Neuro-Fuzzy Investigation of Natural Fiber-Reinforced Concrete

Abstract

Construction industries often utilize the conventional cement-based concrete, but it has a significant drawback of being brittle due to its low ability to absorb energy and low tensile strength. In tropical regions, Oil palm empty fruit bunch (OPEFB) fibre is an agricultural-based fiber material which has found wide applications in concrete reinforcement. Fibre-reinforced concrete has demonstrated significant advantages in improving the inherent brittleness of traditional concrete due to its superior tensile force/strain capabilities, and reduced occurrence of cracks. This study proposes a neuro-fuzzy system based predictive model for the compressive strength of OPEFB-reinforced concrete with grid partitioning (GP) clustering. It was revealed that at 7 days, 0.4% OPEFB-fiber in the concrete yielded the maximum compressive strength. At the training phase of the GP-ANFIS model, it produced root mean square error (RMSE), mean absolute deviation (MAD), mean absolute percentage error (MAPE), and R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>-values of 0.513, 0.511, 2.674 and 0.967. The outcome of the study revealed that the neuro-fuzzy model enhances the predictive accuracy and reliability in concrete strength estimation which contributes to advancing knowledge in the field of sustainable construction materials.