Author/s:
Younis M. Alshkane*, Alle A. Hussein*, Kamal A. Rashed*, Diyari A. Mohammed*
*Civil Engineering Department, College of Engineering, University of Sulaimani Iraq
younis.ali@univsul.edu.iq ; alle.hussein@univsul.edu.iq;Kamal.rashed@univsul.edu.iq;diyari.mohammed@univsul.edu.iq
Corresponding author: Younis M. Alshkane
DOI: https://doi.org/10.31972/iceit2024.061
Abstract
Developing reliable models to predict engineering parameters is an effective strategy to reduce the time needed for tests and the overall cost of the project. One of the extensively used parameters in highway construction projects is the California Bearing Ratio (CBR), which is of the significant parameter in road layers. In this study, the goal is to investigate the possibility of predicting CBR values of fine-grained soil from soil index properties for subgrade made from cohesive materials. The simple regression using one variable and multiple regression analysis using multi variable have been investigated to estimate CBR values. Three geotechnical parameters have been utilized in the analysis: Maximum Dry Density (MDD), Optimum Moisture Content (OMC), and Liquid Limit (LL). The results have been validated using Root Mean Square error (RMSE) and Coefficient of determination (R2). From the simple regression analysis (using only one variable), a useful model was developed to predict CBR value using OMC with R2 of 0.95 and RMSE of 0.38 %. From the multi-linear regression analysis, a model to predict CBR from LL, OMC and MDD with R2 of 0.94 and RMSE of 0.4% is also developed. In addition, simple models were developed to estimate the compaction characteristics from the LL index.
Keywords: CBR, Multiple Regression Analysis, LL, MDD, OMC.