A chaotic neural network combined heuristic strategy for multidimensional knapsack problem

Abstract

Multidimensional Knapsack Problem (MKP), as a classic combinatorial optimization problem, is used widely in various fields such as capital budgeting, allocating processors and databases in a distributed computer system. In this paper, a chaotic neural network combined heuristic strategy (TCNN-HS) is proposed for MKP. The proposed algorithm combines heuristic strategy which includes repair operator and improvement operator so that not only the infeasible solution can be overcome, but also the quality of the solutions can be improved. The TCNN-HS is tested on some benchmark problems, which is selected from OR library. Simulation results show that the proposed approach can find optimal solutions for some instances and outperforms TCNN. © 2008 Springer Berlin Heidelberg.