Optimization - Problem Set 5 Practice | MATH 0164, Assignments of Optimization Techniques in Engineering

Material Type: Assignment; Class: OPTIMIZATION; Subject: Mathematics; University: University of California - Los Angeles; Term: Fall 2005;

Typology: Assignments

Pre 2010

Uploaded on 08/30/2009

koofers-user-fh9-1
koofers-user-fh9-1 🇺🇸

7 documents

1 / 1

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Math 164, Lecture 2, Vese
Homework #5, due on Wednesday, November 2nd, 2005
Remarks:
Please review Sections 4.4 and 5.2 from the textbook.
Reminder: midterm exam on Wednesday, November 9, 1-2pm (MS 6229). This will be a
closed note and closed book written examination.
Sections covered for the midterm: 1.2-1.5, 2.2-2.3, 3.1, 4.1-4.4, 5.2, and 6.1.
Office hours with the instructor before the midterm: Monday Nov. 7, time 2-4pm, in MS
7620-D. (please note: on Tuesday Nov. 8, the office hours will be only with the T.A. Tristan Roy).
[1] Consider a linear program with the constraints in standard form
Ax =band x~
0.
(a) Recall the definition of a direction of unboundedness.
(b) Prove that if d6=~
0 satisfies Ad =~
0 and d~
0,then dis a direction of unboundedness for
these constraints.
(c) Prove that, if dis a direction of unboundedness for these constraints, then dcannot be a
direction of unboundedness.
(d) Let {d1, .., dk}be directions of unboundedness for these constraints. Prove that a nonzero
vector d=Pk
i=1 αidi, with αi0 is also a direction of unboundedness.
[2] Suppose that a linear program in standard form, with bounded feasible region, has loptimal
extreme points {v1, v2, ..., vl}. Prove that a point is optimal for the linear program if, and only if,
it can be expressed as a convex combination of {v1, v2, ..., vl}.
[3] Convert the following linear program to standard form:
minimize z=x15x27x3, subject to
5x12x2+ 6x35
3x1+ 4x29x3= 3
7x1+ 3x2+ 5x39
x1 2, x2, x3free.
[4] Solve the following linear program using the simplex method (graph the feasible region, and
outline the progress of the solution).
maximize z= 7x1+ 8x2, subject to
4x1+x2100
x1+x280
x140
x1, x20.
[5] Consider the linear program
minimize z=5x17x2, subject to
3x1+ 2x230
2x1+x212
x1, x20.
(a) Draw a graph of the feasible region and determine two linearly-independent directions of
unboundedness.
(b) Convert the linear program to standard form and determine two linearly-independent direc-
tions of unboundedness for this version of the problem. Verify that the directions of unboundedness
satisfy Ad =~
0 and d~
0.
1

Partial preview of the text

Download Optimization - Problem Set 5 Practice | MATH 0164 and more Assignments Optimization Techniques in Engineering in PDF only on Docsity!

Math 164, Lecture 2, Vese Homework #5, due on Wednesday, November 2nd, 2005

Remarks:

  • Please review Sections 4.4 and 5.2 from the textbook.
  • Reminder: midterm exam on Wednesday, November 9, 1-2pm (MS 6229). This will be a closed note and closed book written examination.
  • Sections covered for the midterm: 1.2-1.5, 2.2-2.3, 3.1, 4.1-4.4, 5.2, and 6.1.
  • Office hours with the instructor before the midterm: Monday Nov. 7, time 2-4pm, in MS 7620-D. (please note: on Tuesday Nov. 8, the office hours will be only with the T.A. Tristan Roy).

[1] Consider a linear program with the constraints in standard form Ax = b and x ≥ ~ 0. (a) Recall the definition of a direction of unboundedness. (b) Prove that if d 6 = ~0 satisfies Ad = ~0 and d ≥ ~ 0 , then d is a direction of unboundedness for these constraints. (c) Prove that, if d is a direction of unboundedness for these constraints, then −d cannot be a direction of unboundedness. (d) Let {d 1 , .., dk } be directions of unboundedness for these constraints. Prove that a nonzero vector d =

∑k i=1 αidi, with^ αi^ ≥^ 0 is also a direction of unboundedness.

[2] Suppose that a linear program in standard form, with bounded feasible region, has l optimal extreme points {v 1 , v 2 , ..., vl}. Prove that a point is optimal for the linear program if, and only if, it can be expressed as a convex combination of {v 1 , v 2 , ..., vl}.

[3] Convert the following linear program to standard form:

minimize z = x 1 − 5 x 2 − 7 x 3 , subject to

    

5 x 1 − 2 x 2 + 6x 3 ≥ 5 3 x 1 + 4x 2 − 9 x 3 = 3 7 x 1 + 3x 2 + 5x 3 ≤ 9 x 1 ≥ − 2 , x 2 , x 3 free.

[4] Solve the following linear program using the simplex method (graph the feasible region, and outline the progress of the solution).

maximize z = 7x 1 + 8x 2 , subject to

    

4 x 1 + x 2 ≤ 100 x 1 + x 2 ≤ 80 x 1 ≤ 40 x 1 , x 2 ≥ 0.

[5] Consider the linear program

minimize z = − 5 x 1 − 7 x 2 , subject to

 



− 3 x 1 + 2x 2 ≤ 30 − 2 x 1 + x 2 ≤ 12 x 1 , x 2 ≥ 0. (a) Draw a graph of the feasible region and determine two linearly-independent directions of unboundedness. (b) Convert the linear program to standard form and determine two linearly-independent direc- tions of unboundedness for this version of the problem. Verify that the directions of unboundedness satisfy Ad = ~0 and d ≥ ~0.