Journal of Engineering, Project, and Production Management, 2026, 16(5), 2025-355
Predicting Cost-Benefit Trade-Off Path in Urban Rail Transit Construction Schemes Using a LightGBM Model
Associate Professor, School of Rail Transportation, Hunan Railway Professional Technology College, ZhuZhou, 412001, China, E-mail: hnrpcdan@sina.com (corresponding author).
Project Management
Received December 23, 2025; revised February 25, 2026; accepted March 9, 2026
Available online June 17, 2026
Abstract: Urban rail transit projects are characterized by large investment scales, long industrial chains, and diversified benefits, and their cost-benefit trade-offs directly affect project sustainability. Based on cost-benefit theory and machine learning methods, this paper first clarifies the research questions of quantifying cost-benefit trade-off relationships and predicting dynamic trade-off paths for urban rail transit projects, and defines the decision-making connotation of the trade-off path as the dynamic change trend of cost-benefit balance under the joint action of multi-dimensional influencing factors. Then, it constructs a cost-benefit trade-off prediction framework integrated with the LightGBM model. Using panel data of 32 newly-built rail transit projects in 15 first-and second-tier cities (Chengdu, Wuhan, Nanjing, Beijing, Shanghai, Guangzhou, Shenzhen, etc.) in China from 2018 to 2023, including detailed construction cost accounting, operational benefit statistics and urban macroeconomic data, 12 characteristic indicators were identified from three dimensions: construction costs, financial benefits, and national economic benefits. Model parameters were optimized through grid search to achieve an accurate prediction of cost-benefit trade-off paths. The results show that the LightGBM model's coefficient of determination (R²) is 0.892, and its Mean Squared Error (MSE) is 0.037, both of which are significantly better than those of traditional regression models. The added value benefits passenger flow density, unit cost, and land are the core factors affecting the trade-off path, with their weight accounting for more than 45%; based on the model prediction, a trade-off optimization strategy of “gradient cost control and diversified benefit improvement” is proposed, which is applicable to rail transit projects in different tier cities in China under the current urban development stage and can provide a scientific basis for decision-making in rail transit projects.
Keywords: Urban rail transit, lightgbm model, cost-benefit trade-off, path prediction, parameter optimization.
Copyright © Journal of Engineering, Project, and Production Management (EPPM-Journal).
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Requests for reprints and permissions at eppm.journal@gmail.com.
Citation: Hou, D. (2026). Predicting Cost-Benefit Trade-Off Path in Urban Rail Transit Construction Schemes Using a LightGBM Model. Journal of Engineering, Project, and Production Management, 16(5), 2025-355.
