Soft Computing Modeling Including Artificial Neural Network, Non-linear, and Linear Regression Models to Predict the Compressive Strength of Sustainable Mortar Modified with Palm Oil Fuel Ash

Abstract

Producing sustainable concrete and mortar is the idea that have been investigated by many researchers in the world through using waste materials in the mortar or concrete compositions to reduce the thread on the environment. In order to predict the compressive strength of mortar, this article proposes statistical models utilising linear regression (LR), nonlinear regression (NLR), and artificial neural network (ANN) based on experimental data collected from prior research in the field. The pozzolanic material used in mortar is agricultural waste, specifically Palm Oil Fuel Ash (POFA). In order to choose the most efficient model, the proposed models were evaluated using several statistical parameters. When compared to alternative models (Linear regression, nonlinear regression, and ANN), the one developed using ANN proved to be the most efficient in terms of approach, giving lower values for root mean square error (RMSE) and mean absolute error (MAE) which were 5.11, and 4.175 respectively. The suggested ANN model performed well according to the scatter index (SI), and the coefficient of determination value (R2) value was 34% more than the R2 in the LR model and 23% greater in the NLR model.