A Particle Swarm Optimization Model for Automatic Pricing and Replenishment of Vegetables using K-means Clustering Algorithm

Fresh produce supermarkets face complex challenges when dealing with the supply of vegetable items. to meet these challenges, supermarkets need scientific methods to guide pricing and replenishment decisions, considering multiple factors and supporting adjustments to adapt to the needs of the market. First, a heat map of Spearman's correlation coefficient between each vegetable category is plotted to analyse the correlation between the categories. Then, a K-means clustering analysis is performed for each vegetable item, and the vegetable items are divided into four categories to analyse the relationship between the same categories sold in bundles. Next, a dynamic mixed-integer programming model is established for the daily replenishment quantity and pricing strategy to maximise profit, and the model is solved by particle swarm algorithm to derive the daily replenishment quantity and pricing strategy of each vegetable category in the coming week (1–7 July 2023). Finally, an optimisation model in terms of individual items is developed and the replenishment quantity and pricing strategy for individual items on 1 July are derived.