Building a green and ecological railway transportation system that incorporates the “Dual-Carbon” Strategy is a central focus and challenge in current industry research. In the western mountainous regions with complex engineering geological conditions and fragile ecosystems, it is particularly important to explore the optimal railway route under the framework of the “Dual-Carbon” strategy. By analyzing the characteristics of the geographic environment of the western mountainous areas and the trend of low-carbon railroad construction, and referring to the relevant principles of railroad line selection, the method of quantifying the carbon emissions during the construction phase of the railroad and the carbon sequestration capacity of the land lost as a result of the railroad project’s land occupation is proposed by selecting 23 indicators from the five aspects of engineering adaptability, low-carbon adaptability, economic adaptability, environmental adaptability, and social adaptability as the entry point. The evaluation index system for the preferred railroad line scheme in the western mountainous area, developed in the context of the “Dual-Carbon” strategy, incorporates both subjective and objective weighting of the evaluation indexes using the G1 and CRITIC methods, respectively. Additionally, symmetric cross entropy sorting method is introduced to determine the combined weights of the indexes, effectively balancing the variability and conflicts between the indexes while considering subjective experience and the authenticity of the objective data.
Finally, to address issues in the traditional TOPSIS method—such as the irrationality of directly combining benefit-type and cost-type indicators and the significant difference in their magnitudes—a railroad line scheme preference model based on osculating value improvement TOPSIS approach is developed and validated through engineering case studies. The calculation results indicate that the corrected osculating values of the four line schemes in the region are 0, 1.623, 0.700, and 2.000, respectively. The scheme with the highest corrected osculating value is selected as the optimal choice. The preferred results based on the model align with the actual engineering scheme, verifying the model’s reliability. This approach holds significant value for application in the selection of low-carbon railroad lines in western mountainous areas.