We describe a new branching strategy for binary integer programs in which fixing variables to 1 results in a significant change in the bound obtained by solving the LP relaxation, whereas fixing to 0 has little or no impact. In such cases, branching on variables can be ineffective. The new strategy involves simultaneously branching on a set of variables chosen by solving a bilevel program. Computational results show that this strategy can be effective for combinatorial problems.
Bilevel Branching
ROSSI, FABRIZIO;SMRIGLIO, STEFANO
2009-01-01
Abstract
We describe a new branching strategy for binary integer programs in which fixing variables to 1 results in a significant change in the bound obtained by solving the LP relaxation, whereas fixing to 0 has little or no impact. In such cases, branching on variables can be ineffective. The new strategy involves simultaneously branching on a set of variables chosen by solving a bilevel program. Computational results show that this strategy can be effective for combinatorial problems.File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
