DMCA
Implementation Of A Variance Reduction Based Lower Bound In A Branch And Bound Algorithm For The Quadratic Assignment Problem (1997)
| Citations: | 5 - 1 self |
BibTeX
@MISC{Pardalos97implementationof,
author = {P.M. Pardalos and K. G. Ramakrishnan and M.G.C. Resende and Y. Li},
title = {Implementation Of A Variance Reduction Based Lower Bound In A Branch And Bound Algorithm For The Quadratic Assignment Problem},
year = {1997}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
. The efficient implementation of a branch and bound algorithm for the quadratic assignment problem (QAP), incorporating the lower bound, based on variance reduction, of Li, Pardalos, Ramakrishnan, and Resende (1994), is presented. A new data structure for efficient implementation of branch and bound algorithms for the QAP is introduced. Computational experiments with the branch and bound algorithm on different classes of QAP test problems are reported. The branch and bound algorithm using the new lower bounds is compared with the same algorithm utilizing the commonly applied Gilmore-Lawler lower bound. Both implementations use a greedy randomized adaptive search procedure for obtaining initial upper bounds. The algorithms report all optimal permutations. Optimal solutions for previously unsolved instances from the literature, of dimensions n = 16 and n = 20, have been found with the new algorithm. In addition, the new algorithm has been tested on a class of large data variance problem...
Keyphrases
bound algorithm quadratic assignment problem variance reduction efficient implementation new algorithm algorithm report optimal solution different class adaptive search procedure optimal permutation new data structure computational experiment unsolved instance initial upper bound qap test problem large data variance problem
