Research on Cost Pricing Strategy Based on Random Forest Algorithm and Exponential Smoothing Model

This paper focuses on the application of machine learning and statistical analysis methods to item purchasing and pricing strategies, aiming to explore the relationship between total sales volume and markup pricing, and to propose effective purchasing and pricing models. Firstly, through the preprocessing and descriptive statistical analysis of vegetable sales data, the sales patterns of various types of vegetables are clarified. Then, the vegetable replenishment volume and price in the coming week were predicted using multiple linear regression model and exponential smoothing model to ensure the real-time and accuracy of the strategy. To further enhance the model's prediction performance, this paper introduces the random forest algorithm, which is able to capture the complex nonlinear relationships in vegetable sales volume and pricing strategies by constructing multiple decision trees and integrating the predictions. The random forest model not only enhances the prediction ability for seasonal variations and multidimensional feature data, but also reveals the key influence of each factor on sales through feature importance analysis. Ultimately, this study comprehensively analyzes the advantages and disadvantages of different models and proposes corresponding improvement suggestions, providing a scientific basis for the operation and management of fresh produce supermarkets.