Soil Classification and Correlation of Laboratory CBR Values with CBR Value from Subgrade DCP Test on Reconstruction Project of Mirit – Tambakmulyo Road Section, Kebumen, Central Java

Abstract

The California Bearing Ratio (CBR) value has become the standard for determining the bearing capacity of soil,
especially subgrade, in highway construction. Usually expressed as a percentage, the CBR figure can be obtained in the laboratory (CBR Lab) or the field through the Dynamic Cone Penetrometer (DCP) test. This study examines the relationship between the Laboratory CBR and the Field CBR from the results of the DCP test on 12 soil samples obtained at the research location on the Mirit - Tambakmulyo Road Section (Kebumen Regency, Central Java). The samples were then analyzed using a mathematical formula to obtain a regression equation and correlation value. Based on the AASHTO method, the soil at the research location is classified as A-6 soil type, namely clay soil, with a general assessment of moderate to poor subgrade. For the USCS method, the most dominant soil classification is CL: non-organic clay with low to moderate plasticity, gravelly clay, sandy clay, and dusty clay. The Laboratory CBR value was obtained from 5.59% to 7.90%. Two samples have a CBR Value <6.00%, namely Sample 10 and Sample 11. Meanwhile, for the CBR value from the DCP test in the field, the CBR range is 5.11% to 12.96%. There is only one sample with a CBR Value <6.00%, namely sample 5, which has a CBR figure of 5.11%. The CBR value of 6.00% is the minimum subgrade requirement in the Manual Desain Perkerasan 2024 released by the Indonesian Directorate General of Highways. From the results of the regression calculation and correlation of Field CBR to Laboratory CBR, the equation y = -0.0614x + 7.1492 is obtained with a correlation coefficient (r) = -0.1862. The correlation pattern obtained is a weak negative direct correlation. The coefficient of determination (R2) = 0.035 indicates that the accuracy of the regression model is quite good. This study proves that the correlation between Field CBR value and Laboratory CBR is weak and not statistically significant.