In this paper, for the automatic pricing and replenishment decision-making problem of vegetable commodities, it is found that there are certain seasonal characteristics in the sales of vegetables in each category, and wavelet analysis is carried out on the daily sales data of single products to get the sales cycle of each category. For the vegetable single product with large amount of data, all single products are divided into four categories of common household vegetables, packaged vegetables, seasonal special vegetables and special vegetables by systematic cluster analysis, and the time distribution characteristics of single products are described by the sales time series of typical members in each cluster. Based on the market demand theory, the cost-plus pricing and sales volume data of the categories were fitted, and the relationship was obtained to be negatively correlated. Taking the purchase quantity and pricing of each category as the decision variables, maximizing revenue as the objective function, and taking the historical maximum purchase quantity of each category and the historical maximum purchase quantity of the superstore as the constraints, a nonlinear programming model was established to formulate the pricing and replenishment strategies for vegetable products. By solving the model, the optimal strategy of replenishment volume and pricing for the future seven-day category is derived, corresponding to the maximum profit of ¥3124.75.