An intelligent fuzzy TOPSIS framework using similarity and kernel distance measures for disc cutter selection in tunnel boring machines

The selection of appropriate disc cutters plays a crucial role in improving the efficiency, reliability, and sustainability of tunnel boring machine (TBM) operations. In mechanized excavation projects, improper cutter selection may lead to excessive tool wear, increased energy consumption, and reduced operational performance. Therefore, intelligent decision-support approaches are required to assist engineers in selecting optimal cutting tools under uncertain geological and operational conditions. This study presents a comparative evaluation of five proximity measures within the fuzzy TOPSIS method: four similarity measures for generalized fuzzy numbers and one Gaussian kernel distance for intuitionistic fuzzy numbers. Five computational approaches are investigated, including four fuzzy similarity methods and one kernel-based distance method. The kernel distance maps intuitionistic fuzzy numbers into a higher-dimensional feature space through a Gaussian kernel, where distances are bounded and better separated. Linguistic evaluations of performance criteria are converted into trapezoidal fuzzy numbers to construct the fuzzy decision matrix, enabling the integration of expert judgment and uncertainty in the decision-making process. The proposed framework is applied to evaluate multiple cutter alternatives based on relevant mechanical, operational, and performance-related criteria. The results indicate that although all evaluated methods consistently identify the same optimal disc cutter, noticeable variations occur in the ranking of the remaining alternatives. Among the examined techniques, Chen’s similarity method and the kernel-based approach demonstrate greater sensitivity and improved discrimination capability when addressing complex nonlinear relationships among decision criteria. Furthermore, the robustness of the proposed framework is verified through sensitivity analysis using an intuitionistic fuzzy TOPSIS model with kernel distance. The findings highlight the potential of advanced fuzzy decision-making models as intelligent tools for supporting equipment selection in mechanized tunneling. By facilitating more reliable cutter selection, the proposed approach contributes to improved operational efficiency, reduced tool consumption, and more sustainable tunneling practices within modern smart mining and underground construction systems.