Strength and Predictive Modeling of Corn Cob Ash Blended Concrete Using Multi-Output Artificial Neural Network Approach

Abstract

The growing demand for sustainable building materials has motivated significant research into alternative cementitious binders. Corn Cob Ash (CCA), an agricultural by-product rich in silica, has potential as a partial cement replacement in concrete. This study investigates the mechanical performance of M20 concrete incorporating CCA at replacement levels of 0%, 10%, 20%, and 30% by weight of cement. Compressive strength was evaluated at 7, 28, 56, and 90 days. Results indicate a gradual reduction in strength with increasing CCA content, although mixes containing up to 20% CCA demonstrated comparable strength development to conventional concrete. Additionally, a multi-output Artificial Neural Network (ANN) model was developed to predict compressive strength at different curing ages using mix parameters as input features. The ANN architecture (16-8-4) was trained using Leave-One-Out Cross-Validation due to limited experimental samples. The model achieved satisfactory prediction capability, demonstrating the feasibility of machine learning for strength forecasting in sustainable concrete systems. The findings suggest that CCA can be used as a partial cement replacement up to 20% without significant compromise in strength, contributing to eco-friendly and resource-efficient construction.