Random Forest-Based Restocking and Pricing Prediction for Vegetable Items

Abstract

The intolerance to storage of vegetable commodities in supermarkets makes automatic pricing and replenishment decisions for vegetable commodities particularly important. This paper takes the measured data of a superstore as an example to formulate a set of effective pricing and replenishment decisions for vegetable commodities, which is a comprehensive consideration to ensure the balance of supply and demand, and to reduce the losses of the superstore and the loss rate of commodities. First of all, the sales of each category and single product in different time periods were counted, and the Pearson correlation coefficient was calculated to obtain the distribution pattern of the sales volume of each category and single product of vegetables. Then, the relationship between the total sales volume of and the cost-plus pricing of each vegetable category is analyzed, and a random forest model is established to predict the total replenishment volume and pricing strategy in the coming week. Finally, the replenishment quantity and pricing strategy of individual items are given to maximize the revenue of the superstore under the premise of trying to meet the market demand for each category of vegetable goods. The model established in the paper, which basically solves the given problem, has strong practicality and high computational efficiency.