A binary particle swarm optimization for solving the bounded knapsack problem

Abstract

Bounded knapsack problem (BKP) is a classical knapsack problem. At present, methods for solving the BKP are mainly deterministic algorithms. The literature that using evolutionary algorithms solve this problem has not been reported. Therefore, this paper uses a binary particle swarm optimization (BPSO) to solve the BKP. On the basis of using the repair and optimization method to deal with the infeasible solutions, an effective method of using BPSO to solve the BKP is given. For three kinds of large-scale BKP instances, the feasibility and efficiency of BPSO are verified by comparing the results with whale optimization algorithm and genetic algorithm. The experimental results show that BPSO is not only more stable, but also can obtain the approximation ratio closer to 1.