Compressive Strength Prediction Using Machine Learning Techniques for Recycled E-waste Concrete

Conventional concrete is the most used building material because of its many inconsistent attributes. Engineers mostly use compressive strength (CS) as one of the most important parameters when building concrete constructions. Typically, this characteristic is established by costly laboratory experiments, wasting money, materials, and time. On the other hand, E-waste, such as Printed Circuit Boards (PCB), is being generated in alarmingly large quantities. As a result, it is critical to plan ahead of time. E-waste, such as non-metals, such as a piece of PCB, can be recovered and utilized as a concrete fixing. As a result, we could partially substitute the material to get the desired concrete characteristics. According to their findings, E-waste is likely to be used as an aggregate alternative. More use of this tends to reduce the use of conventional aggregates in concrete. Thus, it’s critical to look into alternative aggregates. Artificial intelligence, on the other hand, and its many uses are examples of creative technology that has been successfully applied to scientific applications. Support Vector Machine (SVM) and Artificial Neural Network (ANN) models are commonly employed to resolve engineering problems. The research scope of the study is to examine recycled E-waste concrete for construction using three models: Random Forest (RF), SVM, and ANNs. Pre-processing data, statistical procedures, and data visualization techniques are used to obtain a better knowledge of the database. Finally, the acquired results demonstrate great efficiency compared to previous works that also captured the cohesion.