Abstract: Device failure is the most common dynamic event in shop floor scheduling, which easily affects the original production scheduling plan. In order to ensure that processing production can quickly make adjustments to the original scheduling plan after encountering device failure and reduce the negative impact caused by device failure, the scheduling problem of no-idle multi-device work center under device failure is studied. Firstly, a no-idle hybrid flow shop scheduling model considering setup time and handling time under device failure is established, and the optimization objectives are completion time, handling times and system stability. Secondly, the NSGA-II-V algorithm is designed for solving the original scheduling and rescheduling. Finally, based on the static scheduling results obtained by NSGA-II-V with the idea of partial rescheduling strategy, the device failure cases are tested, and the test results show that the proposed model and algorithm can better handle the zero-idle hybrid flow shop scheduling problem where the dynamic event is completion time failure.

Key words: no-idle multi-device work center scheduling, genetic algorithm, device failure, setup time, handling time

摘要： 设备故障是车间调度中最为常见的动态事件，易对原生产调度计划造成影响。为保证加工生产在遇到设备故障后能够快速地对原调度方案做出调整，降低设备故障所带来的负面影响，对设备故障下的零空闲多设备工作中心调度问题进行研究。建立了设备故障下考虑调整时间和搬运时间的零空闲多设备工作中心调度模型，优化目标为最小化最大完工时间、搬运次数和系统稳定性；设计NSGA-Ⅱ-V算法用于原调度和重调度的求解；基于NSGA-Ⅱ-V获得的静态调度结果与部分重调度策略的思想，进行设备故障算例测试。测试结果表明，所提模型和算法可较好地处理动态事件为设备故障的零空闲多设备工作中心调度问题。

关键词: 零空闲多设备工作中心调度, 遗传算法, 设备故障, 调整时间, 搬运时间