基于知识进化粒子群算法的模糊交货期流水车间调度问题

Knowledge evolution particle swarm optimization for solving flow shop scheduling problem with fuzzy due date

针对模糊交货期的流水车间调度问题的特点,提出采用知识进化算法和粒子群优化的混合算法来求解问题。该算法首先在多个群体空间内采用粒子群优化寻找局部最优解,然后利用知识进化算法的猜测操作和反驳操作建立以群体空间知识为基础的一个知识空间,最后通过知识空间的协同进化更新其中的社会知识,从而形成问题的最优解。通过采用所提算法对带模糊交货期的流水车间调度问题的实例进行测试,并比对遗传算法和粒子群优化算法,表明了混合算法的可行性和有效性。

Aiming at the characteristics of flow shop scheduling problem with fuzzy delivery time,the hybrid algorithm based on knowledge evolution algorithm and particle swarm optimization was proposed to solve the problem.Particle swarm optimization was used by the algorithm to find a local solution in multi-group space,and the surmise operation and the forecast operation as well as rebut operation of knowledge evolution algorithm were used to built a knowledge space based on group space knowledge.The social knowledge in knowledge space was updated by coevolution,thus the optimum solution of the problem was formed.The example of flow shop scheduling problem with fuzzy delivery time was tested by proposed algorithm,and the result showed the feasibility and effectiveness of the hybrid algorithm.