Date of Award


Publication Type


Degree Name



Civil and Environmental Engineering


Artificial neural network, Data-based modelling, Machine learning, Multigene genetic programming, Reinforced concrete, Shear friction


F. Gherib


S. Cheng




Shear friction theory describes the mechanisms by which shear forces are transferred across concrete-to-concrete interfaces. Shear transfer across a plane involves a complex interaction of several phenomena, such as concrete surface condition and cohesion, concrete strength, and steel reinforcement strength and reinforcement ratio. Existing empirical equations for shear friction have been developed using limited sample data sets;thus, their accuracy is limited to the range covered by the data. In order to overcome this limitation, the present thesis proposes two machine learning models drawn from an extensive database to predict the shear friction capacity in reinforced concrete (RC) with a high degree of accuracy. The first method uses the artificial neural network (ANN)framework, which is shown to be effective in analyzing nonlinear data. The second model,which is based on genetic programming (GP), for generating an analytical equation for estimating RC members' interface shear friction capacity. To achieve optimal accuracy, the hyper parameters of each model have been appropriately tuned. Several statistical metrics were used to evaluate the performance of the proposed models. A comparison with the ACI318 (2019), AASHTO (2012), CSA A23.3 (2019), and an empirical equation developed by Kahn and Mitchel (2002) is conducted. It is shown that the two proposed models provided superior accuracy in terms of correlation and error between predicted and actual values. Additionally, sensitivity analyses were carried out to identify the most significant factors that affected the RC shear friction capacity. In both models, the concrete compressive strength was found to be the factor having the greatest influence on the RC shear friction.

Available for download on Sunday, June 02, 2024