Optimization of Internet Platform Supply Chain Management based on Machine Learning Algorithm

The field of supply chain management is undergoing profound changes. Machine learning algorithms, especially Long Short-Term Memory Network (LSTM), show great potential in processing time series data and optimizing supply chain management. The purpose of this study is to explore the application of LSTM in the supply chain management of Internet platform, and compare it with Random Forest (RF) and Convolutional Neural Network (CNN). By comparing inventory turnover rate, order fulfillment cycle time and perfect order rate, the performance of different algorithms in supply chain management was evaluated. Research methods include the description of algorithm principle, the collection and pretreatment of supply chain data, model design, training and evaluation. The final data showed that the inventory turnover rate of LSTM was up to 97.6%; the order fulfillment cycle time of LSTM was kept at a low level, ranging from 3.3 days to 6.3 days; the perfect order rate was up to 87.8%. The conclusion of this study provides theoretical and practical guidance for the intelligentization of supply chain management in the future.