Scribd
Upload
61 views6 pages

Practice Problem Set 3: OA4201 Nonlinear Programming

This document provides the solutions to 5 practice problems evaluating potential local minima for nonlinear functions using first and second order necessary/sufficient conditions. For each problem it is determined whether the given options can be ruled out, confirmed, or their status is inconclusive based on applying the conditions to the derivatives evaluated at each potential minimum.

Uploaded by

fatihy73
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
61 views6 pages

Practice Problem Set 3: OA4201 Nonlinear Programming

This document provides the solutions to 5 practice problems evaluating potential local minima for nonlinear functions using first and second order necessary/sufficient conditions. For each problem it is determined whether the given options can be ruled out, confirmed, or their status is inconclusive based on applying the conditions to the derivatives evaluated at each potential minimum.

Uploaded by

fatihy73
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/ 6

Practice Problem Set 3

OA4201 Nonlinear Programming


Use the first order necessary condition, the second order necessary condition, and the
second order sufficient condition to evaluate the given solutions as potential local minima
for each of the following functions.
 

1. f(x) = x4 – 2x3
(a) x* = 0
(b) x* = 1
(c) x* = 3/2 

2. f(x) = ex – x
(a) x* = -1
(b) x* = 0
(c) x* = 1 

3. f(x) = x12 + x1x2 + (x2 – 1)2


(a) x* = [0 0]’
(b) x* = [0 1]’
(c) x* = [-2/3 4/3]’  

4. f(x) = (x1-5)2 + x1(x2 – 3)2


(a) x* = [0 0]’
(b) x* = [0 3+sqrt(10)]’
(c) x* = [5 3]’ 
 

5. f(x) = x12 + x1x2x32 + x2 x3


(a) x* = [0 0 0]’
(b) x* = [0 1 0]’
(c) x* = [1 1 1]’ 
Practice Problem Set 3 Solution

OA4201 Nonlinear Programming


Use the first order necessary condition, the second order necessary condition, and the
second order sufficient condition to evaluate the given solutions as potential local minima
for each of the following functions.
 

1. f(x) = x4 – 2x3

f '(x) = 4x3 – 6x2

f ''(x) = 12x2 – 12x

(a) x* = 0

f '(x*) = 0: First order necessary condition satisfied.

f ''(x*) = 0: Second order necessary condition satisfied (since y*0*y > 0 for all y).
Second order sufficient condition not satisfied (since y*0*y = 0 for all y).

x* = 0 cannot be ruled out as a local minimum, and also cannot be confirmed using these
conditions.

(b) x* = 1

f '(x*) = -2: First order necessary condition not satisfied.

x* = 1 can be ruled out as a local minimum.

(c) x *= 3/2

f '(x*) = 0: First order necessary condition satisfied

f ''(x*) = 9: Second order necessary condition satisfied (since y*9*y > 0 for all y).
Second order sufficient condition satisfied (since f '(x*) = 0 and y*9*y > 0 for
all y ≠ 0).

x* = 3/2 can be confirmed as a local minimum.


 
 
2. f(x) = ex – x

f '(x) = ex – 1

f ''(x) = ex

(a) x* = -1

f '(x*) = -.6321...: First order necessary condition not satisfied.

x* = -1 can be ruled out as a local minimum.

(b) x* = 0

f '(x*) = 0: First order necessary condition satisfied.

f ''(x*) = 1: Second order necessary condition satisfied (since y*1*y > 0 for all y).
Second order sufficient condition satisfied (since f '(x*) = 0 and y*1*y > 0 for
all y ≠ 0).
x* = 0 can be confirmed as a local minimum.

(c) x* = 1:

f '(x*) = 1.718...: First order necessary condition not satisfied.

x* = 1 can be ruled out as a local minimum.

 
3. f(x) = x12 + x1x2 + (x2 – 1)2

2

2 –1

2 1

1 2

(a) x* = [0 0]’

∗ 0


2
First order necessary condition not satisfied. x* = [0 0]’ can be ruled out as a local
minimum.

(b) x* = [0 1]’

∗ 1


0
First order necessary condition not satisfied. x* = [0 1]’ can be ruled out as a local
minimum.

(c) x* = [-2/3 4/3]’  


 
∗ 0


0
First order necessary condition satisfied.
 
 
∗ 2 1

1 2

Check second order necessary and sufficient conditions:

∗ 2 1
 2 2 2
1 2
∗
 0 ∀ , so second order necessary condition is satisfied.
∗
f (x*) = 0 and  0 ∀ 0, so second order sufficient condition is
satisfied. x* = [-2/3 4/3]’ can be confirmed as a local minimum.
(This one is also easy to do using eigenvalues!)
4. f(x) = (x1-5)2 + x1(x2 – 3)2

2 5 –3

2 –3

2 2 3

2 3 2

(a) x* = [0 0]’

∗ 1

0
First order necessary condition not satisfied. x* = [0 0]’ can be ruled out as a local
minimum. 

(b) x* = [0 3+sqrt(10)]’

∗ 0


0
First order necessary condition satisfied. 

∗ 2 2√10

2√10 0

∗ 2 2√10
 2 4√10
2√10 0

Second order necessary condition not satisfied for y1 = 1, y2 = -1. x* = [0 3+sqrt(10)]’ 


can be ruled out as a local minimum. (Again, this one is easy to do with eigenvalues.)

(c) x* = [5 3]’ 
 
∗ 0

0
First order necessary condition satisfied. 
 
∗ 2 0

0 10
By inspection, the second order necessary and sufficient conditions are satisfied (since
∗
 is diagonal with positive diagonal terms, and since the first order necessary
condition holds). Therefore, x* = [5 3]’ can be confirmed as a local minimum.
5. f(x) = x12 + x1x2x32 + x2 x3

2

2

2 2
 0 2 1
2 2 1 2

(a) x* = [0 0 0]’  
 
0
∗
0
0
First order necessary condition satisfied.

2 0 0
∗
 0 0 1
0 1 0
 
∗
 2 0 for 0 1 1 ′ 
Second order necessary condition not satisfied. x* = [0 0 0]’ can be ruled out as a local
minimum.
 

(b) x* = [0 1 0]’ 
 
0
∗
0
1
First order necessary condition not satisfied. x* = [0 1 0]’ can be ruled out as a local
minimum.
 

(c) x* = [1 1 1]’ 
 
3
∗
2
3
First order necessary condition not satisfied. x* = [1 1 1]’ can be ruled out as a local
minimum.

You might also like