A MIXED PARTICLE SWARM OPTIMIZATION ALGORITHM FOR CONTINUOUS FLOW-SHOP SCHEDULING PROBLEM
This paper presents a mixed particle swarm optimization MPSO algorithm for solving continuous flow-shop scheduling problem where no delay is allowed between the processing of consecutive tasks of one job. The proposed algorithm utilizes two learning mechanisms for the particles: moving toward the best or moving away from the worst. Most of the time, the particles attempt to update its velocity by moving toward the reference directions represented by the personal best and global best particle positions. The velocity index, a measure of dispersion of particles, is calculated after each particle move. A small velocity index indicates that the particles are clustering in a small area and may be trapped in local optima. Whenever the velocity index is smaller than a preset value, the particles are forced to disperse by applying the other learning mechanism, i.e., moving away from the directions represented by the personal worst and global worst particle positions. The proposed algorithm is tested using the benchmark problems and the experimental results show that MPSO is more efficient than GA in solving the continuous flow-shop scheduling problems both in terms of solution quality and speed.
Continuous Flow-shop Scheduling Mized Particle Swarm Optimization Dispersion Indez Metaheuristics Learning mechanisms.
Voratas Kachitvichyanukul Dao Duc Cuong
Industrial Engineering and Management, School of Engineering and Technology, Asian Institute of Tech Industrial Engineering and Management,School of Engineering and Technology, Asian Institute of Techn
国际会议
上海
英文
1-6
2009-08-02（万方平台首次上网日期，不代表论文的发表时间）