Vegetable Automatic Pricing and Replenishment Decision Based on ARIMA Model and Particle Swarm Optimization Algorithm

This paper aims to analyze and address the automatic pricing and replenishment decision problem for vegetable products. Initially, a comprehensive analysis of the correlation between individual vegetable items and categories is conducted using Spearman and Kendall correlation analyses. The study employs optimization techniques such as the particle swarm algorithm to solve a constructed 0–1 programming model. Additionally, various predictive models, including Copula, Informer, and Radial Basis Function Neural Network (RBF), are utilized to obtain detailed data on future sales volume, purchase price, wholesale price, and more. MATLAB, PYTORCH, and other software tools are employed for data visualization, model fitting, and other operations to analyze the correlation between different categories and individual vegetable items from multiple perspectives. Based on the constructed planning model, the paper provides replenishment quantities and pricing strategies for each vegetable category under different constraints, aiming to maximize the benefits for supermarkets.