Research on cargo volume prediction and personnel arrangement based on logistics network sorting centers

Abstract

In today's increasingly competitive e-commerce logistics industry, various levels of sorting centers on the logistics chain serve as the core nodes and capillaries of the logistics network, bearing the increasingly heavy sorting and transportation functions. Therefore, using the cargo volume of sorting centers to statistically predict future cargo volume and based on this, plan personnel arrangements for each station. This significantly improves the operational efficiency and economic benefits of each station, thereby more reasonably allocating resources, improving service quality, reducing labor waste, improving employee job satisfaction, and achieving the optimization and sustainable development of the logistics industry. This article uses the cargo volume statistics data of 57 sorting centers in a logistics network to analyze the temporal changes of cargo volume at each station on different dates and different time periods on the same day. Starting from the flow of goods within the network and considering the mutual influence of goods volume between stations, machine learning is used to extract the inherent changes in various sorting centers. Regression algorithms based on time series are used to predict the future goods volume for a period. Then, based on the predicted goods volume at each station, a linear optimization is established by considering the constraints of formal and temporary workers to solve the scheduling problem, to achieve a relatively balanced number of personnel for different shifts and minimize the total number of workers and labor costs.