A Study on Cargo Volume Prediction and Staff Scheduling Based on Support Vector Regression Model and Integer Planning

In pursuit of enhancing the management efficiency and reducing operational costs within the logistics network sorting centers, this research constructed a Particle Swarm Optimization (PSO)-Supported Vector Regression (SVR) model coupled with an Integer Programming staff scheduling model. Utilizing tools such as MATLAB, Excel, and Python, the study conducted processing and analysis of the cargo volume data, revealing the periodic trends of the cargo volume. After eliminating outliers, the optimized SVR model was employed for forecasting, and weights were established to adjust features and mitigate the impact of non-interactive sorting centers. The optimal staff allocation for each sorting center was determined through Integer Programming. This research provides an efficient set of strategies for cargo volume prediction and workforce scheduling, contributing to the optimization of cost-effectiveness in logistics management.