Sushant Poudel

The incorporation of waste ground glass powder (GGP) in concrete as a partial replacement of cement offers significant environmental benefits, such as reduction in CO₂ emission from cement manufacturing and decrease in the use of colossal landfill space. However, concrete is a heterogeneous material, and the prediction of its accurate compressive strength is challenging due to the inclusion of several non-linear parameters. This study explores the utilization of different machine learning (ML) algorithms: linear regression (LR), ElasticNet regression (ENR), a K-Nearest Neighbor regressor (KNN), a decision tree regressor (DT), a random forest regressor (RF), and a support vector regressor (SVR). A total of 187 sets of pertinent mix design experimental data were collected to train and test the ML algorithms. Concrete mix components such as cement content, coarse and fine aggregates, the water–cement ratio (W/C), various GGP chemical properties, and the curing time were set as input data (X), while the compressive strength was set as the output data (Y). Hyperparameter tuning was carried out to optimize the ML models, and the results were compared with the help of the coefficient of determination (R²) and root mean square error (RMSE). Among the algorithms considered, SVR demonstrates the highest accuracy and predictive capability with an R² value of 0.95 and RMSE of 3.40 MPa. Additionally, all the models exhibit R² values greater than 0.8, suggesting that ML models provide highly accurate and cost-effective means for evaluating and optimizing the compressive strength of GGP-incorporated sustainable concrete.

The incorporation of waste ground glass powder (GGP) as a partial replacement in cement offers significant environmental benefits, such as reduction in CO2 emission from cement manufacturing and decrease in the use of colossal landfill space. However, concrete is a heterogeneous material and the prediction of its accurate compressive strength is challenging due to the inclusion of several nonlinear parameters. This paper explores the utilization of different machine learning (ML) algorithm; Linear Regression (LR), Elastic Net regression (ENR), K-nearest Neighbor Regressor (KNN), Decision Tree Regressor (DT), Random Forest Regressor (RF), and Support Vector Machine (SVM). A total of 188 sets of pertinent mix design experimental data are collected to train and test the MLalgorithms. Concrete mix components such as cement content, coarse and fine aggregates, water-cement ratio (W/C) ratio, various GGP chemical properties, and curing time are set as input data (X) while the compressive strength is set as the output data (Y). The hyperparameter tuning is carried out to optimize the ML models, and the results are compared with the help of the coefficient of determination (R2) and root-mean-square error (RMSE). Among the algorithms used, SVM demonstrates the highest accuracy and predictive capability among all models with R2 value of 0.95 and RSME of 3.40 MPa. Additionally, all the models exhibit R2 value higher than 0.8, suggesting ML models provide highly accurate and cost-effective means for evaluating and optimizing the compressive strength of GGP concrete.

The significant anthropogenic carbon dioxide (CO₂) emissions from cement production and the disposal of the majority of post-consumer waste glass into landfill sites have increased environmental pollution. In order to reduce the environmental impact, ground glass pozzolan (GGP) as a partial cement replacement has drawn interest from the concrete industry. This review examines the potential of GGP as a supplementary cementitious material (SCM), exploring the chemical composition of pozzolans, the different types of glass used for GGP, and the impact of glass color on pozzolanic reactivity. In addition, this study gathers the most recent research articles on the fresh and mechanical properties of concrete incorporating GGP. Key findings show that the incorporation of GGP in concrete improves the modulus of elasticity and the compressive, tensile, flexural, and punching strengths due to the pozzolanic reactions. The results indicate that GGP, made from waste glass, has pozzolanic properties that form additional strength-enhancing calcium silicate hydrate (C-S-H) gel and densify the concrete matrix. Additionally, the life cycle assessments of GGP-incorporated concrete demonstrate reductions in energy consumption and CO₂ emissions compared to conventional concrete, supporting a circular economy and sustainable construction practices.

Cement production is one of the major sources of anthropogenic CO2 emissions in the world. In order to reduce its environmental impact, ground glass pozzolan (GGP) and other supplemental cementitious materials (SCMs) have drawn interest as potential substitutes. This review examines the potential of GGP as a SCM exploring the chemical composition of pozzolans, different types of glass used for GGP, and the impact of glass color on pozzolanic reactivity. Furthermore, the study gathers the most recent research articles on the fresh and mechanical properties of concrete including GGP, highlighting its benefits for workability, density, setting time, compressive strength, tensile strength, flexural strength, and punching strength. The results show that GGP, made from recycled waste glass, has pozzolanic properties that enhance the performance and sustainability of concrete. By incorporating GGP as a partial replacement for cement, the carbon footprint of concrete production can be lowered, supporting a circular economy and sustainable construction practices. This review paper emphasizes the technical viability and environmental advantages of using GGP in concrete, addressing the urgent need to reduce the environmental impact of cement production.

The use of fibre-reinforced concrete (FRC) has been substantially increasing in the last few years, in different fields of the construction industry. Recently, many experiments have been performed to observe the short- and long-term mechanical behaviour of FRC, and several models have been formulated to capture its mechanical response. In this work, the mechanical behaviour is simulated through the Lattice Discrete Particle Model (LDPM) and its extension to fibre-reinforced cementitious composites (LDPM-F). This paper aims to provide insights into the calibration process and potential pitfalls in a case where only limited experimental data are available—in this case, unconfined uniaxial compression and three-point bending tests on different mixes of polypropylene and steel fibre-reinforced concretes. As a first step, a sensitivity analysis is performed to weight the effect of each governing mesoscale parameter on the simulated macroscale behaviour. Then, for each mix at issue, different sets of model parameters are identified as capable of accurately matching the experimental evidence. As a validation, each calibrated set is used to simulate energy absorption tests on round panels. The validation stage shows that one of the identified sets, for the FRC with polypropylene fibres, accurately matches the round panels’ response, while the others result in acceptable predictions. For the mix with steel fibres, instead, none of the sets captures the experimental results, likely due to the different post-cracking behaviour detected in fracture and energy absorption tests. Finally, a parametric study showcases how the LDPM-F might serve as tool to optimise the mix design without extensive experimental investigations.