Random Forest Regression Model Application for Prediction of China’s Railway Freight Volume

Yang Wang () and Xiaochun Lu ()
Additional contact information
Yang Wang: Beijing Jiaotong University
Xiaochun Lu: Beijing Jiaotong University

A chapter in Collaborative Logistics and Intermodality, 2021, pp 91-120 from Springer

Abstract: Abstract Purpose: The China Railway has an important impact on the transport of domestic energy products. The Chinese Prime Minister sees railway freight as a barometer of the Chinese economy; therefore, the study of China’s Railway freight is meaningful. During the past 5 years, from 2012 to 2016, China Railway freight volume continually declined, leading to a very serious situation. It is important to predict the volume of rail freight because it indicates the development of the Chinese economy. The prediction of China’s railway freight by a traditional regression model is not very effective because it is too sensitive to changes in statistical data. In particular, economic changes in China are now too large, resulting in significant changes in railway freight volume. In this chapter, we aim to use an machine learning model to predict China’s railway freight volume and attempt to determine whether the random forest regression model is more effective than the conventional forecasting method. Design/methodology/approach: In this chapter, random forest regression is applied to quantitatively predict railway freight volume. Six independent variables were collected from Jan 2001 to Dec 2016 in relation to China’s railway freight. After data analysis, a random forest regression model of China’s railway freight volume was built using the R language. To obtain the most suitable regression model, the random forest regression error is contrasted with the multiple linear regression model. The result shows that random forest regression model performed better than linear regression. Findings: The results in this study indicate the following: (1) the random forest regression model is able to predict railway freight volume using the selected variables. (2) By comparison of the variance and the normalized mean square error (NMSE) of different models, the best random forest regression model is obtained, and this model performs accurate prediction. (3) Compared with the multiple linear regression model, the random forest regression model exhibits superiority in prediction accuracy, robustness and fitness. (4) Although coal makes up the largest proportion of railway freight, refined oil production also has a large impact.

Keywords: Railway freight volume; Prediction model; Random forest regression (search for similar items in EconPapers)
Date: 2021
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-030-50958-3_6

Ordering information: This item can be ordered from
http://www.springer.com/9783030509583

DOI: 10.1007/978-3-030-50958-3_6

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().