DocumentCode
1711020
Title
Facility location with nonuniform hard capacities
Author
Pal, Martin ; Tardos, Eva ; Wexler, Tom
Author_Institution
Cornell Univ., Ithaca, NY, USA
fYear
2001
Firstpage
329
Lastpage
338
Abstract
The authors give the first constant factor approximation algorithm for the facility location problem with nonuniform, hard capacities. Facility location problems have received a great deal of attention in recent years. Approximation algorithms have been developed for many variants. Most of these algorithms are based on linear programming, but the LP techniques developed thus far have been unsuccessful in dealing with hard capacities. A local-search based approximation algorithm (M. Korupolu et al., 1998; F.A. Chudak and D.P. Williamson, 1999) is known for the special case of hard but uniform capacities. We present a local-search heuristic that yields an approximation guarantee of 9 + ε for the case of nonuniform hard capacities. To obtain this result, we introduce new operations that are natural in this context. Our proof is based on network flow techniques.
Keywords
approximation theory; computational complexity; facility location; heuristic programming; search problems; LP techniques; approximation guarantee; constant factor approximation algorithm; facility location; linear programming; local-search based approximation algorithm; local-search heuristic; network flow techniques; nonuniform hard capacities; proof; special case; uniform capacities; Approximation algorithms; Capacity planning; Cost function; Linear programming; Writing;
fLanguage
English
Publisher
ieee
Conference_Titel
Foundations of Computer Science, 2001. Proceedings. 42nd IEEE Symposium on
Print_ISBN
0-7695-1116-3
Type
conf
DOI
10.1109/SFCS.2001.959907
Filename
959907
Link To Document