An Exact Primal-Dual Penalty Method Approach to Warmstarting Interior-Point Methods for Linear Programming

Hande Benson (bensondrexel.edu)
David Shanno (shannorutcor.rutgers.edu)

Abstract: One perceived deficiency of interior-point methods in comparison to active set methods is their inability to efficiently re-optimize by solving closely related problems after a warmstart. In this paper, we investigate the use of a primal-dual penalty approach to overcome this problem. We prove exactness and convergence and show encouraging numerical results on a set of linear and mixed integer programming problems.

Keywords: Interior-point methods, linear programming, warmstarting, penalty methods

Category 1: Linear, Cone and Semidefinite Programming (Linear Programming )

Citation: Working paper, September 2005

Download: [PDF]

Entry Submitted: 09/01/2005
Entry Accepted: 09/02/2005
Entry Last Modified: 09/01/2005

Modify/Update this entry