Scribd
Upload
5 views

Karmarkars Algorithm For Linear Programming

Uploaded by

threemonthsmaster
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
5 views

Karmarkars Algorithm For Linear Programming

Uploaded by

threemonthsmaster
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 13

Karmarkar's

Algorithm for
Linear
Programming
Karmarkar's algorithm is a powerful method for solving linear programming
problems, offering a significant improvement over traditional simplex
methods.

GROUP : E6

PRAGATI BAJAJ(2023UEE2012)

ANSHIKA BANSAL(2023UEE2013)

SHIVAM SHEKHAWAT(2023UEE2015)

VINEET SINGH PARIHAAR(2023UEE2016)


Input and Output
Input Output

The input to Karmarkar's algorithm is a linear The output is the optimal solution x* that maximizes
programming problem in the standard form: Maximize the objective function while satisfying all constraints.
ca, subject to Ax < b, x ≥ 0.
Algorithm Steps

1 Initialization
Choose an initial feasible interior point x0 such that Ax0 <= b and x0 > 0. If no feasible point is known, perform a Phase 1 process to find an initial feasible point.

2 Linear Transformation
Apply a projective transformation to the problem to map the feasible region to a simpler geometric form (often a ball in a transformed space). This transformation simplifies the geometry and makes the problem easier to solve in
subsequent steps.

3 Iterative Procedure
For k = 1, 2, 3,... (until convergence):

4 Compute the Direction dk


Calculate the direction of movement dk towards the center of the feasible region using the central path method. The direction is based on the gradient of the objective function and the geometry of the transformed feasible region.

5 Compute the Step Size αk


Determine the optimal step size αk by solving the following:

αk = min(xk/dk)

where xk is the current point and dk is the direction of movement.This ensures the updated point remains feasible (i.e., it stays inside the feasible region).

6 Update the Point


Update the current point to the new point xk+1 using the formula: xk+1 = xk + αkdk. This step ensures that the new point is closer to the optimal solution while remaining within the feasible region.

7 Termination
Repeat steps 3a-3c until the algorithm converges. Convergence is typically determined when the objective function value no longer improves significantly, or the change in the current point xk is very small, satisfying a predefined tolerance.

resulting xk is the optimal solution x*.


Example 1: Resource
Allocation Problem
This example demonstrates the application of Karmarkar's
algorithm to a resource allocation problem. A company produces
two products, P1 and P2, using two resources, R1 and R2. The goal
is to maximize profit by determining the optimal production
quantities of each product, subject to resource constraints.

LP Formulation

The problem is formulated as a linear program: Maximize Z = 3x1


+ 4x2, subject to 2x1 + x2 ≤ 100 (Resource R1), x1 + 2x2 ≤ 80
(Resource R2), x1, x2 ≥ 0.
CODE FOR QUESTION(KARMAKAR's METHOD)

SOLUTION(KARMAKAR'S METHOD)
USING SIMPLEX METHOD
Example: Supply Chain Optimization Problem
Problem Description: A company operates several warehouses, each with a
certain storage capacity, and ships products to a number of retail stores. The
objective is to determine the optimal shipping plan that minimizes the total
transportation cost while meeting the demand at each store and respecting
the capacity limits at the warehouses
CODE FOR QUESTION(KARMAKAR's METHOD)

SOLUTION(KARMAKAR'S METHOD)
USING SIMPLEX METHOD

SOLUTION
Advantages of Karmarkar's
Algorithm
1 Polynomial Time 2 Improved Convergence
Complexity
It often converges faster than
Karmarkar's algorithm has a simplex methods, especially
polynomial time complexity, for problems with many
making it more efficient for constraints or variables.
large-scale linear programming
problems compared to simplex
methods, which can have
exponential time complexity in
the worst case.

3 Handling Degeneracy
Karmarkar's algorithm is less susceptible to degeneracy issues, which
can cause simplex methods to cycle and fail to converge.
Applications of
Karmarkar's Algorithm

Resource Allocation Production Planning


Optimizing the allocation of Determining the optimal production
resources, such as labor, materials, schedule and quantities to meet
and capital, to maximize efficiency demand while minimizing costs.
and profitability.

Transportation Finance
Finding the most efficient routes Optimizing investment portfolios,
and transportation plans to managing risk, and making financial
minimize costs and delivery times. decisions.
Conclusion
Karmarkar's algorithm is a powerful tool for solving linear
programming problems, offering significant advantages in terms of
efficiency, convergence, and handling degeneracy. Its applications
extend across various fields, making it a valuable technique for
optimization.
THANK YOU!

You might also like