Characterization of Soft Soil Strength Properties and Settlement Behavior

Abstract

Background: Soft soils present significant challenges in civil engineering practice due to their high compressibility, low shear strength, and susceptibility to excessive settlement under structural loading. Accurate characterization and prediction of settlement behavior are therefore essential for safe and economical foundation design, particularly in regions underlain by soft cohesive soils. Methodology: In this study, soil samples were obtained from three locations, with two trial pits per location, at depths of 0.5 m and 1.0 m, using a systematic site investigation approach. Laboratory testing was conducted in accordance with relevant ASTM standards to determine key geotechnical parameters, including moisture content (MC), optimum moisture content (OMC), cohesion (C), internal friction angle (∅), maximum dry density (MDD), unconfined compressive strength (UCS), shear strength (SS), void ratio (e₀), hydraulic conductivity (K), and consolidation characteristics. Principal Component Analysis (PCA) was applied to identify dominant soil parameters controlling behavior, while multivariate linear regression models were developed to predict total settlement, unconfined compressive strength, shear strength, and permeability Results: The results indicate pronounced spatial variability in soil properties across locations and depths, reflecting heterogeneity typical of soft soils. Correlation analysis, regression modeling, and PCA consistently identified moisture content, void ratio, permeability, and compressibility-related parameters as the primary controls on settlement behavior. The regression models exhibited high predictive accuracy, with coefficients of determination (R²) ranging from 97.0% to 99.9%, confirming strong quantitative relationships between soil properties and measured responses. PCA revealed that the first principal components are dominated by void ratio and hydraulic conductivity, representing a compressibility–permeability mechanism, while strength- and density-related parameters govern secondary components, reflecting soil compactness and structural resistance. Conclusion: This study demonstrates that integrating laboratory-based soil characterization with statistical tools provides both predictive and interpretative insights into soft soil behavior. While PCA–regression techniques are established in geotechnical analysis, their application to site-specific soft soils enables a clearer identification of critical parameters influencing settlement. The developed regression models and PCA framework offer practical decision-support tools for foundation design and settlement risk assessment in soft soil environments, while highlighting the need for broader datasets to enhance regional generalization.