Research On Flow Shop Scheduling Problem Under Blocking And Moment Uncertainty

Production scheduling play a very important role in improving production efficiency and reducing operating costs.It exist widely in production process such as manufacturing,textile,pharmaceutical and food processing.At home and abroad,research on scheduling models,solution methods and related applications is not only fruitful,but also continuous.However,due to the complexity of the process industry production process,there is still room for continuous improvement in terms of congestion and uncertainty.In this thesis,we have read and analyzed relevant domestic and foreign literature,and carried out the following work in view of the blocking status and uncertain processing time:(1)According to the description of the production scheduling,the different types of scheduling are classified.Then,the current situation of the scheduling problem at home and abroad is analyzed and sorted,and the method for solving the scheduling problem is summarized.Finally,the trend of research on production scheduling at home and abord is analyzed and prospected.(2)The blocking flow shop scheduling problem is studied to realize production efficiency optimization with the aim of minimizing makespan.First,we analyze the characteristics of the scheduling problem and establish a corresponding mathematical model.Then,we designed an improved greedy iterative algorithm(IIG)based on variable neighborhood search conbined with the characteristics of the blocking scheduling problem,improved the generation of the initial solution,and designed a new search operator to further improve the current solution.Simulation experiments on large-scale instances show that better solutions can be obtained than other algorithms.(3)The no-wait flow shop scheduling problem under uncertain processing time is studied,and we propose a deterministic model and solution method.In order to improve the degree of freedom of the uncertainty modeling,fuzzy variables of the intervalvalued fuzzy sets(IVFS)are used to describe the uncertainty of the processing time.In terms of modeling,based on the confidence,a chance-constrained programming model of R-NWFSP is constructed with the optimization goal of minimizing makespan.In terms of solution technique,a two-layer simulated annealing algorithm(MNS-TLSAA)based on multi-neighborhood structure was developed to solve the deterministic model efficiently.Experimental results show that designed algorithm has good stability,and the reliability of the deterministic model and solution strategy is verified by sensitivity analysis.(4)Based on the research of the above flow shop scheduling,we apply the related model and method to the reheating furnace scheduling in actual production environment.First,we analyze and sort out the actual processing requirements such as furnace capacity constraints and hot rolling continuity constraints for slab production,and establish a scheduling optimization model based on the total heating time after all slabs reach the rolling heating temperature.Then,we combined the characteristics of the problem to apply the above optimization methods and strategies to the problem.Simulation experiments based on analysis and data collection on steel production confirmed that better target values can be achieved.