Research on Optimization of Demand Forecasting-Based Inventory Control Systems

With the upgrading of production and operation methods of fresh agricultural products in China, people's requirements for the timeliness and quality of fresh products are increasing. However, currently, China's fresh supply chain faces serious premium issues, high logistics and storage costs, and great losses and high costs at the retail end, which are not conducive to the healthy development of the market. Too small inventory can affect customers' choices, while too large inventory can increase corporate costs. To ensure fresh agricultural products appear on consumers' tables in higher quality, this paper, against this background, analyzes the profile and inventory management status of P Supermarket, and through the analysis and forecasting of historical sales data, finds problems such as unreasonable inventory control and low accuracy of customer demand forecasting. This paper focuses on using a scientific demand forecasting optimization research model to analyze and solve the problems existing in P Supermarket. It also uses Matlab for inventory optimization control, applies the particle swarm algorithm for 500 iterations to obtain the optimal inventory level and determine the optimal order quantity.