Scribd
Upload
9 views

Lecture 1

- Linear programming (LP) involves optimizing a linear objective function subject to linear constraints. It has wide applications and is one of the most widely applied methodologies. - The simplex method and ellipsoid method provide polynomial-time algorithms for solving LP problems. More recently, the interior point method was developed, which walks through the interior of the feasible region to find an optimal solution. - LP lays the foundation for more advanced optimization topics like nonlinear programming, integer programming, and robust optimization. Current research integrates interior point methods with techniques like decomposition and parallelization to solve very large real-world LP problems.

Uploaded by

Huo Zimu
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
9 views

Lecture 1

- Linear programming (LP) involves optimizing a linear objective function subject to linear constraints. It has wide applications and is one of the most widely applied methodologies. - The simplex method and ellipsoid method provide polynomial-time algorithms for solving LP problems. More recently, the interior point method was developed, which walks through the interior of the feasible region to find an optimal solution. - LP lays the foundation for more advanced optimization topics like nonlinear programming, integer programming, and robust optimization. Current research integrates interior point methods with techniques like decomposition and parallelization to solve very large real-world LP problems.

Uploaded by

Huo Zimu
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 24

LECTURE 1:

INTRODUCTION

Shu-Cherng Fang
North Carolina State University – Fall 2011
Outline
• What is Linear Programming?
• Why to study Linear programming?
• How to study Linear Programming?
• History of Linear Programming
• How to solve an LP problem?
• Where to go?
What is Linear Programming (LP)?
• Optimize a linear objective function of
decision variables subject to a set of linear
constraints.
• Example
Graphic representation
General form
Why study LP ?
• Wide applications:
- One of the most widely applied methodologies.
- 85 % of Fortune 500 companies had used LP models.
• Passage to advanced subjects:
- Nonlinear Programming
- Network Flows
- Integer Programming
- Conic Programming
- Semidefinite Programming
- Robust Optimization
Reference: Operations Research, Vol. 50, No. 1, 2002
Example – manpower allocation

• = man-hours of skill i working on project j


How to study LP ?
• Geometric intuition
• Algebraic manipulation
• Computer programming
History of Linear Programming
• Conceived by G. B. Dantzig (1947).
– mechanized planning tool for a deployment, training,
and logistical supply program in USAF.
• Named by T. C. Koopmans & G. B. Dantzig (1948).
• Simplex Method proposed by G. B. Dantzig (1948).
- L. V. Kantorovich worked on a restricted class in 1939.
• Kantorovich & Koopmans received the Nobel Prize in
Economic Science (1975).
– Theory of optimum allocation of resources.
History of Linear Programming
• Ellipsoid Method proposed by L. G.
Khachian (1979).
– First polynomial-time algorithm for LP.
• Interior-Point Method proposed by N.
Karmarkar (1984).
– First “good” polynomial-time algorithm
for LP.
How to solve an LP problem?
• What’s special about LP?
- linear constraints shape the feasible domain as a
convex polyhedral set with a finite number of vertices.
- linear objective function
provides a linear
contour of each
fixed value.
Fundamental theorem of LP
• For a linear programming problem, if
its feasible domain is not empty, then
its optimum is either unbounded or is
attained at least at one vertex of the
feasible domain.
How to find an optimal solution?
• Fact:

• Question: How to find ?


Different solution methods
• Enumeration Method
– Find all vertices and choose the optimal one by
comparison.
Simplex Method
• Basic ideas: (walking on the boundary)
Step 1: Start at a vertex.
Step 2: If current vertex is optimal, STOP! Otherwise,
Step 3: Move to a better neighboring vertex. Go To
Step 2.
Is Simplex Method Good?
Is there a polynomial-time algorithm for
LP?
• Yes! - Ellipsoid Method by L. G. Khachian 1979.

Basic ideas:
Ellipsoid method
Ellipsoid method
Is ellipsoid method good?
Is there a good polynomial-time algorithm
for LP?
• Yes! - Interior Point Method by Karmarkar in 1984.
• Basic ideas: (walk through interior)

Step 1: Start with an interior


solution.
Step 2: If current solution is
good enough, STOP. Otherwise,
Step 3: Check all directions for
improvement and move to a
better interior solution. Go
to Step 2.
Is interior point method good?
Where to go?
• Integration of Interior-point methods to develop hybrid
algorithms for solving very large size LP for real-world
applications by exploring:
- special structure.
- sparsity.
- decomposition.
- parallel computation.
• Nonlinear optimization with linear constraints.
• Conic Programming.
• Semidefinite Programming.
• Robust optimization

You might also like