Khalid Saqer Alotaibi and Khaled Galal
Khalid Saqer Alotaibi, Assistant Professor, Department of Civil and Construction Engineering, College of Engineering, Imam Abdulrahman bin Faisal University, Dammam, Saudi Arabia, kalotaibi@iau.edu.sa
Khaled Galal,Professor, Department of Building, Civil and Environmental Engineering, Concordia University, 1515 St. Catherine West, Montreal, QC, Canada, khaled.galal@concordia.ca
ABSTRACT
Masonry is the oldest and most widely used building material in historical and modern structures. However, predicting the axial compressive strength of masonry works is still a challenge for the
scientific community considering its diversity and complexity, which does not depend only on the mechanical properties of materials, but also depends on the geometrical characteristics for each
component. This paper aims to predict the compression capacity of clay brick masonry prisms by using a machine learning technique. Artificial neural networks (ANNs) have attracted a lot of attention in recent years in many scientific fields for their capability to develop prediction models by training data. The ability of artificial neural networks (ANN) to predict the compressive strength of clay brick masonry prisms against experimental tests available in published journals will be examined. The proposed ANN model considered the essential parameters that would affect
masonry strength, including the strength and the type of clay brick unit, mortar type, and slenderness ratio. After the ANN model was trained with the available 326 datasets, the proposed ANN model demonstrated a good ability to predict the compressive strength of clay brick masonry prisms. The model achieves a high value of the coefficient of determination around 0.91 which is better than the empirical models reported in published literature.
KEYWORDS: brickwork, machine learning, prediction, strength
