Logistics Network Sorting Center Cargo Volume Prediction and Personnel Scheduling Based on LSTM-ARIMA

Abstract

With the rapid development of the e-commerce industry, the demand for logistics has increased sharply, and the management efficiency issues of sorting centers have become increasingly prominent. This article provides daily and hourly cargo volume predictions, transportation route optimization, and staff scheduling for 57 centers over the next 30 days to meet user demands and transmission speed requirements. This paper proposes a comprehensive framework that combines the LSTM-ARIMA model for cargo volume prediction, calculates the cargo volume change rate based on the actual sorting center data through route comparison for correction, and optimizes staff arrangement using the linear regression model. This framework effectively addresses issues such as insufficient accuracy in cargo volume prediction, cargo volume deviations caused by changes in transportation routes, as well as excessively high total person-days and unbalanced hourly person-efficiency in employee scheduling.