PREDICTING MASONRY WALL COMPRESSIVE STRENGTH USING ARTIFICIAL NEURAL NETWORKS

Abstract

The application of artificial neural networks (ANNs) has significantly advanced the resolution of complex engineering problems in civil engineering, particularly in data analysis, design, and structural performance prediction. They play a crucial role in materials science by enabling the assessment of mechanical properties of construction materials and structural systems. This paper focuses on the use of ANNs for predicting the compressive strength of masonry walls, a key parameter for reliable structural design. The objective is to provide a systematic and critical review of relevant studies, highlighting the employed neural network models and their predictive accuracy compared to conventional approaches. The findings indicate that ANNs offer more reliable and accurate predictions of compressive strength, especially when experimental data are limited or heterogeneous, and support the development of auxiliary engineering tools and software applications.