Optimization I: Sheet 2

maximize 3x1 ` 2x2 minimize x1

(c) subject to 3x1 ` 2x2 ď 12, (e) subject to x1 ´ x2 ď 2,


10x1 ` 3x2 ď 30, 2x1 ` 2x2 ě 1,
x1 , x2 ě 0. x1 , x2 ě 0.

maximize 2x1 ` x2 minimize ´ 2x1 ´ 3x2

(d) subject to x1 ` x2 ě 2, (f) subject to x1 ` x2 ě 3,


x1 ` x2 ď 5, ´ x1 ` 2x2 ě ´4,
x1 ě 1, x2 ě 0. x1 , x2 ě 0.

4. The simplex method is a typically efficient algorithm for solving linear programs. How-
ever, no version of it has been shown theoretically to solve any linear program in poly-
nomial time. In 1972, Klee and Minty came up with an example in n dimensions that
requires 2n ´ 1 iterations. These problems are of the form
n
ÿ
maximize 10n´j xj
j“1
i´1
ÿ
subject to 2 10i´j xj ` xi ď 100i´1 , i “ 1, 2, . . . , n,
j“1

xj ě 0, j “ 1, . . . , n.

Use the simplex method to solve the three-dimensional Klee–Minty problem with the
basic feasible solution corresponding to x1 “ x2 “ x3 “ 0 as your initial solution. It
should take exactly seven iterations.

5. While solving a standard form problem (minimization), we arrive at the following


tableau, with x3 , x4 , and x5 being the basic variables:

x1 x2 x3 x4 x5
x3 ´1 1 1 0 0 4
x4 α ´4 0 1 0 1
x5 γ 3 0 0 1 β
δ 2 0 0 0 10

Bachelor in Applied Mathematics, IE University Page 2 of 3


[email protected]
Optimization I: Sheet 2

The tableau entries α, β, γ, δ are unknown parameters. Find some parameter values for
each of the following statements that will make the statement true.
(a) The current solution is optimal and there are multiple optimal solutions.
(b) The optimal cost is ´8.
(c) The current solution is feasible but not optimal.

6*. Consider the problem of minimizing cJ x over a polyhedron P . A vector d is said to be


a feasible direction at x if there exists a positive scalar λ for which x ` λd P P . Prove
that a feasible solution x is optimal if and only if cJ d ě 0 for every feasible direction d
at x. Specifically, x is the unique optimal solution if and only if the inequality is strict.

Bachelor in Applied Mathematics, IE University Page 3 of 3


[email protected]

You might also like