Automated pricing of vegetable commodities has been continuously developed and applied in recent years. More and more superstores utilize computer technology and data analysis methods to automatically calculate and adjust commodity prices through algorithms based on market supply and demand, cost, market competition and other factors in order to achieve efficient pricing decisions. This study aims to provide a series of management strategies for hypermarkets, including conducting large-scale data processing for each sales item, forecasting and fitting relevant sales, planning future replenishment and pricing strategies to increase profits and reduce wastage. First of all, data organization and cleaning are carried out, and by drawing category sales charts and single-item sales change charts, we can initially understand the distribution pattern of sales volume, and find that there are strong temporal sales volume differences between vegetable categories and single items. Then an ARIMA time series model is established to fit and analyze the daily sales volume of each category of vegetables, so as to deeply analyze its distribution law. Finally, a multi-objective planning model was established by considering a variety of factors. In it, we take the maximization of the total revenue of the superstore as one optimization objective, and at the same time satisfy the commodity demand of the market for each vegetable, and consider the damage situation as well as the correction of the shelf life. We also set constraints to ensure that the total number of saleable items is between 27 and 33 and to satisfy that the order quantity of each item is not less than the minimum display quantity of 2.5 kg. Finally, we use a greedy algorithm to perform an optimal estimation to determine the replenishment quantity and pricing strategy for the final individual items.