Spare Parts Demand Forecasting and Inventory Optimization Based on Machine Learning Algorithms

Abstract

This study proposes a spare parts demand prediction model based on ensemble learning. By integrating the advantages of multiple algorithms, the model significantly improves prediction accuracy, with a Mean Absolute Percentage Error (MAPE) of 12.1%. Experimental results show that the model exhibits excellent performance in various scenarios, including high-value spare parts, seasonal spare parts, and long-term trend prediction, with a prediction accuracy rate of over 88%. Compared to traditional methods, the model not only has stronger non-linear pattern handling capabilities but also possesses good interpretability and scalability. The research results provide a new solution for the field of spare parts demand prediction and have important practical significance for improving inventory management efficiency.