In this talk a greedy randomized adaptive search procedure (GRASP) is presented for computing approximate solutions to the NP-hard three-index assignment problem (AP3). A F ORTRAN implemen-tation of this GRASP is tested on several problem instances of the AP3. A parallelization strategy that combines GRASP and path re-linking is also proposed and tested for the problem. Computational results indicate that the GRASP and the parallel hybrid heuristic provide good approximate solutions to a variety of AP3 instances.
CITATION STYLE
Aiex, R. M., Pardalos, P. M., Pitsoulis, L. S., & Resende, M. G. C. (2000). A GRASP for computing approximate solutions for the three-index assignment problem: Invited talk. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1800, p. 504). Springer Verlag. https://doi.org/10.1007/3-540-45591-4_67
Mendeley helps you to discover research relevant for your work.