Research On Modeling And Optimization Algorithms Of Dyeing Vat Scheduling

Printing and dyeing process is one of the most important links in the textile industry chain.Due to the dyeing process takes a long time,in which a lot of constraints must be taken into account,and it is high randomness,dyeing vat scheduling is very complicated and highly technical.Manual scheduling not only consumes a lot of manpower and material resources,but also lacks the control to the quality of production schedules.The most important thing is that manual scheduling has become the fault between the upper information management system and the bottom production control systems.With the development of the textile industry,the production requirements are constantly changing,and the existing research results have been unable to meet the application requirements well.Studying a reasonable dyeing vat scheduling model and designing an effective intelligent optimization algorithm not only can help improve the efficiency of the enterprise and reduce production energy consumption,but also is important for realizing the automated production of the dyeing shop and the intelligent regulation of the scheduling.Therefore,by considering the actual requirements,this paper studies the dyeing vat scheduling problem model and its optimization algorithms.The main work and results are as follows:First,the printing and dyeing production process and its characteristics are summarized,and the existing dyeing vat scheduling model is expanded based on the specific problem background.Taking the washing cost,dyeing vat switching cost,and delay cost as the optimization goals,five constraints are added: the constraint for fluorescence prohibited product,the constraint for vat maintenance time,the concurrent constraint for vat loading and unloading,the constraint for the first vat of a product,and the constraint for the batch-splitting of a product.Then,the constraint model of dyeing vat scheduling is established.Then,using the product as the basic scheduling unit,a sliding time window heuristic scheduling algorithm is designed based on the constructive algorithm,in which batching product and scheduling batches are in a regular order.When batching the products,the ordinary products(the products previously scheduled)and the newly added products are processed separately.The ordinary products are divided into batches by splitting and combining,while the newly added products directly obtain the batching plan from the original plan.When scheduling batches,the dyeing vat and processing time of batches are adjusted to generate a feasible scheduling plan for the product.Numerical experiments show that the proposed sliding time window heuristic scheduling algorithm is efficient.Compared with the results of manual scheduling,the solution quality is higher.Finally,in order to further optimize the scheduling results,an exploratory study on product batch scheduling optimization under ideal conditions is conducted.By relaxing the concurrent constraint for the loading and unloading of a vat,using the batch sequence of a dyeing vat as the basic scheduling unit,a batch scheduling model based on column generation is established,and a branch pricing algorithm is designed to solve the model.In the solution process,according to the characteristics of the problem,several key solutions of the algorithm are proposed.Among them,the price sub-problem is designed with two solving algorithms,those are the price sub-problem algorithm based on improved greedy search and variable neighborhood search.Combined with numerical experiments,the feasibility of the model and algorithm are verified,and the two price sub-problem algorithms are analyzed.The experimental results show that the target values of most calculation examples are obviously optimized,and the price sub-problem algorithm based on variable neighborhood search is good at finding better quality solution in an acceptable time.