Regression Model for Predicting Soaked CBR from UCC

Abstract

ABSTRACT. Flexible pavement is a multi-layered structure with a subgrade layer acting as the pavement structure’s foundation. The performance and strength of soil for its use as a subgrade are ascertained by its California Bearing Ratio (CBR) value. CBR test is a technically extensive and labor-intensive mechanism that could cause delays in carrying out construction projects, which would raise the construction cost. Therefore, highway engineers need to design a predictive model for quick assessment of the CBR of subgrade soil. In this research, eight specimens of disturbed soil were obtained from Rawalpindi Division, Pakistan. All soil samples were subjected to laboratory testing and categorized according to the AASHTO soil Classification System. The specimens were subjected to soaked CBR, and unconfined compression tests based on Maximum Dry Density (MDD) and Optimum Moisture Content (OMC) values, acquired from the Modified Proctor Compaction test. An empirical correlation between soaked CBR and unconfined compressive strength test is developed by the Suitable Trend-Line Method in Microsoft excel. From the single linear regression model, the value of the coefficient of correlation is found (0.98) indicating a very good correlation between soaked CBR and UCC strength test.

Keywords: Regression Model, Correlation Coefficient, California Bearing Ratio (CBR), UCC strength.