@article{BAO2026, 
author = {Xueying BAO and Haitao REN and Beisheng LIU and Hui LI and Zilong LI},
title = {Research on carbon emission calculation and influencing factors of railway track-laying bases},
year = {2026},
journal = {Journal of Chongqing University},
volume = {49},
number = {4},
pages = {50-62},
keywords = {railway track-laying base, carbon emissions, random forest regression, interpretable machine learning, influencing factors},
url = {https://www.sciopen.com/article/10.11835/j.issn.1000-582X.2024.285},
doi = {10.11835/j.issn.1000-582X.2024.285},
abstract = {As a major temporary facility in railway construction, the carbon emissions generated by track-laying bases constitute a significant source of embodied carbon during the materialization stage. In this study, a carbon emission measurement model for the life cycle of railway track-laying bases is established using the carbon emission factor method. The characteristic emission contributors are then extracted as potential influencing factors, and key factors are identified by feature-importance ranking. Furthermore, an interpretable machine learning model is used to visualize the contribution of these key factors and to analyze their impact mechanisms on carbon emissions. The results show that the total life cycle carbon emissions of a track-laying base range from 4825.134 t to 15122.059 t. Carbon emissions from building materials in the production stage account for the largest share (72% to 86%). According to the importance ranking, the five key influencing factors are base area, foundation treatment method, road hardening method, mechanical track length, and stock track length. The influence of these key factors on carbon emissions are further analyzed by SHAP (Shapley additive explanations) summary plots and dependency scatter plots. The findings provide a theoretical basis for promoting carbon reduction strategies in the construction and operation of railway track-laying bases.}
}