Improving an exact approach for solving separable integer quadratic knapsack problems

Authors:
Federico Della Croce;Dominique Quadri
Affiliations:
DAI, Politecnico di Torino, Torino, Italy;LIA, Université d'Avignon, Avignon, France 84911
Venue:
Journal of Combinatorial Optimization
Year:
2012

Citing 2
Cited 0

A Genetic Algorithm for the Multidimensional Knapsack Problem

Journal of Heuristics
Computing exact solution to nonlinear integer programming: Convergent Lagrangian and objective level cut method

Journal of Global Optimization

Quantified Score

Hi-index	0.00

Visualization

Abstract

We consider the specially structured (pure) integer Quadratic Multi-Knapsack Problem (QMKP) tackled in the paper "Exact solution methods to solve large scale integer quadratic knapsack problems" by D. Quadri, E. Soutif and P. Tolla (2009), recently appeared on this journal, where the problem is solved by transforming it into an equivalent 0---1 linearized Multi-Knapsack Problem (MKP). We show that, by taking advantage of the structure of the transformed (MKP), it is possible to derive an effective variable fixing procedure leading to an improved branch-and-bound approach. This procedure reduces dramatically the resulting linear problem size inducing an impressive improvement in the performances of the related branch and bound approach when compared to the results of the approach proposed by D. Quadri, E. Soutif and P. Tolla.