Miroslav Radomirović
Research Subjects
Research Interests
Miroslav Radomirović has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Predrag Petronijević
Faculty of Civil Engineering, University of Belgrade, Bulevar kralja Aleksandra 73, 11000 Belgrade, Serbia
3 shared publications - Tatijana Vasiljević
Faculty of Technical Sciences, University of Priština, Knjaza Milosa 7, 38220 Kosovska Mitrovica, Serbia
3 shared publications - Marijana Hadzima-Nyarko
Faculty of Civil Engineering and Architecture, Osijek, Josip Juraj Strossmayer University of Osijek, 31000 Osijek, Croatia
2 shared publications - Miljan Kovačević
Faculty of Technical Sciences, University of Priština, Knjaza Milosa 7, 38220 Kosovska Mitrovica, Serbia
2 shared publications
Latest Publications
- 1 January 2026
- Published by Centre for Evaluation in Education and Science (CEON/CEES)
- No. Session 1_,p. 42-49
Evaluating parametric linear and interaction models for compressive strength estimation of high-performance concrete
Accurate prediction of High-Performance Concrete (HPC) compressive strength is essential for optimizing mix formulations, ensuring structural reliability, and advancing sustainable construction practices. Traditional mix design relies on exhaustive, trial-and-error laboratory testing that consumes substantial material volumes. By minimizing physical mix iterations, predictive modeling directly lowers cement consumption and reduces associated construction-related CO2 emissions. This paper presents a dedicated parametric linear modeling framework, evaluating Simple Linear Regression (SLR), Multiple Linear Regression (MLR), and Polynomial Interaction Models, for estimating HPC compressive strength across an experimental dataset of samples (4.10 to 291.00 MPa). Single-variable linear models demonstrate weak predictive precision (R < 0.36, MAPE > 35 %), underscoring that individual constituents cannot account for overall strength development. Transitioning to Multiple Linear Regression incorporates all mix ingredients simultaneously, while the introduction of multi-variable polynomial interaction terms significantly enhances precision by capturing synergistic cross-product effects between key binders, water content, and curing age. The primary practical advantage of this linear approach lies in yielding explicit, closed-form mathematical equations. Unlike complex, non-parametric "black-box" systems, these transparent linear formulas provide structural engineers with straightforward, immediately applicable tools for site quality control, material optimization, and rapid strength estimation.
- 1 January 2026
- Published by Centre for Evaluation in Education and Science (CEON/CEES)
- No. Session 1_,p. 42-49
Evaluating parametric linear and interaction models for compressive strength estimation of high-performance concrete
Accurate prediction of High-Performance Concrete (HPC) compressive strength is essential for optimizing mix formulations, ensuring structural reliability, and advancing sustainable construction practices. Traditional mix design relies on exhaustive, trial-and-error laboratory testing that consumes substantial material volumes. By minimizing physical mix iterations, predictive modeling directly lowers cement consumption and reduces associated construction-related CO2 emissions. This paper presents a dedicated parametric linear modeling framework, evaluating Simple Linear Regression (SLR), Multiple Linear Regression (MLR), and Polynomial Interaction Models, for estimating HPC compressive strength across an experimental dataset of samples (4.10 to 291.00 MPa). Single-variable linear models demonstrate weak predictive precision (R < 0.36, MAPE > 35 %), underscoring that individual constituents cannot account for overall strength development. Transitioning to Multiple Linear Regression incorporates all mix ingredients simultaneously, while the introduction of multi-variable polynomial interaction terms significantly enhances precision by capturing synergistic cross-product effects between key binders, water content, and curing age. The primary practical advantage of this linear approach lies in yielding explicit, closed-form mathematical equations. Unlike complex, non-parametric "black-box" systems, these transparent linear formulas provide structural engineers with straightforward, immediately applicable tools for site quality control, material optimization, and rapid strength estimation.
- Open Access
- 17 January 2025
- Published by MDPI AG in Computation
- Vol. 13 (1) , 17
Comparative Analysis of Machine Learning Models for Predicting Interfacial Bond Strength of Fiber-Reinforced Polymer-Concrete
This study presents a detailed analysis of various machine learning models for predicting the interfacial bond strength of fiber-reinforced polymer (FRP) concrete, including multiple linear regression, Multigene Genetic Programming (MGGP), an ensemble of regression trees, Gaussian Process Regression (GPR), Support Vector Regression (SVR), and neural networks. The evaluation was based on their predictive accuracy. The optimal model identified was the GPR ARD Exponential model, which achieved a mean absolute error (MAE) of 1.8953 MPa and a correlation coefficient (R) of 0.9658. An analysis of this optimal model highlighted the most influential variables affecting the bond strength. Additionally, the research identified several models with lower expression complexity and reduced accuracy, which may still be applicable in practical scenarios.
- 17 January 2025
- Published by MDPI AG in Computation
- Vol. 13 (1) , 17
Open AccessComparative Analysis of Machine Learning Models for Predicting Interfacial Bond Strength of Fiber-Reinforced Polymer-Concrete
This study presents a detailed analysis of various machine learning models for predicting the interfacial bond strength of fiber-reinforced polymer (FRP) concrete, including multiple linear regression, Multigene Genetic Programming (MGGP), an ensemble of regression trees, Gaussian Process Regression (GPR), Support Vector Regression (SVR), and neural networks. The evaluation was based on their predictive accuracy. The optimal model identified was the GPR ARD Exponential model, which achieved a mean absolute error (MAE) of 1.8953 MPa and a correlation coefficient (R) of 0.9658. An analysis of this optimal model highlighted the most influential variables affecting the bond strength. Additionally, the research identified several models with lower expression complexity and reduced accuracy, which may still be applicable in practical scenarios.
- Open Access
- 25 November 2024
- Published by MDPI AG in Preprints.org
Comparative Analysis of Machine Learning Models for Predicting Interfacial Bond Strength of Fiber‐ Reinforced Polymer‐Concrete
This study presents a detailed analysis of various machine learning models for predicting the interfacial bond strength of fiber-reinforced polymer (FRP) concrete, utilizing multiple linear regression, an ensemble of regression trees, Gaussian Process Regression (GPR), Support Vector Regression (SVR), Multigene Genetic Programming (MGGP), and neural network models. The models were evaluated based on their accuracy. The optimal model identified was the GPR ARD model, which achieved a mean absolute error (MAE) of 1.8953 MPa and a correlation coefficient (R) of 0.9658. Analysis of this optimal model highlighted that the three most influential variables affecting the bond strength are the length of the FRP strip (Lf), the thickness of the strip ( tf), and the compressive strength of the concrete to which the strip is applied (fc'). Additionally, the research identified several models with lower expression complexity and reduced accuracy, which may still be applicable in practical scenarios.
- 25 November 2024
- Published by MDPI AG in Preprints.org
Open AccessComparative Analysis of Machine Learning Models for Predicting Interfacial Bond Strength of Fiber‐ Reinforced Polymer‐Concrete
This study presents a detailed analysis of various machine learning models for predicting the interfacial bond strength of fiber-reinforced polymer (FRP) concrete, utilizing multiple linear regression, an ensemble of regression trees, Gaussian Process Regression (GPR), Support Vector Regression (SVR), Multigene Genetic Programming (MGGP), and neural network models. The models were evaluated based on their accuracy. The optimal model identified was the GPR ARD model, which achieved a mean absolute error (MAE) of 1.8953 MPa and a correlation coefficient (R) of 0.9658. Analysis of this optimal model highlighted that the three most influential variables affecting the bond strength are the length of the FRP strip (Lf), the thickness of the strip ( tf), and the compressive strength of the concrete to which the strip is applied (fc'). Additionally, the research identified several models with lower expression complexity and reduced accuracy, which may still be applicable in practical scenarios.