Strength and stiffness prediction models of expansive clays blended with sawdust ash

Abstract

Strength and stiffness characteristics are major concern for selecting any geomaterial. However, laboratory testing of these characteristics is time associated, laborious, and high cost. So, there is a need of intelligence tools to estimate the strength and stiffness of geomaterial. The impact of sawdust ash on the stiffness and strength properties of combined expanding clays is discussed in this research. The combined expansive clays underwent tests for California bearing ratio (CBR), unconfined compressive strength (UCS), optimal moisture content, maximum dry density, plasticity characteristics (liquid limit and plastic limit), and differential free swell (DFSI). According to test results, adding more sawdust to the blended clays improves their performance. This study also investigates the artificial neural network (ANN) model that considers six input variables to forecast the CBR and UCS of blended clays. The findings demonstrate that the ANN model performs more accurately for the CBR and UCS models. This clever method may help manage the under- or overestimation of additive dosage and reduce project costs.