The utilization of expansive soils for the construction of landfill liners and subgrade facilities without stabilization leads to volume changes due to seasonal change between wet and dry. This necessitated the industrial- and agro-industrial-based waste materials reconstitution of expansive soils to fulfil sustainability requirements for the builtenvironment. In this research paper, multiple datasets were collected from mixes of bagasse ash (BA) and lime (Lm) blend reconstituted expensive soil and deployed in the training and validation interface of advanced machine learning (AML) techniques to predict the unconfined compressive strength (UCS) of the treated soil for their usein landfill liner and subgrade application. The relative importance values for each input parameter were evaluated, such as compaction parameters (MDD and OMC), plastic limit (PL), LL, Lm, and BA. The results of all developed models were observed and collected. The relations between calculated and predicted values show that the GP produced a parametric line of fit expression of y = 0.999 × with performance indices as MAE 14.80 kPa, MSE 400.7 kPa, RMSE 20.00 kPa, and R2 of 0.950, EPR produced a parametric line of fit expression of y = 0.992 × with performance indices as MAE 11.6 kPa, MSE 270.9 kPa, RMSE 16.50 kPa, and R2 of 0.963, and ANN produced a parametric line of fit expression of y = 0.997 × with performance indices as MAE 4.26 kPa, MSE 30.8 kPa, RMSE 5.55 kPa, and R2 of 0.996. The results show that the ANN outperforms the GP and the EPR having produced the least error values, the highest coefficient of determination (R2) and zero outliers beyond the ± 25% performance fit envelop and can be concluded that BA has a remarkable influence in the stabilization of expansive soils and its utilization.
