Research On Optimization Method Of Production Supply Chain Of New Retail Enterprises Based On Combination Model

With the vigorous development of the market economy,the market demand pattern is constantly changing,and major enterprises are also in the stage of rapid development.Under the supply chain environment,enterprises have been unable to independently complete production planning arrangements,and traditional production planning management solutions have been unable to meet the needs of the market.Based on the supply chain environment,"new" retail enterprises need to cooperate with each other in each link of the supply chain when formulating production plans,continuously exchange and share information,and complete cooperative development.As a "new" retail enterprise,how to manage the growing business data and how to accurately extract valuable information for the company’s sales and production forecast from the massive data has become a new problem faced by the "new" retail enterprise.In this paper,combined with enterprise sales and production data,the combined forecasting model based on LSTM-BP improves the forecasting accuracy of sales and production,and studies the optimization method of production planning for supply chain,aiming to improve the development competitiveness of new retail enterprises in the supply chain environment.The main research contents of this paper are as follows:1.Propose and construct a commodity sales forecasting model based on the LSTM-BP combination model.Based on the operating data of Beijing A company,combined with market environmental factors,actual production conditions of the company and theoretical research,this paper firstly extracts and analyzes the data of the product production process,and divides the massive sales data;Considering the impact of sales activity strategies on sales results,calculate the influencing factors of commodity activities,and build a combined commodity sales prediction model based on LSTM-BP;finally,for the combination prediction model and the single prediction model,the parameters are continuously adjusted in the model training process,and the The experimental results are compared and analyzed to improve the accuracy of the combined sales forecast model,and finally obtain the optimal forecast results.2.Propose and build a production forecasting model in the supply chain environment.According to the characteristics of production planning in the supply chain environment,the impact of safety stock on production planning is mainly analyzed.Based on this,combined with warehousing data,calculate the safety stock of products,build a production forecasting model in the supply chain environment,and continuously improve the accuracy of production forecasting through experimental data results.3.Design and implement a production data analysis system based on supply chain.First,the system mainly implements data query and analysis for inventory,sales and production data;secondly,it implements a commodity sales forecast model and a production forecast model,and digitally analyzes and displays the forecast results.The whole system provides a better data analysis platform for the sales and production data of Beijing A company,optimizes the production plan of Beijing A company,and solves problems such as backlog of commodity inventory and difficulty in capital turnover.According to the production planning optimization problem in the supply chain environment,combined with the LSTM-BP combined commodity sales forecasting model and enterprise warehouse storage data,the paper analyzes the factors of product safety stock,constructs a production forecasting model based on supply chain,and provides a feasible method for enterprise production planning.Optimize the program and conduct experimental data analysis.The test results show that the proposed optimization scheme of production calculation can effectively predict the market sales demand,improve the rationality of the production plan of the enterprise,and at the same time reduce the backlog of inventory goods.