Data di Pubblicazione:
2007
Abstract:
Interior Point (IP) algorithms for Min Cost Flow (MCF) problems have been
shown to be competitive with combinatorial approaches, at least on some
problem classes and for very large instances. This is in part due to
availability of specialized crossover routines for MCF; these allow early
termination of the IP approach, sparing it with the final iterations
where the KKT systems become more difficult to solve. As the crossover
procedures are nothing but combinatorial approaches to MCF that are only
allowed to perform few iterations, the IP algorithm can be seen as a
complex ``multiple crash start'' routine for the combinatorial ones. We
report our experiments about allowing one primal-dual combinatorial
algorithm to MCF to perform as many iterations as required to solve the
problem after being warm-started by an IP approach. Our results show that
the efficiency of the combined approach critically depends on the accurate
selection of a set of parameters among very many possible ones, for which
designing accurate guidelines appears not to be an easy task; however, they
also show that the combined approach can be competitive with the original
combinatorial algorithm, at least on some ``difficult'' instances.
Tipologia CRIS:
01.01 Articolo in rivista
Keywords:
Min Cost Flow problems; Interior Point algorithms; Relaxation Method; Crossover
Elenco autori:
Frangioni, Antonio; Gentile, Claudio
Link alla scheda completa:
Pubblicato in: