改进狼群算法求解柔性作业车间调度问题

Improved wolf pack algorithm for solving flexible job⁃shop scheduling problem

针对传统群智能优化算法求解柔性作业车间调度问题时存在寻优后期收敛速度慢、易陷入局部最优等问题,提出一种改进狼群算法(IWPA)。首先,构建以最小化最大完工时间为优化目标的柔性作业车间调度问题(FJSP)模型;其次,采用混沌初始化操作和二进制串调整策略以提高初始种群的质量;为提高算法的收敛速度和全局搜索能力,对探狼和猛狼的位置更新公式分别进行改进;采用类似于遗传算法(GA)中的选择、交叉操作不断对最佳加工序列进行扰动,以改善算法的局部搜索性能;最后,采用新的种群淘汰机制更新狼群来丰富种群多样性。通过与多种群智能优化算法的仿真实验对比,证明所提出的改进狼群算法对求解FJSP问题可行、合理且高效。

An improved wolf pack algorithm(IWPA)is proposed to solve the flexible job⁃shop scheduling problem(FJSP)because the traditional swarm intelligence optimization algorithms have the deficiencies of slow convergence and are susceptible to falling into local optimum when they are used for FJSP.The flexible⁃job shop scheduling model which takes minimization of the maximum completion time as optimization purpose is built.Chaos initialization and binary string adjustment are used to improve the quality of the initial population.The location update formulas of the scout wolf and the fierce wolf are improved to enhance the convergence speed and global search ability of the algorithm.In addition,the selection operation and crossover operation similar to those in the genetic algorithm(GA)are adopted to continuously disturb the optimal processing sequence,so as to improve the local search performance of the algorithm.The new population elimination mechanism is adopted to update the wolf packs to enrich the population diversity.In comparison with several swarm intelligence optimization algorithms,the proposed IWPA based on the disturbance mechanism is feasible,reasonable and efficient for solving the FJSP.