Optimization research on logistics network cargo volume prediction based on ARMA model

Abstract

With the rapid development of power grid logistics, predicting the cargo volume of logistics network sorting centers and reasonable personnel scheduling play a crucial role in their development. This article studies the problem of cargo volume prediction and personnel scheduling in logistics network sorting centers. A time series prediction model, logistics graph and network analysis model, and 01 integer programming model are established to obtain a more reasonable cargo volume prediction and personnel scheduling plan. To predict the daily and hourly cargo volume of 57 sorting centers for the next 30 days, an ARMA time series prediction model was established based on time series data. Firstly, the original sequence was preprocessed, and then the ARMA model was used to predict and analyze the daily and hourly cargo volume of each sorting center. Finally, a goodness of fit test was conducted on the model, and it was found that the model passed the test. When there are changes in the transportation routes between sorting centers in the next 30 days, this article establishes a graph and network model to analyze the transportation routes, analyze the changed routes, and finally dynamically adjust the predicted daily and hourly cargo volume for the next 30 days based on the built model. After calculation, the final dynamic adjustment result is obtained.