A Multi-Objective Optimization Approach for Tunnel Construction Methods Incorporating Uncertainty Quantification and Probabilistic Risk Assessment: A Case Study of the Chongqing Central Park Metro Station

Traditional methods for selecting tunnel construction techniques often overlook the uncertainty of surrounding rock parameters and the probability of support failure, leading to limitations in decision-making under complex geological conditions. This paper presents a multi-objective optimization approach for comparing tunnel construction methods, utilizing statistical moment analysis, dimensionality reduction techniques, and conditional failure probability analysis. Using the results of the stratigraphic structural numerical model, the impact of surrounding rock parameter uncertainty on support structure deformation is quantified through the statistical moment method. Additionally, by combining stress-deformation linear fitting, a dynamic evolution model for the probability of support failure is developed. Using Chongqing Rail Transit Line 10's Central Park East Station as a case study, this analysis compares three tunnel construction methods, the double-wall pilot tunnel method, the nine-section segmental temporary support method, and the nine-segment rock column method-based on multi-objective optimization criteria, including construction period, cost, and reliability. The results indicate that the nine-segment rock column method demonstrates optimal overall benefits by ensuring both the stability of the surrounding rock (with a maximum arch deformation of 14.3 mm) and the reliability of the support (with a failure probability of less than 0.8%), while simultaneously reducing the construction period by 20.8% and lowering costs by 25% to 30%. On-site monitoring confirmed the effectiveness of the rock pillar self-supporting system in stress redistribution and deformation control. The study integrates uncertainty analysis and probabilistic risk assessment into tunnel construction method decision-making, providing a scientific foundation for selecting construction methods in tunnel engineering. It demonstrates theoretical innovation and offers broad engineering applicability.