Title
2D cutting stock problem: a new parallel algorithm and bounds
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.
Year
DOI
Venue
2007
10.1007/978-3-540-74466-5_85
Euro-Par
Keywords
Field
DocType
parallel algorithm,linear speedup,new component,new heuristic,stock problem,comprehensive mpi-compatible synchronization service,time-based balancing scheme,new parallel algorithm,memory machine,important improvement,upper bound,cutting stock problem,branch and bound
Heuristic,Branch and bound,Synchronization,Upper and lower bounds,Computer science,Parallel algorithm,Parallel computing,Distributed memory,Cutting stock problem,Speedup,Distributed computing
Conference
Volume
ISSN
ISBN
4641
0302-9743
3-540-74465-7
Citations 
PageRank 
References 
5
0.50
8
Authors
4
Name
Order
Citations
PageRank
Coromoto León123125.71
Gara Miranda218818.16
Casiano Rodríguez313023.36
Carlos Segura421621.44