Automated pricing and replenishment decisions for vegetables based on price elasticity

In order to solve the problem of replenishment and pricing of fresh supermarket vegetable products, this paper establishes a multi-objective optimization model based on the sales flow, wholesale price and loss rate data from July 1, 2020 to June 30, 2023. Firstly, the data were integrated, cleaned and classified, and the perennial, seasonal and seasonal characteristics of vegetable products were analyzed, and the profit outliers and unsalable products were eliminated. Through time series decomposition and partial correlation analysis, the distribution law and correlation of sales volume of each category and single product are revealed. The dual-logarithmic demand model was constructed, combined with the LSTM to predict the sales volume and cost in the next seven days, and the improved GBest-PSO algorithm was used to optimize the replenishment and pricing strategies, which significantly improved the revenue of supermarkets. Based on the VIKOR evaluation method and the improved NSGA-II algorithm, the mixed multi-objective nonlinear programming problem is solved, the optimal single product replenishment volume and pricing strategy are determined, and the dual optimization of revenue and market demand is realized. In response to question 4, some suggestions are put forward to collect data such as market pricing, consumer price sensitivity and residents' spending power, so as to further improve the scientific decision-making. The research in this paper provides theoretical support and practical guidance for the replenishment and pricing decisions of fresh supermarkets.