The distributed permutation flowshop scheduling problem (DPFSP) has attracted much attention in recent years. In this paper, we extend the DPFSP by considering the sequence-dependent setup time (SDST), and present a mathematical model and an iterated greedy algorithm with a restart scheme (IGR). In the IGR, we discard the simulated annealing-like acceptance criterion commonly used in traditional iterated greedy algorithms. A restart scheme with six different operators is proposed to ensure the diversity of the solutions and help the algorithm to escape from local optimizations. Furthermore, to achieve a balance between the exploitation and exploration, we introduce an algorithmic control parameter in the IG stage. Additionally, to further improve the performance of the algorithm, we propose two local search methods based on a job block which is built in the evolution process. A detailed design experiment is carried out to calibrate the parameters for the presented IGR algorithm. The IGR is assessed through comparing with the state-of-the-art algorithms in the literature. The experimental results show that the proposed IGR algorithm is the best-performing one among all the algorithms in comparison.

分布式置换流水车间调度问题（DPFSP）近年来引起了很多关注。在本文中，我们通过考虑依赖于序列的建立时间（SDST）来扩展DPFSP，并提出一种数学模型和带有重启方案（IGR）的迭代贪婪算法。在IGR中，我们放弃了传统的迭代贪婪算法中常用的模拟退火样接受准则。提出了具有六个不同运算符的重启方案，以确保解决方案的多样性，并帮助算法摆脱局部优化。此外，为了在开发和探索之间取得平衡，我们在IG阶段引入了算法控制参数。此外，为了进一步提高算法的性能，我们提出了两种基于进化过程中建立的工作块的本地搜索方法。进行了详细的设计实验，以校准提出的IGR算法的参数。通过与文献中的最新算法进行比较来评估IGR。实验结果表明，提出的IGR算法是所有算法中性能最好的一种。