An ARIMA-based study of pricing and replenishment decisions for vegetable commodities

Abstract

Fresh produce superstores face many challenges as an important part of the livelihood industry. The short freshness period and variable quality make their daily sales efficiency crucial. A simple, efficient and applicable decision model for supermarkets is beneficial to further promote the healthy and sustainable development of the fresh food market and the overall livelihood industry. In order to predict the daily replenishment and pricing strategy of each vegetable category in the coming week, the key lies in analyzing the supermarket's historical data. Since the sales volume shows strong cyclical fluctuations, the time series model ARIMA is chosen for this purpose to observe the model training and predict the pricing of different categories of products in the next 7 days. Due to the cost-plus pricing mechanism of the superstore, after processing the data, we selected simple linear regression model, multiple linear regression model and nonlinear regression model to fit the data, and opted for multiple linear regression to get the sales volume on the cost-plus pricing of the fitting formula, taking into account the discount rate of the different types of vegetables, we corrected the sales volume of the six types of vegetables, and ultimately obtained the superstore's daily replenishment and pricing strategy for the next seven days.