Artificial neural network based modelling approach for strength prediction of concrete incorporating agricultural and construction wastes

Abstract

Construction debris and agricultural wastes are among the major environmental concerns in the world. Construction debris consumes about 28% of the nation’s landfill facilities. Over 254.5 million tons of rice husk is available for disposal every year. A large amount of these materials can be recycled and reused as aggregate and cement substitutes for general construction and pavements among other works. In this study, effort was made to develop artificial neural network (ANN) model for predicting the 28-day strength of concrete incorporating rice husk ash (RHA) and reclaimed asphalt pavement (RAP) as partial replacements of Portland cement and virgin aggregates respectively. The ANN model predicted the compressive and tensile splitting strengths with prediction error values of 0.648 and 0.072 MPa respectively. The model overpredicted the compressive strength (fc) on average by 0.123 MPa, whereas it underpredicted the tensile strength (fts) by 0.019 MPa. The predicted compressive and tensile strengths deviated on average by 2.088 and 2.905% respectively from experimental results. The results indicate that the ANN is an efficient model to be used as a tool for predicting the compressive and tensile strengths of concrete incorporating RHA and RAP.