Skip to content

Latest commit

 

History

History
This folder contains a simple, big algorithm for solving optimization problems (linear programs) with the simplex algorithm.

Implementation steps found at: http://hahamo.wordpress.com

Key features:
 * self-implemented
 * basic visualization (with future enhancements)
 * works only with a dictionary with introduced slack variables
 * feasibility detection
 * UNBOUNDED as returning value if the entering variable provides, that the problem dictionary is unbounded
 * using Bland's Rule to avoid cycles (eventually bad performance because cycles are rare in random problems)
 * initialization phase
 * converting LP solution to ILP solution if possible

Future improvements:
 * detecting need of slack variables
 * converting problem to standard form
 * improved visualization
 * greedy heuristics besides Bland's Rule
 * largest objective coefficient heuristics