An AI-driven approach for modeling the compressive strength of sustainable concrete incorporating waste marble as an industrial by-product DOI Creative Commons
Ramin Kazemi, Seyedali Mirjalili

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Nov. 5, 2024

Abstract A key goal of environmental policies and circular economy strategies in the construction sector is to convert demolition industrial wastes into reusable materials. As an by-product, Waste marble (WM), has potential replace cement fine aggregate concrete which helps with saving natural resources reducing harm. While many studies have so far investigated effect WM on compressive strength (CS), it undeniable that conducting experimental activities requires time, money, re-testing changing materials conditions. Hence, this study seeks move from traditional approaches towards artificial intelligence-driven by developing three models—artificial neural network (ANN) hybrid ANN ant colony optimization (ACO) biogeography-based (BBO) predict CS concrete. For purpose, a comprehensive dataset including 1135 data records employed literature. The models’ performance assessed using statistical metrics error histograms, K -fold cross-validation analysis applied avoid overfitting problems, emphasize reliable predictive capabilities, generalize them. indicated ANN-BBO model performed best correlation coefficient (R) 0.9950 root mean squared (RMSE) 1.2017 MPa. Besides, distribution results revealed outperformed ANN-ACO narrower range errors 98% predicted points training phase experienced [-10%, 10%], whereas for models, percentage was 85% 79%, respectively. Additionally, SHapley Additive exPlanations (SHAP) clarify impact input variables prediction accuracy found specimen’s age most influential variable. Eventually, validate ANN-BBO, comparison previous studies’ models.

Language: Английский

Research on 3D Printing Concrete Mechanical Properties Prediction Model Based on Machine Learning DOI Creative Commons
Yonghong Zhang, Suping Cui, Bohao Yang

et al.

Case Studies in Construction Materials, Journal Year: 2025, Volume and Issue: unknown, P. e04254 - e04254

Published: Jan. 1, 2025

Language: Английский

Citations

0
Flexural Behavior of Lightweight Sandwich Panels with Rice Husk Bio-Aggregate Concrete Core and Sisal Fiber-Reinforced Foamed Cementitious Faces DOI Open Access
Daniele Oliveira Justo dos Santos, Paulo Roberto Lopes Lima, Romildo Dias Tolêdo Filho

et al.

Materials, Journal Year: 2025, Volume and Issue: 18(8), P. 1850 - 1850

Published: April 17, 2025

The development of sustainable and energy-efficient construction materials is crucial for mitigating the growing environmental impact building sector. This study introduces a new lightweight sandwich panel, featuring core made concrete with rice husk bio-aggregate (RHB) faces constructed from foamed cementitious composites. innovative design aims to promote sustainability by utilizing agro-industrial waste while maintaining satisfactory mechanical performance. Composites were produced 4% short sisal fibers matrices containing 15%, 20%, 30% foaming agent. These composites evaluated density, direct compression, four-point bending. It was found that mixture 20% foam volume demonstrated highest efficiency use in production panels. Concrete mixtures 50%, 60%, 70% bio-aggregates tested density compressive strength used panels densities ranging 670 1000 kg/m3. Mechanical evaluation under flexion shear indicated presence inhibited crack propagation face, enabling creation deflection-hardening behavior. On other hand, increase RHB content led reduction ultimate stress on stress, toughness

Language: Английский

Citations

0
An AI-driven approach for modeling the compressive strength of sustainable concrete incorporating waste marble as an industrial by-product DOI Creative Commons
Ramin Kazemi, Seyedali Mirjalili

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Nov. 5, 2024

Abstract A key goal of environmental policies and circular economy strategies in the construction sector is to convert demolition industrial wastes into reusable materials. As an by-product, Waste marble (WM), has potential replace cement fine aggregate concrete which helps with saving natural resources reducing harm. While many studies have so far investigated effect WM on compressive strength (CS), it undeniable that conducting experimental activities requires time, money, re-testing changing materials conditions. Hence, this study seeks move from traditional approaches towards artificial intelligence-driven by developing three models—artificial neural network (ANN) hybrid ANN ant colony optimization (ACO) biogeography-based (BBO) predict CS concrete. For purpose, a comprehensive dataset including 1135 data records employed literature. The models’ performance assessed using statistical metrics error histograms, K -fold cross-validation analysis applied avoid overfitting problems, emphasize reliable predictive capabilities, generalize them. indicated ANN-BBO model performed best correlation coefficient (R) 0.9950 root mean squared (RMSE) 1.2017 MPa. Besides, distribution results revealed outperformed ANN-ACO narrower range errors 98% predicted points training phase experienced [-10%, 10%], whereas for models, percentage was 85% 79%, respectively. Additionally, SHapley Additive exPlanations (SHAP) clarify impact input variables prediction accuracy found specimen’s age most influential variable. Eventually, validate ANN-BBO, comparison previous studies’ models.

Language: Английский

Citations

1