Research And Application For The Streamlining Of Multi-species And Small Batch Production

In a variety of small quantities of production models,many companies often produce a lot of waste when faced with the production of this type of product due to the lack of systematic production guidance.Problems such as material anomalies,low utilisation of machinery and equipment,high lineside inventory,long production cycles and logistical chaos occur frequently.This paper proposes a comprehensive optimisation theory system combining streamlining and modern optimisation theory for multiple small quantities,based on a synthesis of streamlining theory and modern optimisation theory,combining the advantages of both.In order to meet the characteristics of multiple and small quantities of products in the automotive parts industry,the theory of streamlining production method is firstly used to clarify the direction of optimisation for these products as a whole,and then the key issues involved in the process of streamlining are improved by using the methods of modern optimisation theory.In order to deal with the problems of multiple products,complex production processes and long production cycles,it is necessary to classify products into groups in the process of streamlining.This paper firstly uses the fuzzy clustering method to cluster the products into groups,and uses the F-statistic in fuzzy mathematics to determine the best clusters,and finally uses the grouped data to train IWO-BP neural network and ordinary BP neural network.The results show that the IWO-BP neural network algorithm has faster convergence speed and more accurate convergence accuracy than the ordinary BP neural network algorithm.In response to the problems of high material handling costs and chaotic logistics,a reasonable layout of the equipment on the production site is required in the implementation of streamlining production.In this paper,the layout of the workshop is firstly divided into the layout within the grouping unit and the layout between the grouping units,then the product is designed into several grouping units according to the grouping process,and the layout within the unit is completed by using the "U" layout method,and finally the mathematical model between the grouping units is established with the objective function of minimising the material handling cost.Finally,the mathematical model of inter-cell layout is established with the objective function of minimising material handling cost,and the inter-cell layout is solved by genetic algorithm with hybrid coding technology.Based on the demand for synchronisation of production logistics,this paper designs a synchronised logistics approach to meet the demand.An information system framework is designed to synchronise logistics outside the factory and inside the factory to monitor the production logistics in real time and provide timely feedback on relevant logistics information.An electronic kanban pulling process is designed,using E-kanban to transmit real-time information and finally realise synchronous logistics.