Title: Application of de-noising automatic coding method in freight volume prediction under intelligent logistics

Authors: Zheng Tang

Addresses: Business School, Zhejiang Fashion Institute of Technology, Ningbo 310058, Zhejiang, China

Abstract: With the advent of the information age, there appear many problems in cargo transportation, such as traffic jams, delayed information transmission, and low freight efficiency. The purpose of the study is to make freight transportation better adapt to the intelligent logistics and study the application of de-noising automatic coding networks based on deep learning in freight volume prediction. The de-noising auto-coding network and the stack de-noising auto-coding network are deeply discussed, and a freight volume prediction model based on the stack de-noising auto-coding network is constructed. The de-noising auto-coding prediction method is compared with the traditional prediction method and the deep-learning prediction method of the same kind. According to the comparative analysis, the average error of the stack de-noising auto-coding prediction method is 5.96% in 2019 and 2020, which is smaller than that of traditional prediction methods.

Keywords: intelligent logistics; deep learning; de-noising auto-coding; cargo volume prediction.

DOI: 10.1504/IJGUC.2022.10045599

International Journal of Grid and Utility Computing, 2022 Vol.13 No.1, pp.21 - 29

Received: 04 Mar 2021
Accepted: 07 Jul 2021

Published online: 11 Mar 2022 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article