Prediction of compressive strength of recycled aggregate concrete using artificial neural network and cuckoo search method

Construction waste plays a major role in today's world and it has a huge environmental impact. This can be eliminated by incorporating the waste material as recycled aggregate concrete (RAC) in construction and other projects which has behest for RAC. Compared to natural aggregates, recycled aggregates have a significantly different composition, making it difficult to predict their performance. This study shows how artificial neural network (ANN) and cuckoo search method (CSM) have been used to predict the compressive strength of RAC. The CSM and the ANN have been subjected to training and testing using 121 available data collected from published literature. The ANN and CSM models are developed with the help of six input parameters, such as water-cement ratio, replacement percentage of recycled aggregate concrete, natural coarse aggregate, fine aggregate, recycled coarse aggregate, water absorption. MATLAB software is used to model the artificial neural network and cuckoo search method to predict 28 days compressive strength of RAC. The results of mean squared error (MSE) and coefficient of regression (r2) show that both methods are a useful tool to predict the compressive strength. However, the ANN model reveals better accuracy. (c) 2021 Elsevier Ltd. All rights reserved. Selection and peer-review under responsibility of the scientific committee of the 3rd International Conference on Materials, Manufacturing and Modelling.