Research On Scheduling Method Of Hybrid Flow Body-in-White Welding Production Line Based On Variability

In order to adapt to the diversified order demands of consumers,more and more models are produced on the automobile production line at the same time.The welding production line has developed from the original single product processing to the processing of different configurations and varieties of multiple models,too.Besides,there is a lot of variability in the enterprise production environment.These dynamic uncertainties make the static scheduling method hard to apply on the environment,so the scheduling problem of the production line becomes more and more complex.This thesis is supported by the National Natural Science Foundation of China project “Research on Performance Prediction and Optimization of Semiconductor Manufacturing Systems Based on Variability(Grant No.71671026)”,and the bottleneck prediction and dynamic scheduling of white body welding production line are mainly studied..Firstly,this thesis takes the body-in-white mixed flow welding production line as the research object.According to the characteristics of the body-in-white welding workshop under mixed flow production and based on the constraint theory,the mathematical model of the scheduling problem is established,and the dynamic scheduling algorithm overall flow of the welding production line driving by the bottleneck drift and periodic events is designed.Secondly,the future bottleneck of the production line is predicted,and it is compared with the theoretical bottleneck to estimate whether the bottleneck drifts.According to the combination of the order and the processing capacity of the production line,the identification method of the theoretical bottleneck is given.In the dynamic production environment,the bottleneck is not fixed,and the effect of optimization on the existing bottlenecks in the system is poor.Therefore,this thesis introduces the time series prediction algorithm to predict the next moment process parameters.After extracting the principal components by the kernel principal component algorithm,the bottleneck process is predicted by using the TOPSIS.By comparing the theoretical bottleneck with the dynamic bottleneck,the bottleneck drift status can be obtained.Then,the intelligent optimization algorithm is used to solve the welding production line scheduling problem.With the body sequence as the variable to be optimized,the particle coding and updating strategy is designed,and the bottleneck resource optimization model with the goal of minimizing the maximum completion time,the maximum delivery time ahead,and the maximum similarity to the sequence before scheduled.The bottleneck heuristic algorithm is used to generate the initial solution,and the multi-objective particle swarm optimization algorithm is used to solve the problem.Based on the principle of pulling in front and pushing behind,the operation scheduling of the whole production line is completed.Finally,the dynamic scheduling system for welding production lines based on constraint theory is implemented in Python.According to the requirements,the design and implementation of the scheduling system function module and software interface are carried out.The research results of the production scheduling method based on the constraint theory are applied to the actual situation,which shows that the dynamic scheduling algorithm designed in this paper has practical value.