SAT and Hybrid models of the Car-Sequencing problem
Résumé
We improve the state of the art for solving car-sequencing problems by combining together the strengths of SAT and CP. We compare both pure SAT and hybrid CP/SAT models. Three features of these models are crucial to success. For quickly finding solutions, advanced CP heuristics are important and good propagation (either by a specialized propagator or by a sophisticated SAT encoding that simulates one) is necessary. For proving infeasibility, clause learning in the SAT solver is critical. Our models contain a number of other novelties. In our SAT models, we introduce powerful new encodings for the AtMostSeqCard constraint on which unit propagation simulates an AC propagator. In our hybrid models, we develop a novel linear time mechanism for explaining failure in the same constraint. Our study demonstrates some of the potential and complementarity of SAT and hybrid methods for solving complex constraint models.
Origine : Fichiers produits par l'(les) auteur(s)
Loading...