Intelligent decision support in geotechnical engineering: An advanced model for risk prediction

Geotechnical engineering (GE) is observed as an essential factor in evaluating and managing the uncertainties inherent under subsurface conditions in construction projects. Soil stability, settlement, and foundation performance are all essential considerations in the effective and safe developing of geotechnical constructions. While the global model can forecast the emergence of these situations, the inherent complexity of soil behavior and environmental variables restricts the capacity to provide real-time risk forecasts. Existing models typically lack the yet-to-be defined accuracy and flexibility necessary for application in dynamic geotechnical contexts. The strategy intends to overcome this existing gap by developing an upgraded framework for improved risk prediction using the Enhanced Multi-Objective Optimization-Driven Logistic Regression (EMOO-DLR) method. According to the proposed method, including multi-objective optimization techniques into model risk prediction would result in increased predictive accuracy and flexibility. This research relied on three key geotechnical data parameters such as soil type, moisture content, and bearing capacity. All of these parameters are preprocessed with Min-Max Scaling to normalize their ranges and improve data consistency. Furthermore, Principal Component Analysis (PCA) is utilized to extract features and reduce dimensionality in order to identify and analyze the most important risk variables. The proposed model's effectiveness is evaluated using performance indicators such as R square (R²), Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), and Mean Absolute Error (MAE). The results indicate that the EMOO-DLR model adequately captures and forecasts the prevalent geotechnical hazards, with a reasonable reliability factor incorporated in its measurement. The EMOO-DLR framework provides a comprehensive, efficient, and intelligent decision-making tool for improving risk evaluation, resulting in safer measures, cost-optimized values, and successful risk mitigation in geotechnical engineering. The simple processing of complex, unpredictable settings improve the model for decision support in practice. The system has the ability to change the geotechnical risk management approach by providing engineers and project managers with an accurate, data-driven understanding of the environment, allowing for safer and more sustainable infrastructure construction.