Optimization Of Replenishment and Sales Strategies of Vegetable Commodities Based on Multiple Linear Regression and Non-Linear Programming

Abstract

Automated pricing of vegetable items generates suggested selling prices for the items based on the provided item data, considering the structure of the data. The aim of this paper is to better help superstores plan future replenishment volumes and pricing strategies by analysing sales data for each category and individual item in the vegetable category. Specifically, this paper takes the category as a unit, uses the processed cost-plus pricing, sales volume, purchase price, and loss rate data of each vegetable category to establish a linear regression model, and based on the least squares method, the relationship between cost-plus pricing and total sales volume of each vegetable category is fitted to solve the problem, and the multivariate linear relationship expression with high fit is obtained; based on the cost-plus pricing rule, the future replenishment volume and pricing decision is established based on each category. Based on the cost-plus pricing rule, a nonlinear programming model is established based on the total daily replenishment and pricing decision for each category in the coming week, with the revenue of the superstore as the optimization objective, and considering the constraint of the transport loss rate, the total daily replenishment in the coming week under the maximal revenue is solved by using MATLAB to get the corresponding replenishment strategy, and then the total daily replenishment is substituted into the multivariate linear expression to calculate the corresponding pricing strategy.