改进人工蜂群算法求解置换流水车间调度问题

针对置换流水车间调度问题（PFSP），以最小化最大完工时间为优化目标建立数学模型，设计了一种改进人工蜂群算法。该算法采用反向学习方法和混沌映射来生成初始种群，为使算法能够求解离散的调度问题，采用LRV规则将位置数值映射成工件排列顺序；在雇佣蜂阶段，融入差分进化算法的思想，加入高斯变异算子，使收敛速度加快；在跟随蜂阶段，加入自适应策略，将算法的勘探和开发能力进行平衡；在侦察蜂阶段，加入柯西变异算子，避免陷入局部极值。最后通过比较几种不同的算法，对Car算例以及部分Rec标准算例集进行仿真测试，验证该算法的有效性和优越性。

IMPROVED ARTIFICIAL BEE COLONYLGORITHM FOR SOLVING PERMUTATION FLOW SHOP SCHEDULING PROBLEM

For the permutation flow shop scheduling problem (PFSP), mathematical model is established to minimize the maximum completion time, and an improved artificial bee colony algorithm is designed. The initial population is generated based on the reverse learning method and chaotic mapping. In order to enable the algorithm to solve the discrete scheduling problem, LRV rules are used to map the position values into the job order. In the employed bee phase, the idea of differential evolution algorithm is integrated, and the Gaussian mutation operator is added to accelerate the convergence speed. In the onlooker bee phase, the adaptive strategy is added to balance the exploration and development capabilities of the algorithm. In the scout bee phase, cauchy mutation operator is added to avoid falling into local extremum. Finally, by using several different algorithms, the car example and part of the Rec standard example set are simulated and the results are compared to verify the effectiveness and superiority of the algorithm.