Article
Computer Science, Interdisciplinary Applications
Surya Abisek Rajakarunakaran, Arun Raja Lourdu, Suresh Muthusamy, Hitesh Panchal, Ali Jawad Alrubaie, Mustafa Musa Jaber, Mohammed Hasan Ali, Iskander Tlili, Andino Maseleno, Ali Majdi, Shahul Hameed Masthan Ali
Summary: This study creates regression models based on machine learning to predict the compressive strength of self-compacting concrete (SCC). Through analysis of laboratory data, it is found that the random forest model can accurately predict the compressive strength of concrete.
ADVANCES IN ENGINEERING SOFTWARE
(2022)
Article
Construction & Building Technology
Jun Liu, Baodong Zhang, Yao Li, Bingjie Zhao, Yonggang Deng
Summary: In this study, a mathematical model was established using nonlinear programming software to investigate the influence of steel and polypropylene fibers on the strength of high-performance concrete. The results showed that the strength model of fiber-reinforced high-performance concrete complied with quadratic and cubic functional relations. Steel fiber and hybrid fiber significantly affected the compressive strength, while the amount of polypropylene fiber had no significant effect on splitting strength and flexural strength.
CONSTRUCTION AND BUILDING MATERIALS
(2022)
Article
Construction & Building Technology
M. Shi, Weigang Shen
Summary: This study aims to verify the effectiveness of using automated machine learning (AutoML) for predicting the compressive strength of concrete. By comparing different algorithms and datasets, the results show that Auto-Sklearn can build accurate prediction models without relying on expert experience.
Article
Multidisciplinary Sciences
Sara Elhishi, Asmaa Mohammed Elashry, Sara El-Metwally
Summary: This paper evaluates the performance of eight popular machine learning models in predicting concrete strength and finds that the XGBoost model outperforms others. By employing the SHAP technique to analyze the XGBoost model, researchers also provide insights for decision-making regarding concrete mix design and construction practices.
SCIENTIFIC REPORTS
(2023)
Article
Automation & Control Systems
Mohammad Amin Hariri-Ardebili, Golsa Mahdavi
Summary: This paper proposes three surrogate modeling techniques (polynomial chaos expansion, Kriging, and canonical low-rank approximation) for concrete compressive strength regression analysis. With a benchmark database of high-performance concrete, various sources of uncertainties in surrogate modeling are quantified. The Kriging-based surrogate models outperform the existing predictive models and show more stable results. The selection of a proper optimization algorithm is the most important factor in surrogate modeling.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Construction & Building Technology
Amir Ali Shahmansouri, Maziar Yazdani, Mehdi Hosseini, Habib Akbarzadeh Bengar, Hamid Farrokh Ghatte
Summary: The study demonstrates the potential of using artificial neural networks to predict the compressive strength and electrical resistivity of natural zeolitic concrete, significantly speeding up the process and improving accuracy. Experimental results from 324 different designs of natural zeolitic concrete specimens validate the accuracy and reliability of the model. This has important implications for cost reduction and time saving.
CONSTRUCTION AND BUILDING MATERIALS
(2022)
Article
Polymer Science
Chenchen Luan, Qingyuan Wang, Fuhua Yang, Kuanyu Zhang, Nodir Utashev, Jinxin Dai, Xiaoshuang Shi
Summary: This paper developed practical prediction models for splitting tensile strength and reinforcement-concrete bond strength of FAGC, based on regression analysis of experimental data and literature. The models can be used for design equations and estimating anchorage length accurately.
Article
Engineering, Civil
Tongxu Liu, Zhen Wang, Zilin Long, Junlin Zeng, Jingquan Wang, Jian Zhang
Summary: This study aims to establish an accurate and reliable prediction model for the direct shear strength (DSS) of precast concrete joints (PCJs) using support vector regression (SVR) algorithm. A new correlation matrix-based feature selection method was proposed, and three SVR models with different feature combinations were trained. The results show that the SVR algorithm can accurately and reliably predict the DSS of PCJs, and the proposed feature selection method improves the prediction performance of the SVR model. The influence of each input parameter on the DSS of PCJs is recognized and can provide useful information for future research on predicting the DSS of PCJs.
JOURNAL OF BRIDGE ENGINEERING
(2022)
Article
Multidisciplinary Sciences
Rui Hou, Qi Hou
Summary: This paper establishes a prediction model for the shear strength of ultrahigh-performance concrete (UHPC) beams. Through static shear tests and dynamic load experiments, related parameters for UHPC beam shear strength are determined. A total of 1200 BPNN models were trained and optimized using a genetic algorithm. The final model can reliably predict the shear strength of UHPC beams.
SCIENTIFIC REPORTS
(2023)
Article
Engineering, Civil
M. S. Sandeep, Koravith Tiprak, Sakdirat Kaewunruen, Phoonsak Pheinsusom, Withit Pansuk
Summary: In recent years, machine learning techniques have been widely applied in solving challenging structural engineering problems, providing highly accurate models to replace empirical and semiempirical prediction models. This paper discusses the fundamental terminologies and concepts of commonly used machine learning algorithms for solving structural engineering problems. It also presents a comprehensive literature review on the application of machine learning in predicting shear strength, covering various types of beams. The article concludes with major observations, challenges, and future scope in this field. It serves as a valuable resource for individuals unfamiliar with machine learning but eager to learn more.
Article
Chemistry, Physical
Ayaz Ahmad, Furqan Farooq, Pawel Niewiadomski, Krzysztof Ostrowski, Arslan Akbar, Fahid Aslam, Rayed Alyousef
Summary: Machine learning techniques are widely used in predicting the mechanical properties of concrete. By comparing individual algorithms with ensemble approaches such as bagging, it was found that the ensemble model outperforms decision trees and gene expression programming. Optimization of bagging can improve model accuracy, as demonstrated by various statistical indicators.
Article
Construction & Building Technology
Shi-Zhi Chen, Shu-Ying Zhang, Wan-Shui Han, Gang Wu
Summary: In this study, an ensemble learning algorithm GBRT was used to develop a prediction model for FCI bond strength, which showed higher accuracy compared to traditional models and common machine learning algorithms, proving to be feasible for practical application.
CONSTRUCTION AND BUILDING MATERIALS
(2021)
Article
Chemistry, Physical
Zaineb M. Alhakeem, Yasir Mohammed Jebur, Sadiq N. Henedy, Hamza Imran, Luis F. A. Bernardo, Hussein M. Hussein
Summary: A hybrid model with optimization technique was used to predict the compressive strength of eco-friendly concrete. The results showed that the proposed model had high accuracy and generalization. Furthermore, the factors influencing the compressive strength were explained using the SHAP approach.
Article
Construction & Building Technology
Gaoyang Liu, Bochao Sun
Summary: The mixing ratio of raw materials significantly affects the compressive strength of concrete. An Explainable Boosting Machine (EBM) is used to predict the compressive strength and explain the contribution of mix ratio factors. Bayesian optimization is employed for algorithm selection and hyperparameter optimization. The EBM algorithm shows excellent prediction performance with R2 = 0.93, RMSE = 4.33, and MAE = 3.10, allowing for the interpretation of individual feature contributions and determination of concrete compressive strength.
CASE STUDIES IN CONSTRUCTION MATERIALS
(2023)
Article
Construction & Building Technology
Mehrdad Abdi Moghadam, Ramezan Ali Izadifard
Summary: This study involves simulating the tensile strength of plain concrete and conducting experiments under high-temperature conditions, revealing that the addition of steel fibers can significantly improve the tensile strength of concrete. The established model is able to accurately predict the tensile strength of concrete under high-temperature exposure.
INTERNATIONAL JOURNAL OF CONCRETE STRUCTURES AND MATERIALS
(2021)
Article
Materials Science, Multidisciplinary
Umit Atici
ADVANCES IN MATERIALS SCIENCE AND ENGINEERING
(2016)
Article
Construction & Building Technology
U. Atici, A. Ersoy
CONSTRUCTION AND BUILDING MATERIALS
(2008)
Article
Environmental Sciences
Adem Ersoy, Tayfun Yusuf Yunsel, Uemit Atici
ENVIRONMENTAL TOXICOLOGY
(2008)
Article
Computer Science, Artificial Intelligence
U. Atici
EXPERT SYSTEMS WITH APPLICATIONS
(2011)
Article
Engineering, Industrial
U. Atici, A. Ersoy
JOURNAL OF MATERIALS PROCESSING TECHNOLOGY
(2009)
Article
Construction & Building Technology
U. Atici, A. Ersoy, B. Ozturk
MAGAZINE OF CONCRETE RESEARCH
(2009)
Article
Metallurgy & Metallurgical Engineering
U. Atici, R. Comakli
JOURNAL OF THE SOUTHERN AFRICAN INSTITUTE OF MINING AND METALLURGY
(2019)
Article
Computer Science, Artificial Intelligence
Umit Atici, Adem Ersoy
APPLIED ARTIFICIAL INTELLIGENCE
(2017)
Article
Engineering, Geological
A. Ersoy, U. Atici
ROCK MECHANICS AND ROCK ENGINEERING
(2007)
Article
Mining & Mineral Processing
A Ersoy, U Atici
JOURNAL OF MINING SCIENCE
(2005)
Article
Engineering, Mechanical
A Ersoy, S Buyuksagic, U Atici
Article
Materials Science, Multidisciplinary
A Ersoy, U Atici
DIAMOND AND RELATED MATERIALS
(2004)