Advanced machine learning techniques for predicting compressive strength of ultra-high performance concrete

Abstract

This study presents a robust framework for predicting the compressive strength of ultra-high performance concrete (UHPC) using machine learning models, based on a comprehensive data set of 761 data points derived from various UHPC mix designs. Six models, including K-Nearest Neighbors (KNN), Gradient Boosting Regression (GBR), Random Forest Regression (RFR), Support Vector Regression (SVR), Stacking and eXtreme Gradient Boosting (XGBoost), were evaluated. Among them, XGBoost demonstrated the best prediction accuracy, achieving a coefficient of determination (R²) of 0.969 and a root mean square error (RMSE) of 4.626 MPa, outperforming the other models. The Stacking model also performed well with an R² of 0.960, though it slightly overestimated at higher compressive strength levels. SHapley Additive exPlanations (SHAP) analysis revealed that curing time, silica fume, and aggregate content were the most significant factors influencing compressive strength. Curing time emerged as the dominant factor, significantly surpassing other variables such as silica fume and aggregate content in its impact on compressive strength. This dominance is attributed to its critical role in hydration and compressive strength development, while silica fume and aggregates primarily contributed by enhancing matrix densification and structural integrity. SHAP feature dependency analysis further highlighted complex interactions, particularly between water content and superplasticizer dosage, affecting workability and compressive strength.

Graphical abstract

Keywords

ultra-high performance concrete / compressive strength / machine learning / SHAP / prediction

References

[1] Mathern A, von der Haar C, Marx S. Concrete support structures for offshore wind turbines: Current status, challenges, and future trends. Energies, 2021, 14(7): 1995

[2] Rasul M, Ahmad S, Adekunle S K, Al-Dulaijan S U, Maslehuddin M, Ali S I. Evaluation of the effect of exposure duration and fiber content on the mechanical properties of polypropylene fiber-reinforced UHPC exposed to sustained elevated temperature. Journal of Testing and Evaluation, 2020, 48(6): 4355–4369

[3] IyerN R. An overview of cementitious construction materials. New Materials in Civil Engineering, 2020: 1–64

[4] Ahmad S, Rasul M, Adekunle S K, Al-Dulaijan S U, Maslehuddin M, Ali S I. Mechanical properties of steel fiber-reinforced UHPC mixtures exposed to elevated temperature: Effects of exposure duration and fiber content. Composites. Part B, Engineering, 2019, 168: 291–301

[5] Tabish M, Zaheer M M, Baqi A. Effect of nano-silica on mechanical, microstructural and durability properties of cement-based materials: A review. Journal of Building Engineering, 2023, 65: 105676

[6] Vijayan D S, Devarajan P, Sivasuriyan A. A review on eminent application and performance of nano based silica and silica fume in the cement concrete. Sustainable Energy Technologies and Assessments, 2023, 56: 103105

[7] Yang H, Monasterio M, Zheng D, Cui H, Tang W, Bao X, Chen X. Effects of nano silica on the properties of cement-based materials: A comprehensive review. Construction & Building Materials, 2021, 282: 122715

[8] Wang D, Shi C, Farzadnia N, Shi Z, Jia H. A review on effects of limestone powder on the properties of concrete. Construction & Building Materials, 2018, 192: 153–166

[9] Kumar S, Gupta R C, Shrivastava S, Csetenyi L, Thomas B S. Preliminary study on the use of quartz sandstone as a partial replacement of coarse aggregate in concrete based on clay content, morphology and compressive strength of combined gradation. Construction & Building Materials, 2016, 107: 103–108

[10] Liu J, Wei J, Li J, Su Y, Wu C. A comprehensive review of ultra-high performance concrete (UHPC) behaviour under blast loads. Cement and Concrete Composites, 2024, 148: 105449

[11] Bajaber M A, Hakeem I Y. UHPC evolution, development, and utilization in construction: A review. Journal of Materials Research and Technology, 2021, 10: 1058–1074

[12] Marvila M T, de Azevedo A R G, de Matos P R, Monteiro S N, Vieira C M F. Materials for production of high and ultra-high performance concrete: Review and perspective of possible novel materials. Materials, 2021, 14(15): 4304

[13] Fan D, Yu R, Fu S, Yue L, Wu C, Shui Z, Liu K, Song Q, Sun M, Jiang C. Precise design and characteristics prediction of Ultra-High Performance Concrete (UHPC) based on artificial intelligence techniques. Cement and Concrete Composites, 2021, 122: 104171

[14] Fan D, Zhu J, Fan M, Lu J X, Chu S H, Dong E, Yu R. Intelligent design and manufacturing of ultra-high performance concrete (UHPC)—A review. Construction & Building Materials, 2023, 385: 131495

[15] Li Z, Qi J, Hu Y, Wang J. Estimation of bond strength between UHPC and reinforcing bars using machine learning approaches. Engineering Structures, 2022, 262: 114311

[16] Avcı-karataş Ç. Modeling approach for estimation of ultimate load capacity of concrete-filled steel tube composite stub columns based on relevance vector machine. Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi, 2021, 10(2): 615–626

[17] Cui J, Xing G, Miao P, Zhang Y, Chang Z, Khan A Q. Flexural behavior of RC beams strengthened with BFRP bars and CFRP U-jackets: Experimental and numerical analysis. Journal of Building Engineering, 2024, 97: 110932

[18] Avci-Karatas C. Prediction of ultimate load capacity of concrete-filled steel tube columns using multivariate adaptive regression splines (MARS). Steel and Composite Structures, 2019, 33(4): 583–594

[19] Avci-Karatas C. Application of machine learning in prediction of shear capacity of headed steel studs in steel–concrete composite structures. International Journal of Steel Structures, 2022, 22(2): 539–556

[20] Khan A Q, Awan H A, Rasul M, Siddiqi Z A, Pimanmas A. Optimized artificial neural network model for accurate prediction of compressive strength of normal and high strength concrete. Cleaner Materials, 2023, 10: 100211

[21] Jiang T, Gradus J L, Rosellini A J. Supervised machine learning: A brief primer. Behavior Therapy, 2020, 51(5): 675–687

[22] van Berkel N, Dennis S, Zyphur M, Li J, Heathcote A, Kostakos V. Modeling interaction as a complex system. Human–Computer Interaction, 2021, 36(4): 279–305

[23] Nasteski V. An overview of the supervised machine learning methods. Horizons Series B, 2017, 4: 51–62

[24] Osisanwo F Y, Akinsola J E T, Awodele O, Hinmikaiye J O, Olakanmi O. ,Akinjobi J. Supervised machine learning algorithms: Classification and comparison. International Journal of Computer Trends and Technology, 2017, 48(3): 128–138

[25] Rasul M, Hosoda A. Prediction of occurrence of thermal cracking of RC abutments using artificial neural networks. Journal of Structural Engineering A, 2019, 65: 560–568

[26] Avci-Karatas C. Artificial neural network (ANN) based prediction of ultimate axial load capacity of concrete-filled steel tube columns (CFSTCs). International Journal of Steel Structures, 2022, 22(5): 1341–1358

[27] RasulMHosodaA. Application of artificial neural network in predicting maximum thermal crack width of RC abutments using actual construction data. In: Proceedingds of FIB Symposium 2019 Concrete-Innovations in Materials, Desigh and Structures. Krakow: fib, 2019, 1339–1346.

[28] WangL. Estimating high-performance concrete compressive strength with support vector regression in hybrid method. Multiscale and Multidisciplinary Modeling, Experiments and Design. 2024, 7(1): 477–490

[29] Khan A Q, Naveed M H, Rasheed M D, Miao P. Prediction of compressive strength of fly ash-based geopolymer concrete using supervised machine learning methods. Arabian Journal for Science and Engineering, 2024, 49(4): 4889–4904

[30] KhanA QNaveedM HRasheedM DPimanmasA. Prediction of stress–strain behavior of PET FRP-confined concrete using machine learning models. Arabian Journal for Science and Engineering, 2024, 1–21

[31] Song Y, Zhao J, Ostrowski K A, Javed M F, Ahmad A, Khan M I, Aslam F, Kinasz R. Prediction of compressive strength of fly-ash-based concrete using ensemble and non-ensemble supervised machine-learning approaches. Applied Sciences, 2021, 12(1): 361

[32] Pernía-Espinoza A, Fernández-Ceniceros J, Antonanzas J, Urraca R, Martinez-de-Pison F J. Stacking ensemble with parsimonious base models to improve generalization capability in the characterization of steel bolted components. Applied Soft Computing, 2018, 70: 737–750

[33] Chung K L, Wang L, Ghannam M, Guan M, Luo J. Prediction of concrete compressive strength based on early-age effective conductivity measurement. Journal of Building Engineering, 2021, 35: 101998

[34] Boateng E Y, Otoo J, Abaye D A. Basic tenets of classification algorithms K-nearest-neighbor, support vector machine, random forest and neural network: A review. Journal of Data Analysis and Information Processing, 2020, 8(4): 341–357

[35] Patel A K, Chatterjee S, Gorai A K. Development of a machine vision system using the support vector machine regression (SVR) algorithm for the online prediction of iron ore grades. Earth Science Informatics, 2019, 12(2): 197–210

[36] GholamiRFakhariN. Support vector machine: Principles, parameters, and applications. Handbook of Neural Computation, 2017: 515–535

[37] Ao Y, Li H, Zhu L, Ali S, Yang Z. The linear random forest algorithm and its advantages in machine learning assisted logging regression modeling. Journal of Petroleum Science Engineering, 2019, 174: 776–789

[38] Rabbani A, Samui P, Kumari S. Implementing ensemble learning models for the prediction of shear strength of soil. Asian Journal of Civil Engineering, 2023, 24(7): 2103–2119

[39] Ghasemieh A, Lloyed A, Bahrami P, Vajar P, Kashef R. A novel machine learning model with Stacking Ensemble Learner for predicting emergency readmission of heart-disease patients. Decision Analytics Journal, 2023, 7: 100242

[40] Khan A Q, Deng P, Matsumoto T. Equivalent boundary conditions to analyze the realistic fatigue behaviors of a bridge RC slab. Structural Engineering and Mechanics, 2022, 82(3): 369–383

[41] Chicco D, Warrens M J, Jurman G. The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation. PeerJ. Computer Science, 2021, 7: e623

[42] Huang L, Zhang S S, Yu T, Wang Z Y. Compressive behaviour of large rupture strain FRP-confined concrete-encased steel columns. Construction & Building Materials, 2018, 183: 513–522

[43] Ağbulut Ü, Gürel A E, Biçen Y. Prediction of daily global solar radiation using different machine learning algorithms: Evaluation and comparison. Renewable & Sustainable Energy Reviews, 2021, 135: 110114

[44] Khanal S, Fulton J, Klopfenstein A, Douridas N, Shearer S. Integration of high resolution remotely sensed data and machine learning techniques for spatial prediction of soil properties and corn yield. Computers and Electronics in Agriculture, 2018, 153: 213–225

[45] Chen P, Hsieh H, Su K, Sigalingging X K, Chen Y, Leu J. Predicting station level demand in a bike‐sharing system using recurrent neural networks. IET Intelligent Transport Systems, 2020, 14(6): 554–561

[46] Lynch C J, Gore R. Short-range forecasting of COVID-19 during early onset at county, health district, and state geographic levels using seven methods: Comparative forecasting study. Journal of Medical Internet Research, 2021, 23(3): e24925

[47] Khan M, Lao J, Dai J G. Comparative study of advanced computational techniques for estimating the compressive strength of UHPC. Journal of Asian Concrete Federation, 2022, 8(1): 51–68

[48] Abuodeh O R, Abdalla J A, Hawileh R A. Assessment of compressive strength of Ultra-high Performance Concrete using deep machine learning techniques. Applied Soft Computing, 2020, 95: 106552

[49] Kashem A, Karim R, Malo S C, Das P, Datta S D, Alharthai M. Hybrid data-driven approaches to predicting the compressive strength of ultra-high-performance concrete using SHAP and PDP analyses. Case Studies in Construction Materials, 2024, 20: e02991

[50] VinnakotaS. Understanding the long-term evolution of CSH phases present in cement backfills. In: Proceedings of 37th Cement and Concrete science Conference. London: University College London, 2019.

[51] Han Q, Zhang P, Wu J, Jing Y, Zhang D, Zhang T. Comprehensive review of the properties of fly ash-based geopolymer with additive of nano-SiO₂. Nanotechnology Reviews, 2022, 11(1): 1478–1498

[52] Li S, Cheng S, Mo L, Deng M. Effects of steel slag powder and expansive agent on the properties of ultra-high performance concrete (UHPC): Based on a case study. Materials, 2020, 13(3): 683

[53] Amin M, Zeyad A M, Tayeh B A, Saad Agwa I. Effect of ferrosilicon and silica fume on mechanical, durability, and microstructure characteristics of ultra high-performance concrete. Construction & Building Materials, 2022, 320: 126233

[54] Park S, Wu S, Liu Z, Pyo S. The role of supplementary cementitious materials (SCMs) in ultra high performance concrete (UHPC): A review. Materials, 2021, 14(6): 1472

[55] Liu C, He X, Deng X, Wu Y, Zheng Z, Liu J, Hui D. Application of nanomaterials in ultra-high performance concrete: A review. Nanotechnology Reviews, 2020, 9(1): 1427–1444

[56] Dingqiang F, Rui Y, Zhonghe S, Chunfeng W, Jinnan W, Qiqi S. A novel approach for developing a green Ultra-High Performance Concrete (UHPC) with advanced particles packing meso-structure. Construction & Building Materials, 2020, 265: 120339

[57] Arora A, Almujaddidi A, Kianmofrad F, Mobasher B, Neithalath N. Material design of economical ultra-high performance concrete (UHPC) and evaluation of their properties. Cement and Concrete Composites, 2019, 104: 103346

[58] Ullah R, Qiang Y, Ahmad J, Vatin N I, El-Shorbagy M A. Ultra-high-performance concrete (UHPC): A state-of-the-art review. Materials, 2022, 15(12): 4131

[59] Saleh S, Li Y L, Hamed E, Mahmood A H, Zhao X L. Workability, strength, and shrinkage of ultra-high-performance seawater, sea sand concrete with different OPC replacement ratios. Journal of Sustainable Cement-Based Materials, 2023, 12(3): 271–291

[60] Zhang Y, Zhu Y, Qu S, Kumar A, Shao X. Improvement of flexural and tensile strength of layered-casting UHPC with aligned steel fibers. Construction & Building Materials, 2020, 251: 118893

[61] Huang H, Gao X, Li L, Wang H. Improvement effect of steel fiber orientation control on mechanical performance of UHPC. Construction & Building Materials, 2018, 188: 709–721

[62] Zhang X Y, Yu R, Zhang J J, Shui Z H. A low-carbon alkali activated slag based ultra-high performance concrete (UHPC): Reaction kinetics and microstructure development. Journal of Cleaner Production, 2022, 363: 132416

[63] Di Guida R, Engel J, Allwood J W, Weber R J M, Jones M R, Sommer U, Viant M R, Dunn W B. Non-targeted UHPLC-MS metabolomic data processing methods: A comparative investigation of normalisation, missing value imputation, transformation and scaling. Metabolomics, 2016, 12(5): 93

[64] Singh D, Singh B. Investigating the impact of data normalization on classification performance. Applied Soft Computing, 2020, 97: 105524

[65] RaschkaS. Model evaluation, model selection, and algorithm selection in machine learning. 2018, arXiv: 1811.12808

Conpeting interests

The authors declare that they have no competing interests.

RIGHTS & PERMISSIONS

2025 Higher Education Press