Applications of Artificial Neural Networks in Modeling Compressive Strength of Concrete: A State of the Art Review

Abstract

Cement concrete is widely used throughout the world as a key construction material in civil engineering projects. Being a complex compound comprising of cement, sand, coarse aggregate, admixture and water, its compressive strength is a highly nonlinear function of its constituents, thereby making its modeling and prediction a difficult task. Nature inspired computational techniques, provide an efficient and easy approach for modeling complex, nonlinear or difficult to establish relationships between the independent and dependent variables. Artificial Neural Networks inspired by the learning ability of a human brain, can be regarded as an engineering counterpart of a biological neuron and its highly interconnected and parallel nature, gives them immense ability to learn from past examples capturing unknown relationships, making them a versatile tool for modeling the real world problems. The review paper is an attempt to provide an introduction to artificial neural networks, highlighting its applications as a computational tool for modeling complex functional relationships of various constituents influencing the compressive strength of concrete.