Regression Analysis for Predicting Soil Strength in Bangladesh

This study focuses on establishing a robust relationship between Standard Penetration Test-N values (SPT-N), geotechnical parameters and unconfined compressive strength (qu) using regression analysis. The proposed offers reliable method for estimating qu based SPT-N values. A comprehensive dataset comprising approximately 200 soil samples collected from various boreholes across Dhaka city was utilized. Multiple Linear Regression (MLR), Rando-forest (RFR) AdaBoost techniques were employed to develop unified correlation model. Evaluation metrics including R-squared (R2), Mean Absolute Error (MAE) Root Squared (RMSE), along with Trend-behavior Analysis assess compare the performances of models. Additionally, sensitivity analysis carried out selected model in order importance each parameter used predict qu. Finally, compared against existing empirical models that published previous studies. In terms evaluation Analysis, results showed RFR performed better than others. outperformed others, demonstrating highest R2 score, smallest RMSE MAE lower residuals Hence, provides accurate predictions clayey Bangladesh. Its implementation could ensure more efficient designs, specifically adjusted geological conditions region. While studies have established regional equations parts world, our uniquely has incorporated Plasticity Index (PI) as predictor is calibrated characteristics city. findings this highlight effectiveness applicability predicting Dhaka's properties, thus introducing valuable tool enhancing accuracy assessments design KEYWORDS: Unconfined strength, penetration test-N values, index, linear regression, Random-forest metrics, analysis, Sensitivity