Optimization of Vegetable Replenishment and Pricing Strategies Based on Data Mining

In the digital era, vegetable sales forecasting and strategy optimization are important for superstores. Therefore, we developed an innovative integrated method combining Long Short-Term Memory Network (LSTM), Random Forest (RF), and Genetic Algorithm (GA) to optimize the pricing and replenishment decisions of vegetables on sale. First, the LSTM model was used to accurately predict the future vegetable sales trend to provide a scientific basis for decision-making. Next, the key factors in the sales data were analyzed using RF to identify the variables that affect the sales volume. Finally, GA was used to optimize the price and replenishment strategy based on the prediction results to maximize profits and balance market demand. This integrated approach improves the accuracy of sales prediction and optimizes the replenishment and pricing decision-making process through intelligent algorithms. Then, the operational efficiency of superstores improves significantly. The developed effectively responds to market changes, supports superstore management, and demonstrates its potential for applications in the retail industry.