Linear programming via a quadratic penalty function
Pinar Mustafa, C.
ZOR. Zeitschrift fur Operations-Research
Physica-Verlag, Heidelberg, Germany
Please cite this item using this persistent URLhttp://hdl.handle.net/11693/25748
We use quadratic penalty functions along with some recent ideas from linear l1 estimation to arrive at a new characterization of primal optimal solutions in linear programs. The algorithmic implications of this analysis are studied, and a new, finite penalty algorithm for linear programming is designed. Preliminary computational results are presented.
- Research Paper