Skip to content
master
Go to file
Code

README.md

Introduction

At the very current status the project is composed by three chunks:

*   PILAL   (Pathetically Inadequate Linear Algebra Library)
*   Simplex (The real simplex algorithm implementation)
*   The solver

Simplex depends on PILAL, whilst the solver depends on Simplex. The project can be built either under Windows (there is a MS Visual Studio 2008 solution which can be automatically coverted to a 2010 solution without complications), Linux (make) or MacOS (make).

The project is currently working acceptably, however some optimizations and improvements could be added in the future:

*   Optimization to exploit upper and lower limitations in variables
*   Branch-and-bound for integer linear programming
*	Multi-threading (for branch and bound?)
*   PILAL and Simplex compiled as libraries under MacOS and Linux
*   Better language to define problems
*   Doxygen documentation

Basically, anything that can improve performance and usability is a good feature candidate.

How to use it

All you have to do to compile the project is:

make

If you want to clean up all the dependencies files and object files just type:

make wipe

The "lib" directory was used in a previous Linux version to compile PILAL and Simplex as libraries, but was discontinued because I wanted the Makefile to work under MacOS as well and I had not enough time to learn how to write portable Makefiles. "lib" is currently used by the Visual Studio solution to compile PILAL and Simplex as static libraries. Once you have built the project, to execute the solver with a problem file (look for some example problems in the "problems" directory) run under MacOS or Linux:

./solver <name_of_the_problem_file>

Under Windows:

solver.exe <name_of_the_problem_file>

Problems have a very straightforward structure, therefore you can learn the syntax from the examples in the "problems" or from the Google Code page.

License

See LICENSE file.

Drop a line

If you use this code for your software, please let me know with a mail message at tunnuz@gmail.com, or not.

About

Multi-platform, self-contained and object-oriented implementation of the Simplex Algorithm in C++

Resources

License

Releases

No releases published

Packages

No packages published
You can’t perform that action at this time.