基于多目标混合进化算法的作业车间混排可变分批节能调度方法

Energy-efficient Job Shop Scheduling with Variable Lot Splitting and Sublots Intermingling Based on Multi-objective Hybrid Evolutionary Algorithm

针对作业车间分批调度问题,集成可变子批划分和子批混排策略,考虑批量划分约束、子批混排加工约束等,建立了最小化能耗和完工时间的混排可变分批调度优化模型,并提出了一种改进多目标混合进化算法。为了协调算法的全局搜索与局部搜索性能,将Jaya算法种群更新机制引入基于分解的多目标进化算法中,同时结合混排可变分批调度问题特征,设计了一种基于子批拆分/合并与关键链相结合的局部搜索策略。基于不同规模算例,对比分析了所提出的算法与其他经典算法的求解性能。实验结果表明,所提出的算法在Pareto解集收敛性和分布性方面具有明显优势,同时所提出的混排可变分批策略可有效降低能耗、缩短完工时间。

For solving the lot streaming job shop scheduling,a strategy was presented integrating variable sublots splitting and sublots intermingling,and a multi-objective optimization model of lot streaming scheduling was established to minimize the energy consumption and makespan.An improved multi-objective hybrid evolutionary algorithm was presented.In order to balance the global and local searching ability of the algorithm,the population updating mechanism of the Jaya algorithm was incorporated into the decomposition based multi-objective evolutionary algorithm.Considering the scheduling characteristics of variable lot splitting and sublots intermingling,a local searching strategy was designed integrating lot splitting/merging with critical path.The performance of the proposed algorithm and the state-of-the-art algorithms were compared under a set of instances of different scales.Experimental results show that the proposed algorithm has good performance on the convergence and distribution of Pareto solution sets.Moreover,the proposed variable lot splitting and sublots intermingling strategy may effectively reduce the energy consumption and makespan.