Dynamic risk assessment for subway station excavation near existing buildings using spherical fuzzy numbers

Purpose Constructing subway stations near existing buildings is a complex engineering challenge. To identify high-risk factors and guide safe construction, a risk factor identification and safety evaluation system is needed. Design/methodology/approach Compared to intuitionistic fuzzy sets, Pythagorean fuzzy sets and similar approaches, spherical fuzzy sets provide a more nuanced representation of fuzzy information. They offer greater computational flexibility and precision in complex fuzzy decision-making scenarios. Building on this, the study constructs a full-cycle dynamic risk factor identification and safety evaluation system based on spherical fuzzy numbers, triangular fuzzy numbers and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method. The model uses expert coefficients to measure the weight of different expert judgments and the coefficient of variation to assess the consistency of expert judgments. By combining subjective expert judgments with objective on-site data, the evaluation system is built through three steps: system establishment, data processing and risk factor identification. Findings The applicability of the model was validated through the construction of the Xueshan Road Station of the Jinan Metro. An evaluation system comprising four criteria and 14 risk factors was established, covering geological conditions, construction monitoring, buildings and technical and management variables. A comprehensive evaluation of six construction stages was conducted, and the identified high-risk factors were consistent with the on-site conditions. Sensitivity analysis further verified the accuracy and stability of the model. Originality/value The model is applicable to various engineering challenges, providing a robust basis for early warning of construction risks.