Please use this identifier to cite or link to this item: http://repositorio.unicamp.br/jspui/handle/REPOSIP/328254
Type: Artigo
Title: A Variation On The Interior Point Method For Linear Programming Using The Continued Iteration
A variation on the interior point method for linear programming using the continued iteration
Author: Berti, Lilian F.
Oliveira, Aurelio R. L.
Ghidini, Carla T. L. S.
Abstract: In this paper, we present a proposal for a variation of the predictor-corrector interior point method with multiple centrality corrections. The new method uses the continued iteration to compute a new search direction for the predictor corrector method. The purpose of incorporating the continued iteration is to reduce the overall computational cost required to solve a linear programming problem. The computational results constitute evidence of the improvement obtained with the use of this technique combined with the interior point method.
In this paper, we present a proposal for a variation of the predictor-corrector interior point method with multiple centrality corrections. The new method uses the continued iteration to compute a new search direction for the predictor corrector method. The purpose of incorporating the continued iteration is to reduce the overall computational cost required to solve a linear programming problem. The computational results constitute evidence of the improvement obtained with the use of this technique combined with the interior point method
Subject: Linear Programming
Interior Point Methods
Continued Iteration
Continuous Optimization
Programação linear
Métodos de pontos interiores
Métodos iterativos (Matemática)
Country: Alemanha
Editor: Springer
Citation: Mathematical Methods Of Operations Research. Springer Heidelberg, v. 85, p. 61 - 75, 2017.
Rights: fechado
Identifier DOI: 10.1007/s00186-016-0558-9
Address: https://link.springer.com/article/10.1007/s00186-016-0558-9
Date Issue: 2017
Appears in Collections:IMECC - Artigos e Outros Documentos
FCA - Artigos e Outros Documentos

Files in This Item:
File SizeFormat 
000397237500005.pdf549.37 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.