Research On Integrated Modeling And Optimization For Production Scheduling Of Polyester Fiber

As the core part of the modern industrial manufacturing execution system,production scheduling technology can improve the economic and social benefits of enterprises.Intelligent manufacturing with the goal of digitization,networking,and intelligence is the core technology of the new round of industrial revolution.Against the background of the development of intelligent manufacturing,enterprises generate a large amount of valuable data,while how do we obtain effective information from the massive data and apply it to practice to overcome uncertainty remains an urgent problem to be solved.This paper takes the hybrid production process of a domestic polyester fiber workshop as the research object,and an integrated upstream and downstream production scheduling model is established to optimize the workshop production efficiency.Besides,a dynamic scheduling method based on data-driven robust optimization and scheduling knowledge mining under uncertainty is proposed.The specific work is as follows:(1)Propose a static scheduling method for the whole process of polyester fiber hybrid production.Focusing on the complexity and coupling of the production process of the polyester fiber hybrid production process,an integrated MILP model that synergizes the upstream continuous process and the downstream discrete process is proposed.Firstly,we establish the continuous-time event-driven model of the upstream process,and then establish the model of the downstream process considering the batch size and processing sequence.The material and time constraints are used to combine the two models.The optimizer is used to solve the pre-scheduling results.Finally,the case verifies the efficiency of the integrated model.(2)Propose a discrete process dynamic scheduling method based on data mining.To solve the problem that the uncertainty of the discrete production process causes the static pre-scheduling results to deviate from the actual situation,and considering the needs of quick response to the workshop changes,a data mining-based dynamic production scheduling method is proposed.It can mine potential scheduling rules from static pre-scheduling with better solutions and re-direct dynamic workshop production scheduling.Cases are used to verify that the proposed method meets the workshop performance requirements and quick response to workshop changes,proving the effectiveness of the method.(3)Propose a dynamic scheduling method for the whole process of polyester fiber hybrid production under uncertainty.Focusing on the problem of uncertainty in the upstream and downstream of the polyester fiber hybrid production scheduling process,a data-driven dynamic scheduling method combining robust optimization and data mining is proposed.Firstly,the uncertainty set perturbation interval is established by kernel density estimation,then the robust optimization method is adopted to reduce the influence of the uncertainty of material processing time.Considering that the fluctuations in the upstream process affect the downstream production,the data mining dynamic scheduling method is used to overcome the uncertainty of raw material arrival.Finally,the feasibility of the proposed method will be proved by a case.