The construction industry remains one of the leading contributors to global carbon emissions, with cement production alone responsible for approximately 7–8% of total CO₂ emissions. An urgent shift toward sustainable practices has resulted in growing adoption of industrial waste materials, such as fly ash, slag, red mud, and silica fume, as alternative binders or supplementary cementitious materials (SCMs) in concrete. These eco-friendly materials support circular economy principles, reduce clinker content, and help to mitigate environmental impacts.
However, the inherently heterogeneous nature of industrial waste introduces significant variability in chemical composition, reactivity, and long-term performance, making reliable mix design and prediction a challenge. Traditional experimental and trial-and-error methods are not only time-consuming and expensive, but also struggle to capture the complex, nonlinear interactions between multiple input parameters.

Artificial Intelligence (AI) and Machine Learning (ML) techniques offer powerful alternatives. By leveraging large, multivariate datasets from experimental and real-world studies, AI/ML models can accurately predict and optimize concrete properties ,including compressive strength, durability, setting time, and more, streamlining the discovery and deployment of sustainable binders.

Aims and Scope

This Research Topic aims to showcase the latest advances and applications of AI and ML in sustainable concrete systems using industrial waste-based binders. We welcome contributions that demonstrate predictive modeling capabilities, data-driven optimization, and decision-support tools that accelerate material innovation and reduce reliance on conventional experimental protocols. By embracing digital and AI-powered methods, the construction sector can realize greater efficiency, consistency, and sustainability, reducing manual effort and supporting data-driven decision-making.

Themes

Prospective authors are encouraged to address, but are not limited to, the following themes:
1. Data Collection and Processing
o Standardization, compilation, and feature engineering of experimental datasets involving waste-based binders.
o Data cleaning approaches tailored for AI/ML applications in construction materials.
2. Predictive Modeling
o Development of AI/ML models (e.g., ANN, SVM, XGBoost, Random Forest) for predicting compressive strength, workability, and durability parameters.
o Benchmarking against traditional regression and statistical methodologies.
3. Optimization and Decision Support
o Use of genetic algorithms, swarm intelligence, or hybrid ML for binder mix design optimization.
o AI-powered decision support tools for selecting sustainable binder systems and reducing manual mix design labor.
4. Uncertainty, Interpretability, and Explainability
o Interpretable AI models (e.g., SHAP, LIME) for elucidating variable influence.
o Sensitivity analysis and uncertainty quantification in model predictions.
5. Integration with Sustainability Metrics
o Coupling ML predictions with life cycle assessment (LCA) data to enable low-carbon and resource-efficient mix designs.
o Tools for predicting embodied energy or carbon footprint.
o Case studies of AI/ML-driven binder design using real industrial waste streams.

Article Types
We invite contributions in the form of:
o Original Research Articles
o Review Articles
o Case Studies / Field Studies
o Short Communications / Technical Notes

Invitation

By fostering cross-disciplinary collaboration among engineers, material scientists, and AI researchers, this Research Topic seeks to accelerate the adoption of predictive, data-driven sustainable concrete practices. Embracing advanced computational tools not only streamlines material development but can also help reduce routine workloads, enabling the industry to focus human expertise on high-value innovation.

Are you interested in contributing a manuscript to this Research Topic?

Application of AI/ML for Predictive Modeling in Waste-based Binder Performance in sustainable concrete