Research On Supply Chain Collaboration Model For Intelligent Manufacturing

With the deep integration of technologies such as big data,Internet of Things,and intelligent control with the manufacturing industry,traditional manufacturing has begun to transform to intelligent manufacturing.As an efficient way of supply chain management,supply chain collaboration is of great significance to the development of intelligent manufacturing,and it has attracted widespread attention and research in the academic community.Aiming at the problems of supply chain demand uncertainty in the intelligent manufacturing scenario,low level of collaboration between manufacturing companies,large environmental impact,low service quality,etc.By collecting historical demand data,establishing a supply chain demand forecasting model,realizing the collaborative optimization of the supply chain based on demand forecasting,and improving the collaboration level and user satisfaction of intelligent manufacturing enterprises.The research content is as follows:(1)A LSTM-based supply chain demand forecasting model is proposed.Based on the time series data collected by the intelligent manufacturing supply chain,this thesis proposes a predictive model called IPSO-LSTM-corr.The model uses an improved particle swarm algorithm based on LSTM to optimize the number of hidden layer neurons and learning rate of the network.In addition,a correction layer for correcting the output results is added to the output of the LSTM,so as to realize automatic parameter optimization and provide more accurate prediction results.The experimental results show that the prediction effect and training speed of the IPSO-LSTM-corr model are significantly better than that of the LSTM model,which lays a foundation for the establishment of a subsequent supply chain collaboration model.(2)A collaborative optimization model of supply chain for intelligent manufacturing is proposed.Based on the results of the prediction model,this thesis establishes a supply chain collaborative optimization model.Taking a two-level supply chain with multiple production stages as an example,this thesis considers the cost of the supply chain and the carbon emission as another optimization objective,and considers the impact of production,transportation and inventory costs on carbon emission.In order to solve the proposed multi-objective optimization model,this thesis improves agmopso algorithm,improves the search ability and convergence speed of the algorithm by improving the inertia weight update method,and proposes a local evolution strategy to solve the problem of too few solutions in the external archive.The experimental results show that the proposed collaborative optimization model can significantly reduce the cost and carbon emissions of the supply chain in the intelligent manufacturing scenario,and the improved algorithm has good optimization ability and convergence speed.