Prediction of compressive strength of concrete with rubber fibers using artificial neural networks

A non-recyclable material that enters the environment is used car tires. Research shows that used tires are made of materials that, due to their non-decomposition under normal conditions, cause pollution and damage to the environment. According to research, one method of removing these materials is to use rubber waste in concrete. Therefore, in this study, aggregate composites were replaced by waste rubber particles the compressive strength of concrete was estimated by artificial neural network using the input parameters water to cement ratio, superplasticizer additive and granulation weight composition. The results of this study were compared with other related research studies and confirmed the superiority and high accuracy of the artificial neural network obtained in this study. The a-20 engineering index of the neural network was determined to be one and the error of 99% of the data was less than 15%, indicating the appropriate approximation of the compressive strength of concrete containing waste rubber particles by the artificial neural network. In addition, the results of the sensitivity analysis using the Millen method indicated a 40% effect of the weight of the superplasticizer additive as a sensitive parameter in this type of concrete.