Abstract
This work introduces a set of important improvements in the resolution of the Two Dimensional Cutting Stock Problem. It presents a new heuristic enhancing existing ones, an original upper bound that lowers the upper bounds in the literature, and a parallel algorithm for distributed memory machines that achieves linear speedup. Many components of the algorithm are generic and can be ported to parallel branch and bound and A* skeletons. Among the new components there is a comprehensive MPI-compatible synchronization service which facilitates the writing of time-based balancing schemes. © Springer-Verlag Berlin Heidelberg 2007.
Cite
CITATION STYLE
León, C., Miranda, G., Rodríguez, C., & Segura, C. (2007). 2D cutting stock problem: A new parallel algorithm and bounds. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4641 LNCS, pp. 795–804). Springer Verlag. https://doi.org/10.1007/978-3-540-74466-5_85
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
