Data-Driven Fresh Food Supply Chain Management and Market Competitiveness Research

Vegetable commodities have the characteristics of short freshness period, variety, and special purchase time. Therefore, it is particularly important to predict and manage the purchase and shipment of vegetable commodities. Good management can significantly improve the profits of supermarkets and save costs. For the relationship between vegetable categories, we studied the seasonal distribution of sales volume of each category of vegetables based on Spearman correlation analysis, and then studied the cyclical pattern of vegetables based on the ARIMA algorithm model with a sliding window of seven days. We also predicted the sales trend of vegetables and predicted the cost data based on the LSTM model, and predicted the vegetable prices and inventory for the next seven days based on the particle swarm optimization algorithm, obtaining the optimal results.