Vegetables, as an indispensable consumer goods in the daily life of consumers, their price and supply have a great impact on the sellers, urban and rural residents and the whole country. In this paper, the sales flow and loss of 251 vegetables in a supermarket within three years are selected as the test data. On the basis of the analysis of the historical sales volume, price and loss data of vegetables, the vegetable price and supply prediction model based on the taboo search algorithm is put forward from the perspective of sellers. First of all, according to the recent sales data, the average daily average sales volume of various single products is obtained, and the varieties that meet the display demand are determined as available varieties, and 36 available varieties are obtained. Further use of 0-1 planning to determine 27-33 available varieties. The linear regression was used to analyze the relationship between the price of each item and obtain the linear regression equation. Significance analysis of the equations showed a high degree of fit. Finally, with the benefit as the objective function, the optimization model is established combined with the above constraints and solved using the taboo search method. The results of the model solution show that under the constraints, the maximum revenue of July 1 is 1787.61 yuan, and the replenishment quantity and pricing of 28 single products sold at this time can be obtained. Finally, the reliability of the model was further tested by using the time-series prediction. This paper can not only effectively predict the price and sales volume of vegetables, but also formulate reasonable vegetable replenishment and pricing strategies for the sellers, but also has important guiding significance for the sales of other commodities.