DATA-DRIVEN APPROACH FOR PREDICTING CONCRETE STRENGTH USING ARTIFICIAL NEURAL NETWORKS

Predicting the strength of concrete is a crucial task in the construction industry as it directly impacts the safety and reliability of structures. Traditional methods for predicting concrete strength often rely on physical testing, which can be timeconsuming, expensive, and labour-intensive. In recent years, data-driven approaches leveraging artificial neural networks (ANNs) have gained significant attention due to their ability to capture complex relationships between input variables and concrete strength. This presents a data-driven approach for predicting concrete strength using artificial neural networks. The proposed methodology involves collecting a comprehensive dataset comprising various input parameters such as cement content, water-to-cement ratio, aggregate type, and curing time, along with corresponding concrete strength values obtained from experimental tests. The dataset is pre-processed and divided into training and testing sets to train and evaluate the performance of the ANN model. The artificial neural network architecture consists of multiple layers of interconnected nodes, allowing the model to learn and recognize intricate patterns in the data. The training process involves optimizing the network's parameters using backpropagation and gradient descent algorithms, minimizing the difference between predicted and actual concrete strength values. The including mean squared error, coefficient of determination, and root mean squared error. The proposed data-driven approach using artificial neural networks achieves accurate predictions of concrete strength. The ANN model exhibits strong generalization capabilities, effectively capturing the nonlinear relationships between input variables and concrete strength. Compared to traditional testing methods, the developed approach significantly reduces the time and cost required for concrete strength prediction. In the field of concrete strength prediction by introducing a data-driven approach based on artificial neural networks. The proposed methodology provides an efficient and accurate means for estimating concrete strength, enabling engineers to make informed decisions during the design and construction phases. Furthermore, the study highlights the potential of data-driven techniques in enhancing various aspects of civil engineering, paving the way for further advancements in the field.