Example 1

consider the function: f(x,y) = x2 — y2

These first derivatives are zero at x* = 0 and y* = 0. The Hessian

matrix of f is: &2 ƒ &2ƒ
&x2 &x &y
2 0
H( f)= &2ƒ &2 ƒ =
0 -2
&y &x &y2

The Hessian matrix of f at (x*,y*) = H

The determinant H1 = 2 (positive), and the determinant
H2 = 2(-2) -0(0) = -4 (negative). Then H is indefinite.
Since this matrix is neither positive definite nor negative definite, the
point (x* = 0, y* = 0) is a saddle point.
 Example 2
Find the critical points of the function:

SOLUTION: The necessary conditions for the existence of

an extreme point a re: ….
(1)

… (2)

From (1) x1 = 0 or (- 4), and from (2) x2 = 0 or (- 8 . Then

3 3
these equations are satisfied at the points:
8 4 4 8
(0,0), (0,- ), (- , 0), and ( - ,- )
3 3 3 3
 To find the nature of these extreme points, we
have to use the sufficiency conditions. The second-
order partial derivatives of f are given by:

The Hessian matrix of f is given

by:

H( f
)=
H(f )
 1.7 CONSTRAINED PROBLEMS
1.7.1 Multivariable Optimization With
Equality Constraints
We consider the optimization of continuous functions
subjected to equality constraints:

Where:

Here m is less than or equal to n; otherwise (if m > n), the

problem becomes overdefined and, in general, there will be no
solution. There are several methods available for the solution of
this problem: The Constrained variation, Jacobian method,
Methods of direct substitution, and Lagrange multipliers.
1.7.1.1 Method of Direct Substitution
For a problem with n variables and m equality
constraints, it is theoretically possible to solve
simultaneously the m equality constraints and
express any set of m variables in terms of the
remaining n — m variables. When these expressions
are substituted into the original objective function,
there results a new objective function involving only
n — m variables. The new objective function is not
subjected to any constraint, and hence its optimum
can be found by using the unconstrained
optimization techniques.
 Example 1

Either x1 or x2 can be eliminated without difficulty.

Solving for x1,
…….
(1)
Substitute for x1 in the Objective Function, the new equivalent
objective function in terms of a single variable x2 is:

The constraint in the original problem has now been eliminated, and
f(x2) is an unconstrained function with one independent variable.
 We can now minimize the new objective function
by setting the first derivative of f equal to zero,
and solving for the optimal value of x2:

f”(x) = 28 (positive), then X* is a local minimum.

Once x2* is obtained, then, x1* can be directly obtained via
the relation (1): , then:

f min = 4(1.071)2 + 5(1.286)2 = 12.85714

 Solving Using Excel
Problem:

B3: = D3^2
C3: = E3^2
F4: =SUMPRODUCT(B4:C4;B3:C3)
F6: = SUMPRODUCT(D6:E6;D3:E3)
 Example 2
The profit analysis model:
Max the profit z = v.p – cf – v.cv …….….. (1)
The demand is represented by: v = 1,500 – 24.6p ………... (2)
Where: v = volume (quantity), p = price,
cf = fixed cost = $10,000, cv = variable cost = $8 per unit.
Substituting values of cf and cv into (1), weobtain:
z = v.p -10,000 – 8v …..…… (3)
Substituting (2) in (3):
z = 1500p – 24.6p2 – 10,000 – 8(1,500 – 24.6p)
z = 1696.8p -24.6p2 -22,000 …..……. (4)
dz
= 1696.8p -49.2p = 0 for the critical points, then:
dp
p* = 34.49
d2z
= - 49.2 (negative), then p* is a local maximum.
dp 2
Substituting in (2): v* = 1500 – 24.6(34.49) = 651.55
Substituting in (3): zmax = (651.55)( 34.49) – 10,000 – 8(651.55) = 7259.56
 1.7.1.2 Lagrange Method

• The basic features of the Lagrange multiplier

method is given initially for a simple problem of
two variables with one constraint.
• The extension of the method to a general
problem of n variables with m constraints is
given later.
Problem with Two Variables and One Constraint.

Consider the proble m

subject to: :
Define Lagrange functio n
:
λ is called the Lagrange multiplier.
L is treated as a function of the three variables x1, x2, and λ.
Theorem: Necessary Conditions for Extremum:
 Theorem: Sufficient Condition
A sufficient condition for f(X) to have a relative
minimum at X* is that the quadratic, Q, defined by:
62L
Q= d 1d 2
6s16s2
evaluated at X = X* must be positive definite for all values
of dX for which the constraints are satisfied.
- If Q is negative definite for all choices of the admissible
variations dX, X* will be a constrained maximum of f(X).
-It has been shown by Hancock that a necessary condition
for the quadratic form Q, to be positive (negative) definite
for all admissible variations dX is that each root of the
polynomial Z, defined by the following determinantal
equation, be positive (negative):
 0

1
 Example
Find the solution of the following problem using the
Lagrange multiplier method:
f(x,y) = x-1y-2
Subject to: g(x,y) = x2 + y2 - 4 = 0
The Lagrange function is:
L(x,y,λ) =f(x,y) + λg(x,y) = x -1y -2 + λ(x2 + y2 - 4 )
The necessary conditions for the extreme of f(x, y) give:
&L 1
&x
= -x-2y-2 +2λx = 0 ……. (1) λ = x-3y -2……… (4)
2
&L
= -2x-1y-3 +2λy = 0 ……. (2) λ = x-1 y-4 ……… (5)
&y
&L
= x2 +y2 -4 = 0 ……. (3) From 4 , 5 : 1 x -3y -2 = x -1y -4
&λ 2
1 1 -2 1
x -3 y -2 = x-1y -4 y = x2y-4 y2 = x2
2 2 2
1 1 2
X* = y* Or: x2 = y ………… (6)
2 2
(3): x2 +y2 - 4 = 0
1 2
From (6): y2 +y2 = 4 y2 = 8 y* = 2
2 3 3
2
Substituting in (6): x* =
3

3 1 81
(5): λ = x -1 y -4 λ* = =9 3
2 16 16 128
The determinant:
0

-z 0
1

-z
=0

1
0
1.218 -z 0.344 2.309

0.344 0.974 -z 3.266 =0

2.309 3.266 0
 1.218-z 0.344 2.309

0.344 0.974-z 3.266 =

2.309 3.266 0

0.974-z 3.266 0.344 3.266 0.344 0.974-z

(1.218-z) - (0.344) + 2.309
3.266 0 2.309 0 2.309 3.266

(1.218-z) (0.974- z) (0) – (3.266)(3.266) - (0.344) -(3.266)(2.309) +

+ (2.309) (0.344)(3.266) – (0.974-z)(2.309) =0

Z= - 0.344 (negative), Then x*,y* is a relative maximum

f*(x*,y*) = 0.3248
Determinant Calculation

1
 R2 R2 + 8R1

-3[1(1)-2(0)] = -3(1) = -3
http://matrix.reshish.com/determinant.php
Necessary Conditions for a General Problem

The equations can be extended to the case of a general

problem with n variables and m equality constraints:
=f(x1, x2, ……, xn)
subject to:
The Lagrange function, L, in this case is defined by introducing one
Lagrange multiplier λj for each constraint gj (X) as:

The necessary conditions for the extremum of L, are given by:

 Sufficient Condition
A sufficient condition for f(X) to have relative minimum
at X* is that the quadratic, Q, defined by:

evaluated at X = X* must be positive definite for all values

of variations dX for which the constraints are satisfied.
If Q is negative definite for all choices of the admissible
variations dxi, X* will be a constrained maximum of f(X).
It has been shown by Hancock that a necessary condition for the
quadratic form Q, to be positive (negative) definite for all
admissible variations dX is that each root of the polynomial Zi,
defined by the following determinantal equation, be positive
(negative):
 Where:

,
This equation on expansion, leads to an (n — m)th-order polynomial
in z. If some of the roots of this polynomial are positive while the
others are negative, the point X* is not an extreme point.
 1.7.2 Multivariable Optimization With
Inequality Constraints
Consider the following problem: Minimize f(X)
subject to: gj(X) ≤ 0, j = 1,2,. . .,m

Kuhn-Tucker Conditions
The conditions to be satisfied at a constrained minimum
point, X*. These conditions are, in general, not sufficient to
ensure a relative minimum. However, there is a class of
problems, called convex programming problems for which
the Kuhn-Tucker conditions are necessary and sufficient for
a global minimum.
 Kuhn-Tucker Conditions
The Kuhn-Tucker conditions can be stated as follows:

Note that if the problem is one of maximization or if the

constraints are of the type gj ≥ 0, the λj have to be
nonpositive. On the other hand, if the problem is one of
maximization with constraints in the form gj ≥ 0, the λj have
to be nonnegative.
Types of Nonlinear Programming
▪ Nonlinear objective function,
linear constraints.
▪ Nonlinear objective function
and nonlinear constraints.
▪ Linear objective function and
nonlinear constraints.
32
 Nonlinear Objective Function and Linear
Constraints:
The Great Western Appliance Company sells two models of toaster ovens,
the Micro toaster (X1) and the Self-Clean Toaster Oven (X2).
The firm earns a profit of $28 for each Micro toaster regardless of the
number sold. Profits for the Self-Clean model, however, increase as more
units are sold because of fixed overhead. Profit on this model may be
expressed as 21 X2 + 0.25 X 22.
Great Western’s profit is subject to two linear constraints on production
capacity and sales unit time available.
Max: 28X1 + 21X2 + 0.25X22
Subject to:
X1 + X2 ≤ 1000 (units of production capacity),
0.5X1 + 0.4X2 ≤ 500 (hours of sales time available).
X1, X2 ≥0.
When an objective function contains a squared term and the problem
constraints are linear, it is called a quadratic programming problem.
33
An
AnExcel
ExcelFormulation
Formulationof Great Western
of Great Western
Appliance’sNonlinear
Appliance’s NonlinearProgramming
ProgrammingProblem.
Problem.

34
 Max: 28X1 + 21X2 + 0.25X22
Input Screen X1 + X2 ≤ 1000
0.5X1 + 0.4X2 ≤ 500

35
36
Solution
Solutionto
toGreat
GreatWestern Appliance’s
Appliance’s
Western NLPProblem
NLPProblem
using
using Excel
ExcelSolver:
Solver:

37
 Both Nonlinear Objective Function and Nonlinear
Constraints.

• The annual profit at a medium-sized (200-400 beds) Hospital

Corporation-owned hospital depends on the number of medical
patients admitted (X1) and the number of surgical patients
admitted (X2). The nonlinear objective function for
Hospicare is:
• Max. 13X1 + 6X1X2 + 5X2 +1/X2.
• The corporation identifies three constraints, two of which are also
nonlinear, that affect operations. They are
• 2X12+ 4X22 ≤ 90 (nursing capacity, in thousands of labor-
8X 1 –
days).X 2 1 2 ≤ 61 (marketing budget required, in thousands of
X +
3
2X ≤ 75 (x-ray capacity, in thousands).
$). 38
Max: 13X1 + 6X1X2 + 5X2 + 1/X2
Subject to:
2X1 2 +4X22 ≤ 90
X1 + X23 ≤ 75
8X1 – 2X2 ≤ 61

39
 AnExcel
An Excel Formulation
Formulation of Hospicare
of Hospicare Corp.’s
Corp.’s NLP
NLP Problem:
Problem:

Max: 13X1 + 6X1X2 + 5X2 + 1/X2

2X12 + 4X22 ≤ 90
X1 + X23 ≤ 75 40

8X1 – 2X2 ≤ 61
41
42
 Linear Objective Function with Nonlinear
Constraints
Thermlock Corp. produces massive rubber washers and gaskets
like the type used to seal joints on the NASA Space Shuttles. To do
so, it combines two ingredients; rubber (X1) and oil (X2).
The cost of the industrial quality rubber used is $5 per pound and
the cost of the high viscosity oil is $7 per pound. Two of the three
constraints Thermlock faces are nonlinear. The firm’s objective
function and constraints are

Min. 5X1 + 7X2

Subject to:
3X1+ 0.25X12 + 4X2 + 0.3X22 ≥ 125 (hardness constraint),
13X1 + X13 ≥ 80 (tensilestrength),
0.7X1 + X2 ≥ 17 (elasticity).
43
Excel
ExcelFormulation
Formulationofof Thermlock’s
Thermlock’s NLPNLP
Problem:
Problem:

44
Solutionto
Solution Thermlock’s
toThermlock’s NLP Problem
NLP Using
Problem Using
Excel Solver:
Excel Solver:

45
 Computational Procedures -Nonlinear
Programming
Unlike LP methods:
• One disadvantage of NLP is that the solution procedures
to solve nonlinear problems do not always yield an
optimal solution in a finite number of steps. The solution
yielded may only be a local optimum, rather than a
global optimum. In other words, it may be an optimum
over a particular range, but not overall.
• There is no general method for solving all nonlinear
problems.
• Classical optimization techniques based on calculus, can
handle some special cases, usually simpler types of
problems.
46
 Gradient method (steepest ascent method)

It is an interactive procedure that

moves from one feasible solution to the
next in improving the value of the
objective function.
It has been computerized and can
handle problems with both nonlinear
constraints and objective.

47
 Separable programming
▪ Linear representation of nonlinear
problem.
▪ Separable programming deals with a
class of problems in which the
objective and constraints are
approximated by linear functions.
In this way, the powerful simplex
algorithm may again be applied.
48
In general, work in the area of
NLP is the least charted and
the most difficult of all the
quantitative analysis models.

49