Comprehensive Evaluation of Grouting Effectiveness Combining Qualitative On-Site Tests and Improved Fuzzy Integration with Entropy Weight Method: Case Study of a Mountain Tunnel

Abstract

Grouting can considerably reduce water leakage and rock collapse disasters in mountain tunnels. Evaluating postgrouting effects is crucial for ensuring grouting quality. However, because grouting is a multifaceted problem, the quantitative evaluation of grouting effects remains challenging. Moreover, single-aspect evaluations often yield inaccurate results. To address these problems, this study comprehensively evaluated grouting effectiveness by combining qualitative and quantitative approaches based on three aspects, namely, grouting integrity, continuity, and sturdiness. An improved fuzzy integration using an entropy weight method (EWM) is proposed for quantifying tunnel grouting. As another key approach, numerous on-site tests were conducted to visually assess grouting effects. The conventional analytic hierarchy process (AHP)–fuzzy method was used to validate the EWM–fuzzy method. The results of the improved integration were consistent with those of the AHP–fuzzy method and the on-site tests, demonstrating the reliability of the proposed integration. Moreover, the AHP–fuzzy process was analyzed by assigning all applicable scale values in the nine-scale method, which is typically ignored. This analysis revealed that grouting integrity and continuity significantly influenced scores negatively, contrasting with the positive influence of sturdiness. Importantly, higher scale values did not uniformly improve results; outcomes depended on entries in judgment and membership matrices. The results of the study provide a reference for evaluating and managing tunnel grouting projects and related multifaceted problems. The proposed EWM–fuzzy integration provides a more efficient approach compared with the AHP–fuzzy process, highlighting the potential for using advanced methods. Finally, the enhanced AHP–fuzzy method resolves the shortcomings in single-value assignment within the nine-scale method.