This research addresses the paramount issue of enhancing safety and health conditions in underground mines through the selection of optimal sensor technologies. A novel hybrid MEREC-CoCoSo system is proposed, integrating the strengths of the MEREC (Method for Eliciting Relative Weights) and Combined Compromise Solution (CoCoSo) methods. The study involves a three-stage framework: criteria and sensor discernment, criteria weight determination using MEREC, and sensor prioritization through the MEREC-CoCoSo framework. Fifteen criteria and ten sensors were identified, and a comprehensive analysis, including MEREC-based weight determination, led to the prioritization of “Ease of Installation” as the most critical criterion. Proximity sensors were identified as the optimal choice, followed by biometric sensors, gas sensors, and temperature and humidity sensors. To validate the effectiveness of the proposed MEREC-CoCoSo model, a rigorous comparison was conducted with established methods, including VIKOR, TOPSIS, TODIM, ELECTRE, COPRAS, EDAS, and TRUST. The comparison encompassed relevant metrics such as accuracy, sensitivity, and specificity, providing a comprehensive understanding of the proposed model’s performance in relation to other established methodologies. The outcomes of this comparative analysis consistently demonstrated the superiority of the MEREC-CoCoSo model in accurately selecting the best sensor for ensuring safety and health in underground mining. Notably, the proposed model exhibited higher accuracy rates, increased sensitivity, and improved specificity compared to alternative methods. These results affirm the robustness and reliability of the MEREC-CoCoSo model, establishing it as a state-of-the-art decision-making framework for sensor selection in underground mine safety. The inclusion of these actual results enhances the clarity and credibility of our research, providing valuable insights into the superior performance of the proposed model compared to existing methodologies. The main objective of this research is to develop a robust decision-making framework for optimal sensor selection in underground mines, with a focus on enhancing safety and health conditions. The study seeks to identify and prioritize critical criteria for sensor selection in the context of underground mine safety. The research strives to contribute to the mining industry by offering a structured and effective approach to sensor selection, prioritizing safety and health in underground mining operations.

The main goal of the present study is to investigate the capability of hybridizing the imperialist competitive algorithm (ICA) with an intelligent, robust, and data-driven technique named the light gradient boosting machine (LightGBM) to estimate the compressive strength of geo-polymer concrete (CSGCo). The hyper-parameters of the LightGBM algorithm have been optimized based on ICA and its accuracy improved. The obtained results from the proposed hybrid ICA-LightGBM are compared with the traditional LightGBM model as well as four different topologies of artificial neural networks (ANN) comprising a multi-layer perceptron neural network (MLP), radial basis function (RBF), generalized feed-forward neural network (GFFNN), and Bayesian regularized neural network (BRNN). The results of these models were compared based on three evaluation indices of R², RMSE, and VAF for providing an objective evaluation of the performance and capability of the predictive models. Concerning the outcomes, the ICA-LightGBM with the R² of (0.9871 and 0.9805), RMSE of (0.4703 and 1.3137), and VAF of (98.5773 and 98.0397) for training and testing phases, respectively, was a superior predictor to estimate the CSGCo compared to the LightGBM with the R² of (0.9488 and 0.9478), RMSE of (0.9532 and 2.1631), and VAF of (94.3613 and 94.5173); the MLP with the R² of (0.9067 and 0.8959), RMSE of (1.3093 and 3.3648), and VAF of (88.9888 and 84.9125); the RBF with the R² of (0.8694 and 0.8055), RMSE of (1.4703 and 5.0309), and VAF of (86.3122 and 66.1888); the BRNN with the R² of (0.9212 and 0.9107), RMSE of (1.1510 and 2.6569), and VAF of (91.4168 and 90.5854); and the GFFNN with the R² of (0.9144 and 0.8925), RMSE of (1.1525 and 2.9415), and VAF of (91.4092 and 88.9088). Hence, the proposed ICA-LightGBM algorithm can be efficiently used in anticipating the CSGCo.