Automated Vegetable Pricing and Replenishment Decisions based on PSO

Abstract

The main aspects that supermarkets need to consider are the purchase volume and pricing, especially the freshness of vegetable products, so it is more important to determine the daily replenishment volume. This paper divides 251 typical vegetables into 6 categories for the market, collects the daily sales volume, daily sales unit price and cost price and loss rate of each vegetable in a typical supermarket within three years, and formulates replenishment and pricing strategies for the coming week to maximize the operating profit of the supermarket. This article constructs a "cost-plus pricing" model. First, the regression model is used to analyze the relationship between total sales and cost-plus pricing for different vegetable categories. Secondly, using the single-objective optimization model, taking revenue as the objective function, sales volume and cost markup as the decision variables, considering the constraints such as the relationship between total sales and pricing, and solving with particle swarm optimization to find appropriate replenishment and pricing strategies, so as to maximize the revenue of supermarkets in the coming week. Finally, using time series forecasting, the sales volume of each vegetable category in the next week is predicted by using the sales flow of the previous three years, and the replenishment volume obtained by the above optimization model is compared to detect the reliability of the model, and the results show that the reliability of the model results is high.