Both the problem of disposing of agricultural waste and the production of carbon dioxide are reduced by the use of rice husk ash and ordinary concrete. The evaluation of the compressive strength of rice husk ash concrete has, however, presented additional difficulties. In order to forecast the compressive strength of RHA concrete, this research suggests a brandnew hybrid artificial neural network model that has been optimised utilising a reptile search method with circle mapping. The suggested model was trained using 192 concrete data with 6 input parameters (age, cement, rice husk ash, super plasticizer, aggregate, and water) and its prediction performance was compared to that of five existing models. To assess the accuracy of the produced models' predictions, four statistical indicators were used. According to the performance evaluation, the suggested hybrid artificial neural network model had the best prediction accuracy in terms of R2, VAF, RMSE, and MAE (0.9709, 3.4489, 2.6451), all of which were highly satisfactory. The suggested model outperformed previously created models on the same data in terms of predicted accuracy. Age is the most crucial factor for predicting the compressive strength of RHA concrete, according to the results of the sensitivity analysis.