Artificial neural networks for predicting creep with an example application to structural masonry

Numerous creep models with limited accuracy have been developed within the last few decades to predict creep of concrete and masonry structures. The stochastic nature of creep deformation and its reliance on a large number of interdependent parameters (e.g., the brick and mortar types, the relative humidity, and the history and level of applied loading) make developing a single, general, and yet accurate mathematical model almost impossible. Artificial neural networks (ANNs) have been recently introduced as an efficient artificial intelligence modeling technique for applications incorporating a large number of variables. Artificial neural networks have proven successful in many instances where conventional mathematical modeling techniques were not as accurate or capable. Here, the potential use of ANNs in predicting creep is examined. A new ANN model is applied to the prediction of creep of structural masonry. The ANN developed is able to predict the creep performance with an excellent level of accuracy compared with that of conventional models.