Spring 2002 Course Announcement!

WHICH COURSE? Math 5050 - Linear Programming

The idea behind linear programming is simple: you have a quantity you want to maximize or minimize, that quantity depends linearly on some variables, and each of the variables must satisfy some linear constraints. The main tool for solving problems of this type is a family of algorithms that have been developed since the 1950's and that are known collectively as the Simplex Method. Linear programming problems are so ubiquitous in ``real life'' (where ``real life'' is defined as business, economics, and engineering), and the Simplex Method is so elementary, that solving linear programming problems using the Simplex Method is frequently a topic in pre-Calculus courses.

This course will follow the book pretty closely. I hope to cover most of Part 1 - Basic Theory, some of Part 2 - Network Problems, and some of Part 4 - Integer Programming and Quadratic Programming. For more information about the book, see http://www.princeton.edu/~rvdb/LPbook. In general, the level of mathematics, won't be much higher than in an undergraduate linear algebra class. Rather, the emphasis will be on understanding why the algorithms work, something about the real-world problems they apply to, and how they're efficiently implemented on a computer.

In order to understand some of the issues involved in the computer implementation of the algorithms, it will be an overall goal of the class as a whole (including the instructor) to write the code implementing the algorithms as efficiently as we can. Then we can compare our programs with programs that have been written by others and see which is more efficient (and why!).

See you in Linear Programming!

Last modified: Sun Dec 30 17:27:34 CST 2001