Modelling Marshall Stability of fiber reinforced asphalt mixtures with ANFIS

Abstract

In this study, an Adaptive Neural Fuzzy Inference System (ANFIS) model for predicting the Marshall Stability (MS) of basalt fiber reinforced asphalt concrete mixtures and various mix proportions has been developed. Experimental details were used to construct the model. The amounts of bitumen (%), Fiber (Basalt) Ratio (%) were used as input variables and Marshall Stability (kg) values were used as output variables. Statistical equations were used to evaluate the Developed ANFIS model. Results showed that developed ANFIS model has strong potential to predict Marshall Stability of asphalt concrete using related inputs in a short time. Also, the Marshall Stability of Fiber-Reinforced asphalt concrete and various mix proportions can be found without performing any experiments.