Modeling and Statistical Evaluations of Unconfined Compressive Strength and Compression Index of the Clay Soils at Various Ranges of Liquid Limit

A significant stage in geotechnical engineering is to establish geotechnical properties of soil models to predict the most important soil properties such as unconfined compressive strength (UCS) and compression index (Cc) because they are the main parameters in the state design of the footings, pavements, or stability assessment of existing structures or slopes. This study is focused on developing models to predict the compressive strength and Cc for the clay soils as a function of Atterberg limits, natural moisture content, dry density, void ratio, and fine content (passing ≤ 0.075 mm). The UCS of the soils ranged from 24 to 340 kPa and was quite accurately quantified using the laboratory-tested data and data collected from published research studies. The Cc of the soils varied between 0.0878 to 0.8317, which was also correlated as a function of easy measurable soil properties such as Atterberg limits, natural moisture content, density, void ratio, and fine contents (percentage passing sieve number 200). A wide experimental test results (a total of 253 tested soils) were combined with more than 350 data collected from different academic research studies, and total data were statistically analyzed and modeled. In the modeling process, the most relevant parameters affecting the compressive strength and Cc of soils incorporation ratio (6–41 % of moisture content), plasticity index (7–72 %), dry density (11–19 kN/m3), and fine content (0–100 %). According to the correlation determination, mean absolute error, and the root mean square error, the compressive strength and Cc of soil can be well predicted in terms of liquid limit, plasticity index, moisture content, dry density, void ratio, and percentage passing sieve No. 200 (75 µm) using linear simulation techniques. The sensitivity investigation concludes that the dry density and moisture content are the most important parameters for the prediction of the compressive strength and Cc, respectively, with the training data set.