This paper investigates a three-competing group scheduling problem on serial-batching machines considering the setup time of groups and batches, as well as the job-dependent deteriorating effect, where the setup time of groups and batches depends on their own corresponding start time and deterioration rate, and the jobs’ actual processing time depends on their starting time and deterioration rate. In this problem, the jobs in three groups are processed competitively, and the jobs belonging to the same group are processed together on a machine, meanwhile, the jobs from each group are divided into batches. The objective is to minimize the makespan of one group with deterioration effect under the constraint of satisfying the threshold value of another group. Some key structural properties are proposed under the relevant demonstration, and a decision flow chart of scheduling rules is constructed based on these structural properties. Then, an effective improved differential evolution (DE) search algorithm combining the shaking operation from variable neighborhood search is developed to solve the studied scheduling problem. Computational experiments are conducted to evaluate the performance of proposed improved DE algorithm and some other well-known algorithms. The experimental results show that the proposed algorithm is more effective and stable than compared algorithms.

本文考虑了批量生产的组和批次的建立时间，以及与工作有关的恶化效应，研究了批量生产机器上的三竞争群体调度问题，组和批次的建立时间取决于各自的相应开始时间，并且劣化率，而作业的实际处理时间取决于其开始时间和劣化率。在该问题中，竞争性地处理三个组中的作业，并且在计算机上一起处理属于同一组的作业，同时，将每个组中的作业分为几批。目的是在满足另一组阈值的约束下使具有劣化效果的一组的制造时间最小化。在相关论证中提出了一些关键的结构特性，基于这些结构属性，构造出调度规则的决策流程图。然后，开发了一种有效的改进的差分进化（DE）搜索算法，该算法结合了来自可变邻域搜索的抖动操作，以解决所研究的调度问题。进行了计算实验，以评估所提出的改进的DE算法和其他一些知名算法的性能。实验结果表明，所提算法比同类算法更有效，更稳定。进行了计算实验，以评估所提出的改进的DE算法和其他一些知名算法的性能。实验结果表明，所提算法比同类算法更有效，更稳定。进行了计算实验，以评估所提出的改进的DE算法和其他一些知名算法的性能。实验结果表明，所提算法比同类算法更有效，更稳定。