Scribd
Upload
204 views3 pages

Operations Research Exam Guide

This document outlines the sections and questions for an exam on linear programming and nonlinear programming concepts. It includes questions testing basic concepts, solving a minimum cost network flow problem, using the revised simplex method to solve an LP, applying the Frank-Wolfe algorithm to solve a quadratic program, graphically solving an NLP, and checking if an NLP solution satisfies the Kuhn-Tucker conditions.

Uploaded by

Daninson Camacho
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
204 views3 pages

Operations Research Exam Guide

This document outlines the sections and questions for an exam on linear programming and nonlinear programming concepts. It includes questions testing basic concepts, solving a minimum cost network flow problem, using the revised simplex method to solve an LP, applying the Frank-Wolfe algorithm to solve a quadratic program, graphically solving an NLP, and checking if an NLP solution satisfies the Kuhn-Tucker conditions.

Uploaded by

Daninson Camacho
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

IE505 Final Exam (Take Home)

Spring 2021

1 Basic Concepts (15 points)


Indicate whether True or False.
1. Every global optimum is a local optimum. T or F
2. A totally unimodular constraint matrix is characteristic of every minimum
cost flow problem. T or F
3. A linear program in standard form cannot have nonnegativity constraints.
T or F
4. A graphical solution of LPs and NLPs involving many variables and/or
many constraints is generally impractical. T or F
5. The Phase I method is used to find an initial basic feasible solution for an
LP in standard form. T or F
6. An initial basic feasible solution may be found by inspection (after the
insertion of slack variables) when the LP problem involves constraints in
the form Ax ≤ b with b > 0.
7. an LP displays degeneracy when some basic feasible solution xB contains
xBi = 0. T or F
8. Degeneracy is important because it is theoretically possible that the sim-
plex will not converge (because it revisits some previous bfs). T or F
9. As presented in class and lecture, the Kuhn-Tucker conditions are nec-
essary conditions. That is, every optimal solution must satisfy them for
the NLP of interest (and the appropriate assumptions regarding differen-
tiability are satisfied). T or F
10. The direction vectors used in the Frank-Wolfe algorithm obey
dk = y k − xk
when y k is the solution of the linear approximating program during itera-
tion k and xk is the current working value of the optimal solution in that
iteration. T or F

1
2 Network Simplex (15 points)
Consider the following netwrok:

[]

Figure 1: The network for question 2

First, please find an initial solution and find the minimum-cost flow if the
we are moving 7 unit products from 0 to 5 and 6 unit of products from 0 to 4.

3 Revised Form of the Simplex (20 points)


Consider the following Linear programming problem.
min −2x1 − x2
s.t. x1 + x2 ≥ 5
3x1 + x2 ≤ 6
2x2 ≤ 11
xi ≥ 0
Please solve this problem to optimality

4 Frank-Wolfe Algorithm (20 points)


Consider this program:
1 2 1 2
min (x1 − 5) + (x2 − 5)
2 2

2
s.t. 0 ≤ x1 ≤ 2
0 ≤ x2 ≤ 2

Solve by the Frank-Wolfe algorith using x0 = 0 0 as your initial solution.

5 NLP Graphical Solution (15 points)


Consider this program:
1 2 1 2
min (x1 − 5) + (x2 − 5) (1)
2 2

s.t. x1 + x2 − 2 ≤ 0 (2)
−x1 ≤ 0 (3)
−x2 ≤ 0 (4)

Solve it graphically. Indicate all Kuhn-Tucker points, local optima, and global optima.

6 Kuhn-Tucker Conditions (15 points)


Then show your solution of mathematical program (1) - (4) satisfies the Kuhn-
Tucker conditions.

You might also like