SciELO - Scientific Electronic Library Online

 
vol.28 issue6Quality in Education and Transport Infrastructure from the Territorial Accessibility Pont of View. Caldas Department in Colombia as a Study CaseJPIAspectZ: An Extension of AspectZ for Formal Requirement Specification of JPI Aspect-Oriented Applications author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

  • On index processCited by Google
  • Have no similar articlesSimilars in SciELO
  • On index processSimilars in Google

Share


Información tecnológica

On-line version ISSN 0718-0764

Abstract

BUITRAGO, Oscar Y; RAMIREZ, Andrés L  and  BRITTO, Rodrigo A. Parametric Projections to Escape from Edges in Polyhedrons of the Type Ax ≤ b. Inf. tecnol. [online]. 2017, vol.28, n.6, pp.181-188. ISSN 0718-0764.  http://dx.doi.org/10.4067/S0718-07642017000600019.

The aim of this paper is to propose a procedure to create escape vectors specifically developed to perform an orthogonal projection from an Ax ≤ b type polyhedron’s edge tn its interinr. The prnpnsed prncedure can be implemented in interior-point methods for solving linear programming problems since in these methods various strategies are used to avoid reaching the borders of the polyhedron. A projection procedure was develnped tn the pnlyhedrnn’s interinr starting frnm three scenarins and it was used nn a cnntext case demonstrating its applicability. The results show that the projection carried out in the iteration in which the escape is done not only allows abandoning from a suboptimal solution but it also kept the same objective function value. Therefore, it can be concluded that interior-point methods do not need to avoid the frontier or to adjust values, as it occurs in other applications.

Keywords : escape vector; polyhedron; optimization; orthogonal projections; interior-point method.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )

 

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License