Solving knapsack problem based on discrete particle swarm optimization
GU Qian-qian · Jisuanji gongcheng yu sheji · 2007
Particle swarm optimization(PSO) is a new evolutionary algorithm.Numerical optimization is the primary field of PSO applications.Combining discrete particle swarm optimization(DPSO) with penalty function method and greedy transform method,two new algorithms is proposed for solving Knapsack problem(KP): Discrete particle swarm optimization based on penalty function strategy(PFDPSO) and greedy discrete particle swarm optimization(GDPSO).PFDPSO and GDPSO are compared with hybrid particle swarm optimization(Hybrid_PSO) in Ref.[7].The numerical results show that GDPSO is most excellent for solving knapsack problem.The ability in solving KP of GDPSO exceeds Hybrid_PSO.But PFDPSO is more inefficient.These indicate that it is more efficient and practical method that DPSO combines with greedy transform method for solving knapsack problem.