|
|
|
|
|
|
|
|
Discrete Optimization Using Mathematica
|
|
|
|
|
|
Organization: | Wolfram Research, Inc. |
|
|
|
|
|
|
2002 World Multiconference on Systemics, Cybernetics, and Informatics (SCI 2002)
|
|
|
|
|
|
Orlando, FL
|
|
|
|
|
|
Many important classes of optimization problems are discrete in nature. Examples are the standard problems of integer programming (including the ever-important "knapsack problems"), permutation assignment problems (e.g., the notorious "traveling salesman problem"), set coverings, set partitioning, and so on. Problems of practical interest are frequently too large to be solved by brute-force enumeration techniques. Hence arises a need for more refined methods that use various tactics for sampling and searching the domain in pursuit of optimal solutions. Version 4.2 of Mathematica has a flexible package for performing global optimization. It includes powerful functionality from the category of evolutionary programming. Ways to apply this technology to various problems in discrete optimization will be discussed. We will present details of how to code problems so that built-in Mathematica functions can digest them and will illustrate with a variety of examples. We will also discuss some practical tuning considerations.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| lichtblau.nb (71 KB) - Mathematica Notebook |
|
|