Chapter 1
Introduction
Optimization problems are abundant in every day life, and all of us face such problems fre
quently, although we may not always be aware of the fact! Obvious examples are, for instance,
the use of your GPS to Fnd a shortest route from your home to your work place in the morning,
or the scheduling of trains on the rail connections between Waterloo and Toronto. There are
however many more, less obvious examples. How, for example, does the region of Waterloo
determine the structure of its electricity network? How are schedules for buses determined?
And how does a company set the price for a newly developed product? All of these questions
are optimization problems, and in this chapter we show that many of these questions admit a
mathematical formulation.
1.1
Linear Programming
In this section we introduce linear programming, an efFcient and versatile method to optimize
(minimize or maximize) a linear function subject to a system of linear inequality and/or equality
constraints. As unimpressive as this may sound, linear programming is a very powerful tool that
is used in practice to solve instances of optimization problems arising in applications in most
branches of industry. In fact, a recent survey (see also [3]) of ±ortune 500 Frms shows that 85%
of all respondents use linear programming in their operations. The roots of linear programming
7
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document8
CHAPTER 1. INTRODUCTION
can be traced back at least a couple of hundred years to the work of Fourier on solutions of sys
tems of linear inequalities. The name
linear programming
, however, was coined only recently,
in the late 1930’s and early 1940’s when the Russian mathematician Leonid Kantorovich and
the American George Dantzig formally de±ned the technique. George Dantzig also developed
the
Simplex algorithm
which to this date remains one of the most popular methods to solve lin
ear programs. Dantzig who worked as a mathematical advisor for the U.S. Air Force initially
applied linear programming to solve logistical problems arising in the military. It did, however,
not take long for industry to realize the technique’s potential, and its use is widespread today.
In this section we will see two typical examples of optimization problems that can be solved via
linear programming.
1.1.1
A production example
Production problems are probably among the most frequent applications of linear programming.
In a typical such application, we are given a company that produces a number of different prod
ucts from a set of resources. Producing a unit of a product requires the use of a certain amount
of each resource, and can be sold at a certain price on the market. The company has limited
amounts of each of the resources available, and is interested in maximizing its revenue from
sales. How much of each product should be produced?
This is the end of the preview. Sign up
to
access the rest of the document.
 Spring '10
 GUENIN
 Linear Programming, Optimization, objective function, KWOil

Click to edit the document details