Interaction Between Biaxial Geogrid and Solid Waste Materials: Laboratory Study and Artificial Neural Network Model Development

Abstract

The growing industrialism has led to the generation of unmanageable waste, which is usually dumped in landfills. Utilizing these materials in reinforced earth structures will not only help in large-scale recycling of the materials but also restrict the overutilization of geo-materials. The aim of the present study is to assess the interaction of biaxial geogrid with two different solid waste materials namely, steel slag (SS) and construction and demolition waste (CDW) and compare its performance with the conventional backfill material, sand. To achieve this, direct shear tests and pullout tests were conducted at various normal stress levels ranging from 25 to 200 kPa. The direct shear results revealed that both steel slag and CDW exhibits higher shear strength compared to sand. Similarly, the pullout test results indicated that the pullout resistance of the geogrid is higher in steel slag and CDW than in sand. The pullout resistance factor (F∗) was evaluated to quantify the interaction of the geogrid with backfills. F∗ values were found to be greater than 1 at lower normal stress (25 kPa), and less than 1 at higher normal stress (200 kPa), indicating a stronger interaction at lower normal stresses. According to interaction coefficient ratio (ICR) values, the interaction of the geogrid with steel slag and CDW at 25 kPa normal stress was 38% and 33% higher, respectively, than that with sand. Furthermore, an artificial neural network (ANN) model was developed using data from the current study as well as data gathered from previous studies to predict the pullout resistance of geogrids in a wide range of geomaterials. An excellent prediction capability was observed with coefficient of determination (R 2) value more than 0.9.