Dynamic Scheduling Of Two Stage Multi Assembly Line Based On PID Control And Finite Buffer

With the rapid development of manufacturing technology and increasing demand of the market,the traditional mass procutcion mode is gradually replaced by the small batch and multi variety mode,and some products due to the structural requirements and the requirements of the agile assembly,a “pre-assembly” two stage procution mode with supply and demand relationship was emerge in the process of assembly.The finished components and parts in upstream are transport to the downstream go through buffer area,this requires scheduling results can not only make the downstream line finish assembly work in time,can also make the upstream line and downstream line coordinated production,at the same time taking into account that the actual production process will appear many factors of disturbance,so the scheduling scheme should also have some robustness.This paper take the SD product's assembly process of NFJ company as the research object,to research the dynamic scheduling problem of two stage multi assembly line with buffer.The research work has a good theory and a practical application value.The main work and achievements of this paper are as follows:.1.The characteristics of two stage assembly line scheduling problem and the problem existing in the current related research field are analyzed,the possibility of applying the dynamic scheduling method in the field of industrial control to the production field is discussed,the idea of solving the scheduling process divided into two parts is proposed.2.The downstream assembly feedback scheduling model is established based on maximize customer satisfaction,the maximum assembly line utilization and minimize the completion time,according to the characteristics of PID controller and fuzzy theory,a PID control optimization algorithm is proposed,and the PID parameters are determined by the differential evolution algorithm,the validity of the algorithm is verified by a numerical example.3.The upstream scheduling model was established based on minimizing the penalty cost and minimizing the buffer according to the demand of downstream line with buffer,an improved multi objective particle swarm optimization algorithm for pareto is designed,the effectiveness of the algorithm is verified by two 6x6 MOFJSSP problem.4.Builded the simulation model of two stage multi-production lines dynamic scheduling with supply and demand relationshio based on Visual Basic and MATLAB,complete simulation interface and program design5.Apply the dynamic scheduling model and algorithm that builded in this paper to NFJ company's workshop,compare the results of this paper with company's actual scheme,verify the feasibility and superiority of the model and solution algorithm.