Scheduling electronic products based on a modified particle swarm optimization

Yong Zhu,Xiaohong Pan

2011-01-01

Abstract:Order-oriented production was satisfied to maximizing the profit, and to the due date of customer order. In such a circumstances, scheduling orders had been an important decision in modern enterprises. A method of scheduling electronic products based on a modified PSO was proposed. A cost model of scheduling orders for electronic products with fuzzy capacity constraints was established. The total cost was constituted of two costs, they were production costs from the assembly and penalty cost arising from tardiness production. A new adaptive particle swarm optimization algorithm with dynamically changing inertia weight (DCWPSO) was proposed to solve the problem. The DCWPSO changed inertia weight based on population diversity and evolution speed in order to balance the trade-off between exploration and exploitation. The application demonstrates the efficiency of the proposed model and DCWPSO algorithm, which is significantly superior to the linearly decreasing weight PSO (LDWPSO) in convergence speed and accuracy.