A novel digital twin-assisted prediction approach for optimum rescheduling in high-efficient flexible production workshops

The optimum reschedules usually need to be considered in the flexible production workshop according to the actual production requirements to ensure the higher efficiency of production line and the better performance of machining operation. Generally, the reschedules after the order arrival can be acquired using the traditional rescheduling methods that might cause a time-consuming process. To solve these problems with respect to the order arrival event, a digital twin-assisted prediction rescheduling approach is proposed to support the efficient reschedules in the flexible production workshops. By considering the differences between traditional reschedules and the predictive ones, the prediction rescheduling method is used to specify the rescheduling strategy of flexible production workshops based on the order arrival hypothesis, which can acquire the optimal reschedules before the order arrival. Simultaneously, the rescheduling model is proposed to consider the dynamic and static parameters in the distributed calculation strategy based on backtracking searching optimization algorithm. By combining the digital twin-assisted production workshop, a case study is use to verify the feasibility of the proposed method. The experimental results show that the predictive rescheduling approach can acquire optimal reschedules before order arrivals, which can significantly reduce the reactive time to further improve real-time performance of production workshops.

Automated summary

In order to ensure higher efficiency and better performance in the flexible production workshop, the optimum reschedules need to be considered based on actual production requirements. Traditional rescheduling methods might be time-consuming for order arrivals. To solve this problem, a digital twin-assisted prediction rescheduling approach is proposed to support efficient rescheduling in flexible production workshops. By considering the differences between traditional and predictive rescheduling, the proposed method can acquire optimal reschedules before order arrivals, reducing reactive time and improving real-time performance of production workshops.