Split Tensile Strength prediction of Coir Fiber Reinforced Concrete using Machine Learning Technique

Coir fiber reinforced concrete (CFRC) is widely used nowadays due to its easily availability, economical and enhancement in strength. Many past literatures show strength enhancement of concrete when coir fiber is added to it. CFRC enhances the tensile properties of concrete. The mechanical behavior of CFRC varies with various parameters such as, fiber length, its diameter, volume of fiber added etc. It is not effective to evaluate the tensile strength of CFRC with existing empirical equations. So, a prediction model for split tensile strength is necessary for CFRC to predict the strength for any grade of concrete using various parameters. The Machine Learning models are very effective method for prediction model development that have complex relationships. So, this paper discusses about tensile strength prediction of various Machine learning models like Multiple Linear Regression (MLR), Support Vector Regression (SVR), Decision Tree (DT) Regression, K Nearest Neighbors (KNN) Regression, Random Forest (RF) Regression, etc. of CFRC by collecting data from previous existing literatures of CFRC. Among that XGBoost shows a better R2 value of 0.95 and a lower Root Mean Square Error value of 0.348 MPa when training the data. Key Words: Machine Learning, Coir fiber Reinforced Concrete, Split Tensile Strength, Prediction Models