Research And Software Design Of Dip-Dye Production Scheduling Optimization In A Dyeing Enterprise

Dyeing process is the core of dip-dye production and also the major polluting process in dyeing enterprises. Vat-based dyeing process scheduling has become a complicated and high technical work with a long processing cycle, large energy consumption, multi-resource constraints and high randomness. Many of the dyeing process scheduling plans are still accomplished manually in China, which is not only time-consuming and laborious, but also very difficult to take the factors affecting the scheduling of vat into account, thus often resulting in imbalance in the use of equipment, serious waste of water, steam, electricity, and other kinds of energy. Reasonable scheduling of vat dyeing production can not only make full use of plant production capacity to ensure order fulfillment, but also reduce energy consumption and waste water emissions. Research on dyeing production scheduling has important practical significance economically and socially.In this thesis, we regarded the van-based dyeing production scheduling problem as a batching machines scheduling problem. Focusing on the needs of a dyeing production enterprise, two optimal scheduling methods of dyeing process were proposed, with the actual production constraints considered, and the corresponding mathematical models were established and solved respectively.(1) As the general solution of the batching machines scheduling problem, we used the concept of slot and continuous-time expression to allocate the vats and orders to the slots as two types of variables--0 and 1, then established the short-term scheduling MILP model of multi-product single stage batch plants with parallel lines for vat arrangement. Different objective functions could be selected based on the sales cycle of the production and the degree of urgency. Moreover, a number of heuristic rules were introduced to reduce the number of integer variables, continues variables and constraints and speed up the solving process. Finally, the branch-and-bound method was used to get the solution.(2) An improved genetic algorithm method was proposed for dyeing process scheduling, where the mathematical model was established based upon the human experience of vat scheduling, taking the delivery period, switching cost and other constraints, the volume differences among vats and the remaining orders in the last plan period into consideration. The improved algorithm used genetic algorithm for global search and climbing algorithm for local search to improve the searching process.Finally, an optimal vat scheduling software system was designed, the optimal scheduling methods proposed in this thesis were realized as two modules, with an manual scheduling module to fine-tune the results. The system framework, database, user interface and maneuverability were considered, and some auxiliary modules were provided to meet the needs of the scheduling-making process.