The vegetable sector significantly impacts various aspects such as securing material supply, optimizing industrial chains, and alleviating poverty. However, the perishable nature of fresh agricultural products results in substantial losses for retailers. Accurately and efficiently forecasting vegetable product sales is advantageous for minimizing food waste and enhancing the profit margins of fresh produce retailers. This study utilized the Prophet model to analyze and predict the time series of fresh product sales. A fresh supermarket was analyzed from July 1, 2020, to July 1, 2023, for statistical purposes. This study considered the seasonality of vegetables (seasonal and perennial) and their popularity (hot-selling, best-selling, and flat-selling products), leading to the classification of individual items. The Prophet model was then employed for forecasting, and the various components of the model are decomposed for time series analysis. The results of the predictions indicate that the performance of the Prophet model was superior to that of the ARIMA model. The model provided excellent forecasting for all types of individual products, effectively analyzing trends, seasonality, and holiday effects. Offline fresh produce sales showed a downward trend, attributable to the pandemic and the growth of e-commerce. With the exception of the Double Eleven shopping festival and New Year’s Day, most holidays positively affected vegetables, with the Spring Festival showing the most significant enhancement effect on flat-selling products. These findings offer valuable insights for enterprises regarding fresh reordering decisions and promotional timing, suggesting that the Prophet model can be efficiently utilized by fresh produce companies in management decision-making.