SLRA and MLRA Based Correlations of Geotechnical Parameters of Fine Grained Soils

Abstract

Due to huge expenditure is involved in testing of soil by employing heavy machinery and costly equipment; it is necessitated to develop statistical correlations between various geotechnical parameters which can be used mostly in design of foundations and pavement constructions. Many techniques are de-veloped for carrying out regression analysis to establish the relationships be-tween response variable and explanatory variables as per the practice are con-cerned. Single linear regression analysis (SLRA) and multiple linear regression analysis (MLRA) are the modeling techniques commonly adopted to develop correlations between the required variables. The MLRA model creates a rela-tionship in the form of linear line that best approximates all the individual data points. The MLRA is a statistical technique that uses several explanatory varia-bles to predict the outcome of a response variable. This paper presents the rela-tionships established between geotechnical parameters of fine grained soils (% of soil passing 0.075mm BIS sieve > 50%) using SLRA and MLRA technique.The geotechnical parameters of fine grained soils such as California Bearing Ratio (CBR), Liquid limit (LL), Plastic limit (PL), Plasticity Index (PI), Opti-mum Moisture Content (OMC), Maximum Dry Density (MDD), Free swell in-dex (FSI) and % fine fraction (FF) are considered from the soil testing carried out on natural soil samples collected at different stretches of a road project from Khandi to Shadnagar in T elangana state. The SLRA and MLRA showed good fits between the geotechnical parameters of fine grained soils with regression coefficients showing above 0.80. The MLRA models developed can be utilized to estimate the CBR of fine grained soils whose % fines more than 50%.