Smart Predictive Modeling for Compressive Strength in Sisal-Fiber-Reinforced-Concrete Composites: Harnessing SVM, GP, and ANN Techniques

The integration of AI in civil engineering marks a transformative shift, streamlining tasks from project planning to infrastructure maintenance. Machine learning enhances accuracy in data analysis, optimizing design and decision-making processes. AI’s role spans structural engineering for design optimization and real-time monitoring of infrastructure health. The application of AI in civil engineering not only boosts efficiency but also introduces innovative approaches for sustainable and resilient urban development. This paper ventures into the forefront of construction innovation by exploring the synergy between sustainable materials and advanced prediction methods. Focusing on integrating sisal fiber, a renewable agricultural waste, into concrete, we employ cutting-edge techniques such as Support Vector Machine (SVM), Gaussian Progress (GP), Artificial Neural Network (ANN), Linear Regression (LR), and Non-linear Regression (NLR) for predicting the compressive strength of Sisal-Fiber-Reinforced-Concrete Composites (SFRCC) based on their mixture composition. The study aims to redefine predictive capabilities for these composite materials, providing insights that bridge sustainability and technological advancement in the dynamic field of construction. The findings reveal that the model derived from the artificial neural network (ANN) demonstrates superior efficiency, exhibiting a higher coefficient of determination and lower RMSE and SI values compared to linear and nonlinear regression, support vector machine, and Gaussian progress methods.