LOGISTICS SORTING OPTIMIZATION BASED ON MACHINE LEARNING

The aim of this research is to achieve accurate prediction of the cargo volume in a logistics sorting center and efficient allocation of personnel. To this end, various machine learning algorithms such as Random Forest, Multilayer Perceptron Regressor (MLPRegressor), Support Vector Machine, and Multiple Linear Regression are comprehensively applied to deeply mine the historical cargo volume data of the sorting center, and features such as lagged cargo volume and moving average cargo volume are constructed. The grid search method is used for parameter optimization, and the optimal model is selected for future cargo volume prediction. On this basis, based on the prediction results, methods such as Linear Programming (LP), graph theory models, and queuing theory are utilized to construct a model for the optimal allocation of personnel. Taking into account factors such as employees' work efficiency and attendance rules, the rational deployment of personnel in the sorting center is realized. The experimental results show that the constructed cargo volume prediction model has a high accuracy, and the optimized personnel allocation scheme is reasonable and feasible. The conclusion indicates that the multi-algorithm fusion strategy can effectively improve the accuracy of cargo volume prediction and the efficiency of personnel allocation in the logistics sorting center, providing strong support for the intelligent management of the sorting center.