As the existing manufacturing systems may not be ready to support flexible decisions for smart production with increasing product mix and shortening product life cycle, it is crucial to rapidly respond to dynamic needs to improve bottleneck productivity and ensure the throughput of the whole manufacturing system. Limitations of the existing approaches can be traced in part to the lack of a framework within which different decisions in real settings can be integrated and aligned in light of the changes of manufacturing contexts. To fill the gaps, this study aims to develop a dynamic scheduling and dispatching system with the constructed discrete event simulation model to optimise the scheduling for bottleneck and associated dispatching rules for remaining processes, while considering the manufacturing system as a whole to empower smart manufacturing and reduce waste for sustainable production. A hybrid genetic algorithm is developed to minimise photolithography capacity loss, while considering the waiting time constraints and determining the optimal dispatching rules for non-bottleneck workstations by the design of experiments and multi-criteria decision analysis to integrate related decisions for the manufacturing system as a whole. An empirical study was conducted for validation. The results have shown its practical viability.

中文翻译：

---

基于仿真的工业3.5多标准性能评估动态调度调度系统和可持续TFT-LCD阵列制造的实证研究

由于现有制造系统可能无法支持智能生产的灵活决策，增加产品组合和缩短产品生命周期，因此快速响应动态需求以提高瓶颈生产力并确保整个制造系统的吞吐量至关重要。现有方法的局限性部分可以追溯到缺乏一个框架，在该框架中，可以根据制造环境的变化整合和调整实际环境中的不同决策。为了填补这一空白，本研究旨在开发一种动态调度调度系统，并利用构建的离散事件仿真模型来优化瓶颈调度和剩余进程的相关调度规则，同时从制造系统整体考虑，赋能智能制造，减少浪费，实现可持续生产。开发了一种混合遗传算法，以最大限度地减少光刻容量损失，同时考虑等待时间约束，通过实验设计和多准则决策分析确定非瓶颈工作站的最佳调度规则，将制造系统的相关决策整合为一个所有的。进行了实证研究以进行验证。结果显示了其实际可行性。同时考虑等待时间约束，通过实验设计和多准则决策分析确定非瓶颈工作站的最优调度规则，将相关决策整合为一个整体。进行了实证研究以进行验证。结果显示了其实际可行性。同时考虑等待时间约束，通过实验设计和多准则决策分析确定非瓶颈工作站的最优调度规则，将相关决策整合为一个整体。进行了实证研究以进行验证。结果显示了其实际可行性。