A Dynamic Equilibrium Allocation Framework for Intelligent Blending Using Metaheuristic Optimization

In tobacco production, maintaining a dynamic equilibrium in the distribution of both the quantity and the variety of flipped tobacco grand across various flipping stations during the silk thread blending process presents a significant challenge. This study develops a mathematical model for dynamic equilibrium allocation by examining the relationship between the brand variety and quantity of mixed materials in production, alongside the material processing requirements at the mixing and turning station. The study further explores the characteristics of both saturated and unsaturated states of the turning station. By harnessing the advantages of metaheuristic algorithms, an intelligent framework for dynamic equilibrium allocation of materials is proposed. A repair operator is designed to transform non-feasible solutions into feasible ones, thereby enhancing the accuracy of results by incorporating the specific characteristics of the flipping box tobacco as it enters the station. The experimental findings indicate that the proposed framework successfully achieves optimal allocation of tobacco processing quantities across various turning stations under saturated conditions. In scenarios characterized by unsaturated conditions, the discrepancy between the maximum quantity and the maximum number of grands remains within a margin of 1. This framework adeptly addresses the application challenges associated with intelligent box turning and blending in silk production, accommodating flexible production and processing requirements, and establishing a solid foundation for future advancements in optimization technologies.