Scribd
Upload
12 views3 pages

NLP With Constraints and Duality in LPP

Non linear programming

Uploaded by

Giridhar meda
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
12 views3 pages

NLP With Constraints and Duality in LPP

Non linear programming

Uploaded by

Giridhar meda
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

Process Optimization

Module III
Duality and NLP with constraints
Examples and Practice problem • Monsoon 2024-25

1. Duality in Linear Programmming Problem (LPP)

Duality Theorem:
If the primal problem has an optimal solution and so does the dual, the optimal values
of the respective objective function are equal.

1.1 Practice Problem:


Convert the following problem into its dual form.

Minimize 10x1 + 20x2


subject to:
3x1 + 2x2 ⩾ 18
x1 + 3x2 ⩾ 8
2x1 − x2 ⩽ 6

2. Non-Linear Programming (NLP) problem with constraints

One way is to solve explicitly for one variable and eliminate all the other variables
from the formulation as done in Direct substitution method. In Lagrange method, the
problem is transformed in terms of the Lagrangian function L(X, λ) and Lagrange mul-
tipliers (λ).

2.1 Direct Substitution: Example problem

Minimize f(X) = 4x1 2 + 5x2 2


subject to:
2x1 + 3x2 = 6

Step 1: Solve for x2 and eliminate x1 from the formulation.

6−3x2
x1 = 2

The objective function is modified accordingly,

Minimize f(X) = 4 × ( 6−3x


2
2 2
) + 5(x2 )2
= 14(x2 )2 − 36x2 + 36 = f(x2 )

1
Examples and Practice problem Duality and NLP with constraints

Step 2: Apply necessary condition with respect to x2

∂f(x2 )
=0 (1)
∂x2

Applying the necessary condition to the equation derived in 2.1, we get

28x2 − 36 = 0 (2)

36
and solving that, we get x∗2 = 28 = 1,286
Step 3: Evaluate x1 and optimal value of the objective function f(x∗1 , x∗2 )
∗

By back-substitution, we get

6 − 3x∗2
x∗1 = = 1,072 (3)
2
and Optimal objective function value,

f(x∗1 , x∗2 ) = 12,8571 (4)

2.2 Direct Substitution: Practice problem

Minimize f(X) = 3x1 2 + 4x2 2 + x1 x3 + x2 x3


subject to:
x1 + 3x2 = 3
2x1 + x3 = 7

2.3 Lagrange multiplier: Example problem

Minimize f(X) = 4x1 2 + 5x2 2


subject to:
2x1 + 3x2 = 6

Step 1: Transform the original problem in terms of the Lagrangian function. In ge-
neral for a problem with only equality constraints, the Lagrangian function is writ-
ten as,

X
m
L(X, λ) = f(x) + λi × [hi − bi ] (5)
i=1

For the given problem, the Lagrangian function expressed in Equation 5 can be
defined as,

L(X, λ) = 4x1 2 + 5x2 2 + λ × (2x1 + 3x2 − 6) (6)

Step 2: Apply Necessary condition to the Lagrangian function given in Equation 6


with respect to the two variables x1 , x2 , and the Lagrange multiplier λ, we get:

∂L(X, λ)
= 8x1 + 2λ = 0 (7)
∂x1

2
Duality and NLP with constraints Examples and Practice problem

∂L(X, λ)
= 10x2 + 3λ = 0 (8)
∂x2
∂L(X, λ)
= 2x1 + 3x2 − 6 = 0 (9)
∂x1
Solving the above three Equation 7, 8, and 9, simultaneously, we get:
x1 = 15/14 = 1,071, x2 = 9/7 = 1,286, and λ = −30/7 = −4,286

Step 3: Calculate the optimal value of objective function by substituting the values
of x1 and x2 in the original formulation. The optimal value of objective function is
12,8571.

2.4 Lagrange multiplier: Practice problem

Minimize f(X) = 3x1 2 + 4x2 2 + x1 x3 + x2 x3


subject to:
x1 + 3x2 = 3
2x1 + x3 = 7

You might also like