A new form of deep learning in smart logistics with IoT environment

The purpose of this study is to introduce a method to fill the gap in urban freight volume prediction in the Internet of Things by utilising deep learning for meeting smart logistics requirements in China. To meet this objective, researchers reviewed relevant literature to demonstrate that deep learning is appropriate for logistics use. Chinese urban transportation system has been selected as the study object with 9 years of data to examine the deep learning application in Chinese urban transportation in the IoT environment. Researchers introduce the deep learning method to predict urban road freight volume and design the prediction model architectures of two new DL algorithms. Model training and parameter adjustment are also tricky in the research of this article. It is necessary to understand the role of each parameter in the algorithm and flexibly use the relevant DL framework in Python to obtain an ideal model through multi-fold cross-validation and multiple trials. The final results show that the transportation freight volume prediction in the Internet of Things by utilising deep learning has excellent prediction effects to meet smart logistics requirements compared with traditional methods.