As an important large-scale temporary project of railway engineering, the carbon emission of track-laying base is the source of carbon emission that cannot be ignored in the materialization stage of railway engineering. The carbon emission factor method was used to establish the carbon emission measurement model in the life cycle of railway track-laying base. Then, the carbon emission characteristics of track-laying base were extracted as influencing factors, and the key influencing factors were identified by feature importance ranking. Finally, the interpretable machine learning model was used to visualize the contribution of key influencing factors to carbon emissions, and analyze the impact mechanism of key influencing factors on carbon emissions. The results show that the total life cycle carbon emission of the track-laying base is 4825.134~15122.059t. The carbon emission ratio of building materials in the production stage of rail laying base was the highest (72%-86%). According to the ranking results of the importance of influencing factors of track laying base, the five key influencing factors are identified as base area, foundation treatment method, road hardening method, mechanical track length and stock track length. The influence of key factors on carbon emissions was analyzed by SHAP summary diagram and dependency scatter diagram. The research results can provide theoretical basis for the research on carbon emission reduction of railway track-laying base.