AUTHOR=Abbas Moutaman M. , Muntean Radu 

TITLE=Predicting mechanical properties of marble powder concrete using artificial neural networks and blockchain-rock for sustainable construction

JOURNAL=Frontiers in Built Environment

VOLUME=Volume 11 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/built-environment/articles/10.3389/fbuil.2025.1594735

DOI=10.3389/fbuil.2025.1594735

ISSN=2297-3362

ABSTRACT=Use of marble powder—an industrial by-product—serves as a supplementary cementitious material (SCM) and ensures sustainability by minimizing environmental impacts of cement manufacturing. This paper suggests a novel use of artificial neural networks (ANN) and Blockchain-Rock technology to enhance predictive accuracy and assure tracking of data in concrete mix optimization. Using an ANN model trained on 629 data sets, the proposed approach achieved high predictive accuracy for mechanical properties of marble powder concrete: Model I reached R2 = 0.99 and RMSE = 1.63 on the test set, while Model II achieved R2 = 1.00 and RMSE = 0.21. These results are superior or comparable to those of other machine learning models, such as a feedforward ANN (R2 = 0.985, RMSE = 1.12) and a general regression neural network (GRNN) (R2 = 0.92, RMSE = 4.83), highlighting the effectiveness of the proposed ANN architecture. This demonstrates the ANN’s ability to efficiently predict compressive and tensile strength of marble powder concrete, substantially reducing the need for standard long-duration tests. Additionally, Blockchain-Rock ensures secure and tamper-free tracking of material origin and concrete mixes, enabling transparency and efficiency in the supply chain. Experiments demonstrate that the addition of marble powder improves concrete strength and durability. Furthermore, ANN-based predictions enable real-time optimization of the concrete mix design. This dual approach offers an extended solution for sustainable construction by leveraging AI-based efficiency and blockchain-based data security. Future work can explore additional enhancements by real-time IoT integration and larger data sets to further improve predictive accuracy and industrial applicability.