A disturbance evaluation method for scheduling mechanisms in digital twin-based workshops

In the workshop scheduling problem, frequent disturbances lead to continuous and frequent rescheduling. This is detrimental to the optimal utilization of production resources, the maximization of production efficiency, and the minimization of operational costs. Therefore, finding an effective method to reduce frequent rescheduling is crucial for stable and efficient workshop operations. This paper introduces digital twin (DT) technology. Serving as a digital replica of a physical system, DT establishes an interactive connection between the physical entity and its digital counterpart. It has been applied in multiple fields. A disturbance evaluation method for scheduling mechanisms based on DT technology is proposed and studied in this paper. Firstly, this method evaluates the impact of disturbances by using a causal factor chart (CFC) and convolutional neural networks (CNN). Then, corresponding scheduling mechanisms are proposed based on the degree of disturbance impact. DT technology is used to provide data and model support throughout the entire process. Through the experimental verification, workshop disturbances were accurately evaluated by this method. Two unnecessary instances of rescheduling were avoided, resulting in a 66.3% reduction in the time to handle disturbances. The experimental results show that the proposed method can enhance the adaptability of scheduling mechanisms and contribute to a more agile response to disturbances.