Logistics sorting cargo volume prediction and personnel scheduling optimization based on ARIMA

ACKNOWLEDGEMENTS: This work was supported by ministry of education industry-university cooperative education project (Grant No.: 231106441092432) and special research project on teaching reform (Grant No.: 30120300100-23-yb-jgkt03)

Abstract

The development of the e-commerce industry is unstoppable, and logistics is an  important part of this industry, and the speed of industry development is largely constrained  by the efficiency of logistics sorting. The prediction of cargo volume and personnel  scheduling in logistics network sorting centers is one of the core issues in logistics  management. With the rapid development of the e-commerce industry, logistics network  sorting centers have to handle a large number of packages every day. Accurately predicting  cargo volume and arranging personnel scheduling reasonably are of great significance for  improving efficiency, reducing costs, and ensuring service quality. Regarding problem one: In  order to ensure the stability of the data and the accuracy of predictions, seasonal adjustments  and Mahalanobis distance are first used to handle missing and outliers in the data, thereby  improving the reliability of subsequent models and the accuracy of predictions. Next, the  SARIMS model will be used to predict the cargo volume of the sorting center for the next 30  days, and the ARIMA time series model will be used to predict the hourly cargo volume of  the sorting center. Finally, residual testing will be conducted to ensure that the results of the  model are more accurate and reliable. Regarding problem two: when predicting the cargo  volume for the next 30 days and each hour, it is necessary to first identify the total cargo  volume corresponding to the changing route. Then, the proportion of transported goods for  each route is calculated using a mathematical model, taking into account factors such as  transportation efficiency, cost, and route length, and comprehensively considering the  rational allocation of goods. Finally, based on the change rate of the total cargo volume, the  final prediction result is revised to make it consistent with reality. Regarding problem three:  categorize the problem as a typical integer linear programming problem and use heuristic  algorithms to solve it. In order to effectively schedule employees, it is necessary to establish a  planning model, define appropriate variables and parameters, and use an objective function  and constraints. I hope to achieve a balance of actual hourly efficiency per day while  minimizing the total number of person days as much as possible. Regarding problem four: In  order to solve the scheduling problem of specific sorting centers, it is necessary to deeply  optimize the factory production line settings. The core goal is used to balance cost effectiveness, rational resource allocation.