Abstract

PEREZ-ORTEGA, Joaquín et al. A New Heuristic Strategy for the Bin Packing Problem. Ing. invest. y tecnol. [online]. 2016, vol.17, n.2, pp.155-168.  Epub Apr 04, 2022. ISSN 2594-0732.

The Bin Packing problem (BPP) is NP-hard, the use of exact methods for solving BPP instances require a high number of variables and therefore a high computational cost. In this paper a new heuristic strategy for solving the BPP instances, which guarantees obtain optimal solutions, is proposed. The proposed strategy includes the use of a new model based on flow arcs. In the proposed model, the number of variables was reduced by previous allocation of objects in bins. Additionally, our approach includes a heuristic that allows reducing the instance size and thereby the solution algorithm search space. To validate the proposed approach, experiments were performed using the test sets hard28, 53nirup, bin1data and falkenauer, all of them well known in the state of the art. The results show that it using our approach is possible to find the optimal solution for all test set. Also, the execution time was reduced in regard the reported time obtained by using the flow arc model. Time reductions were up to 19.7 and 43% for 53nirup and hard28 test set, respectively.

Keywords : heuristic; bin packing; flow arcs; exact model for BPP; lower bound; instances of the BPP.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )