Multi-objective optimization in tunnel line alignment under uncertainty

• Main Authors: Guo, Kai, Zhang, Limao
• Other Authors: School of Civil and Environmental Engineering
• Format: Article
• Language: English
• Published: 2022
• Subjects: Engineering::Civil engineering; Multi-Objective Optimization; Genetic Algorithm
• Online Access: https://hdl.handle.net/10356/160759
• Institution: Nanyang Technological University

New metro lines are being constructed to meet the rapid development of cities. Multi-objectives are required for the tunnel line alignment. A genetic algorithm-based approach is proposed in this research with the aim of realizing the multi-objective optimization for complex construction projects under uncertainty. Different decision variables and multi-objectives can be identified, and a Pareto front of the optimal trade-off solutions can be obtained by using a genetic algorithm to perform the optimization. A final optimal solution, the one that is nearest to the ideal solution, is determined as the suggestion for the decision-making. To test the applicability of the proposed approach, a tunnel line alignment project is studied. The radius and the depth of the tunnel line are analyzed as the decision variables, and the investment, headway, and comfort are determined as the objectives. The optimization is performed by using the non-dominated sorting genetic algorithm (NSGA-II). A particular optimal solution with the objectives of an investment of 559.81 million CNY, a headway of 5.56 min, and a comfort magnitude of 0.8646 is determined as the selected solution. To study the extendibility of the proposed approach, another variable, the fleet size, and more constraints are considered in the tunnel line alignment project. Three different scenarios are analyzed for optimization through the proposed approach. Results demonstrate that the proposed approach can realize optimization under more constraints according to priorities, which gives the project owner great flexibility to achieve particular project aims. The novelty of this research lies in its capabilities of: (1) assessing more than two decision variables and objectives and evaluating the combined effect of them in tunnel alignment projects; (2) generating the optimal solutions for the tunnel line alignment projects, which can help for improved decision-making when conflicting objectives exist; (3) providing great flexibility and extendibility with the proposed approach for achieving priorities in tunnel alignment projects.