MCS-libre Project Homepage

Last Updated: 19 October 2007

Introduction

MCS-libre is a free, open-source, C++ toolkit for performing Monte-Carlo Simulation. It is designed to be "lightweight" for easy and efficient integration into larger simulation programs.

The toolkit is released under GNU's Lesser General Public License. This allows inclusion of the work into proprietary programs, while promoting software liberty as defined by the Free Software Foundation. The code is open-source to leverage the benefits of peer-review in a manner that enhances the collective wealth of society.



News

Version 2.0 is coming out very soon, probably before the end of October 2007. Graphical results from quality testing of the release can be viewed here.

This a significant restructuring of the toolkit, primarily to take advantage of the free Boost C++ Random libraries. Boost has peer-reviewed versions of several of the functions implemented from scratch in MCS-libre 1.1, which MCS-libre 2.0 will use. This will make MCS-libre much smaller, while providing more functionality to the end user. The core of MCS-libre, i.e., the simulation stream manager and data collection facilities, will still be implemented from scratch.

No effort has been made to ensure backward compatibility with version 1.1, since this is a substantial revision. If you need backwards compatibility, contact the author though the download page.



Download

Download MCS-libre here.



Documentation

Documentation for version 1.1 is available in HTML, Postscript, and PDF. Version 2.0, which is coming very soon, has a completely different structure so the documentation offered here will soon be obsolete. New documentation will be posted with the release of version 2.0.



Alternatives to MCS-libre

MCS-libre was designed for speed and flexibility, but you have to “muck around” in C++ to use it. If you are just doing a “quick and dirty” analysis, you may be better served by the free R Project for Statistical Computing environment (an implementation of the S language) or GNU Octave (a free implementation of the Matlab language). Both are excellent resources that are very powerful. R in particular is great for graphical display; the test plots shown here to validate the output of MCS-libre version 2.0 where produced using it.



Author

MCS-libre was written by D. E. Williams to assist discrete-event simulation of hydroponics, dynamic systems, and manufacturing systems. The author holds a B.S. degree in Mechanical Engineering, has completed 25 graduate credits in Controls Engineering, and is looking for employment in the area surrounding Austin, Texas, USA.



Acknowledgements


  SourceForge Logo