Parallelizing Constraint Solvers for Hard RCPSP Instances - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Rapport (Rapport Technique) Année : 2016

Parallelizing Constraint Solvers for Hard RCPSP Instances

Résumé

The Resource-Constrained Project Scheduling Problem (RCPSP) is a well-known scheduling problem aimed at minimizing the makespan of a project subject to temporal and resource constraints. Constraint Programming allows to model and solve RCPSPs in a natural and efficient way, especially when Lazy Clause Generation (LCG) techniques are employed. In this paper we show that hard RCPSPs can be efficiently tackled by a portfolio approach that combines the strengths of different —not necessarily LCG-based— solvers by exploiting their parallel execution on multiple cores. Our approach seeks to predict and run the best solvers for a new, unseen RCPSP instance and also enables the bounds communication between them. This on-average allows to outperform the oracle solver that always chooses the best available solver for any given instance.
Fichier principal
Vignette du fichier
main.pdf (338.84 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01295061 , version 1 (30-03-2016)

Identifiants

  • HAL Id : hal-01295061 , version 1

Citer

Roberto Amadini, Maurizio Gabbrielli, Jacopo Mauro. Parallelizing Constraint Solvers for Hard RCPSP Instances. [Technical Report] Inria Sophia Antipolis. 2016. ⟨hal-01295061⟩
207 Consultations
492 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More