基亏基因片段分解的粒子群算法求解置换Flowshop问题

针对粒子群算法在求解置换流水车间调度问题时容易早熟的现象，提出了一种基于基因片段分解的粒子群优化算法求解置换流水车间调度问题。首先。对工件加工顺序采用了基因片段分解的方法，个体的初始值是随机生成的。但是初始种群采用贪婪方法得到。然后，通过加入综合学习策略和增强基因片段间的合作来提高该算法的全局搜索能力。对基因片段最优解进行交换局部搜索。最后，通过对Rec系列20个子问题的仿真测试，得出该算法在每个子问题上都取得了优于粒子群算法的解。仿真结果表明该算法收敛速度快，且具有较高的求解质量。

Decomposition of gene section particle swarm optimization for the Flowshop problem

Coping with such disadvantages of particle swarm optimization algorithm being easy to run into local optima,the method that particle swarm optimization based on gene scetion decomposition is proposed to be applied to permutation flow shop scheduling algorithm. Firstly, using the method of gene scetion decomposition in the job processed sequence and generated the individual initial value randomly, but obtained the initial population by greed method. Then, improving the global search capability of the algorithm by adding integrated learning strategies and strengthening cooperation among sub- groups, after that , exchanged the best values of gene scetion of local search. Finally, the simulation test for Rec series of 20 sub-problems proves that result of each sub-problems by this algorithm is superior to particle swarm optimization algorithm. It shows that this algorithm has better answers and more rapid convergence.