Abstract: For the multi-objective discrete workshop energy-saving scheduling issue,a multi-neighborhood local improvement search method was suggested to optimize the maximum completion time and the total energy consumption of processing.In accordance with the peculiarities of the issue,OS self-assignment rules were implemented to lower the dimensionality of the solution space,and a machine greedy allocation mechanism was developed to balance the machine load.To optimize the initial solution's quality,a two-rule coordinated population initialization strategy based on the Pareto dominance relation was given.An adaptive global enhanced search operator and a local enhanced search operator were suggested,inspired by the feedback control mechanism of biological hormones.To enhance the algorithm's global search capability and local search depth,to grow the memory pool matrix and to prevent the algorithm from prematurely converging,the two operators were alternately executed.The superiority and stability of the proposed approach for addressing the multi-objective discrete workshop energy-saving scheduling issue were validated by comparative studies.

Key words: discrete workshop, energy-saving scheduling, machine greedy allocation mechanism, global enhanced search operator, local enhanced search operator

摘要： 针对多目标离散车间节能调度问题,以优化最大完工时间和加工总能耗为目标,提出一种多邻域局部增强搜索算法(MLEA)。根据问题特点,引入工序向量(OS)自指定规则对解空间降维,设计一种机器贪婪分配机制,以均衡机器负载,为优化初始解的质量,提出一种基于Pareto支配关系的双规则协调种群初始化方法。受生物激素反馈调节机制启发,提出自适应全局强化搜索算子和局部增强搜索算子,两种算子交替运行,以提高算法的全局搜索能力和局部搜索深度,增加记忆池矩阵,避免算法过早收敛。最后,通过对比实验,验证了所提算法在解决多目标离散车间节能调度问题上的优越性和稳定性。

关键词: 离散车间, 节能调度, 机器贪婪分配机制, 全局强化搜索算子, 局部增强搜索算子