Answers are grounded in this paper and cite their sources.
This analysis employs machine learning methods to forecast CBR values in expansive soil treated with hydrated-lime activated rice husk ash, highlighting the random forest model's efficacy.
The California bearing ratio (CBR) value is a fundamental property used to characterise the strength of the subgrade in road pavement design. CBR calculation in the laboratory is complex, time-consuming, costly, and requires careful execution. A machine learning approach is used to predict CBR of soil treated with hydrated-lime activated rice husk ash (HARHA). The prediction uses an algorithmic approach on A-7-6 expansive soil treated with HARHA, added from 0.1% to 12% in 0.1% increments. Using this approach, 121 distinct data sets were produced in the laboratory which are used to accomplish the stated goals. The data set includes six input parameters: HARHA, liquid-limit, plastic-limit, optimum moisture content, clayey activity, maximum dry density and one output: CBR value. Various models were used, including artificial neural networks (ANN), support vector machine (SVM), Gaussian process regression (GPR), and random forest (RF). The models’ performance was evaluated using statistical measures including coefficient of determination, mean absolute error, root mean square error, relative absolute error, and root relative squared error. The evaluation indicates that the RF model had superior predictive performance followed by ANN, SVM, and GPR model. Moreover, sensitivity analysis shows that maximum dry density is the most influential factor on CBR value.
No takes yet. Share an insight, caveat, or question.
Ahmad et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: