Several Mixed Integer Linear Programming problems present a formulation based on a great number of knapsack constraints. Problems widely addressed in literature with this structure are, among others, Generalized Assignment, Multiple Knapsack, Bin Packing, Capacitated P-median and Single Source Capacitated Facility Location. In general knapsack constraints make these problems very hard to solve. The state of the art on these problems requires to use approaches besed on Lagrangean Relaxation or decomposition approaches like Dantzig-Wolfe and Column Generation tenchniques. In this paper, we present an approach based on the generation of general cutting planes of the polyhedron associated with each knapsack constraints. This approach yelds a lower bound for the LP-relaxation that is the same obtained by the Dantzig-Wolfe decomposition and henceforth stronger than the LP-relaxation. We use this aproach in the solution of the Single Source Capacitated Facility Location problem (SSCFLP) using a Branch-and-Cut algorithm. Computational experience is reported on a large set of test instances available in literature.
A Branch-and-Cut algorithm for the Single Source Capacitated Facility Location problem
2013-05-01
1080240 byte
Conference paper
Electronic Resource
English
Branch-and-Cut-and-Price for Capacitated Connected Facility Location
British Library Online Contents | 2011
|Reliable capacitated facility location problem with service levels
Online Contents | 2018
|