Digital-twin-based job shop multi-objective scheduling model and strategy
Zhuo Zhou,Liyun Xu,Xufeng Ling,Beikun Zhang
DOI: https://doi.org/10.1080/0951192X.2023.2204475
IF: 4.1
2023-05-05
International Journal of Computer Integrated Manufacturing
Abstract:Job scheduling is an important part of production management and an effective method for increasing productivity, improving enterprise competitiveness, and reducing production energy consumption. However, in traditional scheduling, some problems, such as low information transparency, response lag, poor accuracy, and poor optimisation, affect the scheduling performance. Therefore, this study presents a digital-twin-based (DT-based) job shop scheduling strategy to address these problems. First, energy consumption is introduced, and a job shop scheduling model that minimises the completion time, tardiness, and energy consumption is established. Then, a scheduling decision framework based on cloud-edge computing is proposed, and the DT-based job shop system composition and operating mechanism are described. In addition, a DT-based shop scheduling strategy, comprising an overall scheduling mechanism, an accurate processing time determination method that uses a time and space compression ratio simulation, and a data comparison and abnormality determination method, is designed. Moreover, the non-dominated sorting genetic algorithm (NSGA-II) is improved by incorporating a multi-mode crossover and random variation method, along with a variable proportion elite retention strategy. Finally, the effectiveness of the improved algorithm and proposed strategy is verified using a standard dataset and practical processing problem.
engineering, manufacturing,computer science, interdisciplinary applications,operations research & management science