Digital Twin-Based Job Shop Deviation Detection and Real-Time Scheduling

Scheduling is one of the key factors affecting production efficiency. The scheduling deviations caused by the impact of internal and external environment are easy to lead the rescheduling. By introducing the digital twin (DT), the scheduling deviation between actual and planned production can be captured and the deviation can be expressed quantitatively to trigger rescheduling strategies more precisely. For the real-time scheduling problem (RSP), a DT-based deviation detection method and scheduling framework are proposed. A multi-objective model for RSP and a rescheduling strategy based on deviation detection are developed. In addition, an improved non-dominated sorting genetic algorithm (INSGA-II) is proposed to solve the RSP. The algorithm combines the adaptive mutation method and the neighborhood search method. Finally, an assembly job shop is taken as a case study to illustrate the effectiveness and advantages of the proposed framework and methodology.