Research on Vegetable Pricing and Replenishment Strategies Based on ARMA Models

Abstract

This paper explores the importance of pricing and replenishment issues of vegetable items for merchant operations. First, data related to vegetable stocking were systematically collected to assess the correlation between different categories. Through Q-Q plot analysis, it was determined that the data did not conform to a normal distribution, so the Spearman correlation coefficient was chosen for further quantitative analysis. Considering the large number of individual vegetable categories, in order to deal with the correlation between different individual items, a hierarchical clustering method was introduced in this paper, which classified them into three categories: low volume, medium volume and high volume. In order to predict the future vegetable intake, this study adopted the seasonal ARMA model and applied the Pearson's correlation coefficient for correlation analysis by combining the cost utilization rate as well as the total sales volume. In addition, the relationship between sales and cost margin was fitted with the help of support vector machine. Finally, this study constructs a single optimization model with the objective of maximizing the revenue of the superstore. The innovation of this paper is the combination of various statistical and machine learning methods in order to solve the pricing and replenishment problem of vegetable items, aiming to improve the efficiency of merchant operations.