Abstract: China's manufacturing model is gradually evolving to a distributed collaborative production model.Aiming at the Distributed Hybrid Flowshop Scheduling Problem (DHFSP) with the goal of minimizing completion time and total energy consumption,combining the advantages of genetic operators and Iterative Greed algorithm (IG),a multi-objective hybrid iterative greed algorithm based on non-dominant ordering was proposed.In this algorithm,based on the NEH2 rule,a collaborative initialization strategy was proposed to improve the quality of the initial solution;a multi-factory-based crossover operator was designed to increase the diversity of the population,which was helpful to explore more areas of the problem-solving space.According to the characteristics of the problem,a multi-objective local search method was proposed,which enhanced the local search ability of the algorithm and avoided the premature convergence of the algorithm.To verify the effectiveness of the algorithm,MOHIG was compared with three multi-objective optimization algorithms of NSGA-Ⅱ,MOEA/D and JAYA through 360 examples,and the results showed that the two performance indexes of MOHIG algorithm were better than those of the other three algorithms,and MOHIG algorithm was efficient in solving DHFSP.

Key words: distributed hybrid flowshop scheduling, multi-objective optimization, iterative greedy algorithms, energy consumption

摘要： 目前我国制造模式正逐步向分布式协同生产模式演进。针对以最小化完工时间和总能耗为目标的分布式混合流水车间调度问题(DHFSP),综合遗传算子和迭代贪婪算法(IG)的优点,提出了一种基于非支配排序的多目标混合迭代贪婪算法(MOHIG)。在该算法中,基于NEH2规则提出了一种协同初始化策略提高初始解的质量;设计一种基于多工厂的交叉算子增加种群的多样性,有助于探索问题解空间的更多区域;根据问题多工厂调度的特点提出一种多目标局部搜索方法,增强了算法的局部搜索能力,避免算法过早收敛。为了验证算法的有效性,将MOHIG与NSGA-Ⅱ、MOEA/D和JAYA三种多目标优化算法通过360个实例进行了比较,结果显示MOHIG算法的两个性能指标都优于其他三种算法,表明MOHIG算法在求解DHFSP方面具有高效性。

关键词: 分布式混合流水车间调度, 多目标优化, 迭代贪婪算法, 能耗