关键词: 涡流纺, 纺纱车间, 生产调度, 多目标优化, 改进NSGAII

Abstract: To cope with the challenge of gradually changing production of vortex spinning in the textile industry from the traditional single large-volume mode to the small-volume and multi-variety mode, a new scheduling method is proposed in this paper. The method is based on analyzing the order production of a vortex spinning enterprise in Jinhua, and by this method, the article constructs a multi-objective optimization model with the objectives of minimizing the maximum completion time, minimizing the overdue loss and minimizing the number of changeover times. The article also solves the model by the proposed INSGAII_SAA algorithm. The research focuses on improving the production scheduling efficiency to cope with diversified market demands and helping enterprises to reduce the comprehensive cost of production management.
In this paper, firstly, through the analysis of the vortex spinning production process, the vortex spinning scheduling optimization model under the small batch and multi-species mode is established. The model comprehensively considers several production constraints, such as equipment availability, the number of change machines, order priority, etc., to ensure the practical applicability of the model. On this basis, this paper proposes the INSGAII_SAA algorithm, which aims to improve the efficiency and effectiveness of the solution by avoiding the algorithm from falling into problems such as premature convergence. In terms of algorithm improvement, this paper mainly has three major innovations. Firstly, heuristic rules are used to initialize the population to reduce the redundancy of the initial solution space and improve the quality of the population. Secondly, the mechanism of adaptive selection of the cross-mutation operator is introduced. The traditional cross-mutation operation has a trade-off between local search and global search ability, while in this paper, by dynamically adjusting the selection probability of the cross-mutation operator, the algorithm is able to adaptively select the appropriate operator, which improves the algorithm's local and global search ability while guaranteeing the diversity of the population. Thirdly, this paper incorporates the simulated annealing algorithm (SAA) to locally search the individual with the largest crowding distance to further optimize the algorithm's solving ability and effectively avoid the solution set from falling into the local optimum. To verify the effectiveness of the improved algorithm, this paper selects the order data of a vortex spinning enterprise in Jinhua for experiments. The results show that the proposed INSGAII_SAA algorithm outperforms NSGAII in several performance indicators. INSGAII_SAA reduces the maximum completion time by about 8.7% on average, reduces the overdue loss by about 59.89% on average, and reduces the number of changeover times by about 34.91% on average compared to the original NSGAII.
The improved algorithm proposed in this paper shows good results in the experiments, but there are still some limitations. First of all, the production interruption problem that may be caused by factors such as insufficient supply of raw materials or equipment failure is not considered in the model construction process of this paper. In addition, situations such as order insertion or emergency orders that may occur in actual production are also not taken into account in the model. In future research, the scheduling model can be further optimized for these problems to improve the robustness of the model and the adaptability of the algorithm.

Key words: vortex spinning, spinning workshop, production scheduling, multi-objective optimization, improving NSGAII