The multibin-packing constraint captures a fundamental substructure of many assignment problems, where a set of items, each with a fixed number of dimensions, must be assigned to a number of bins with limited capacities. In this work we propose a simple decomposition for multibin-packing that uses a bin-packing constraint for each dimension, a set of all-different constraints automatically derived from a conflict graph, plus two alternative symmetry breaking approaches. Despite its simplicity, the proposed decomposition is very effective on a number of instances recently proposed in the literature. © 2013 Springer-Verlag.
CITATION STYLE
Gualandi, S., & Lombardi, M. (2013). A simple and effective decomposition for the multidimensional binpacking constraint. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8124 LNCS, pp. 356–364). https://doi.org/10.1007/978-3-642-40627-0_29
Mendeley helps you to discover research relevant for your work.