Ensemble learning-based strength prediction and model interpretability analysis of engineered cementitious composites

Abstract

Accurate prediction of compressive strength is essential for improving the performance and durability of Engineered Cementitious Composites (ECC) in construction applications. Traditional methods often fall short in accounting for the complex interactions between material properties, such as fiber type, matrix composition, and curing conditions. To address this challenge, this study presents an advanced ensemble learning framework based on a dataset of 313 ECC samples characterized by 18 key features. The ensemble model integrates three base learners, namely Random Forest (RF), Gradient Boosting Decision Trees (GBDT), and Support Vector Regression (SVR), along with a meta-learner selected from ten candidate models. The proposed ensemble model demonstrates significantly higher prediction accuracy compared to conventional approaches. The results show that the ensemble model achieves a coefficient of determination (R²) of 0.896, a root mean square error (RMSE) of 5.734, and a mean absolute error (MAE) of 4.505, substantially outperforming individual models. Among the evaluated meta-learners, Lasso Regression was identified as the optimal choice. Its regularization capability effectively mitigated overfitting and enhanced generalization, leading to a notable improvement in the final predictive performance of the stacking framework. Furthermore, SHapley Additive exPlanations (SHAP) and Partial Dependence Plots (PDP) were employed for model interpretability and visualization. The analysis reveals that factors such as fiber elastic modulus, silica fume content, and fiber volume fraction significantly contribute to the enhancement of ECC compressive strength. This model provides practical insights for optimizing the design and application of ECC materials.

Keywords

Engineered cementitious composites / Strength prediction / Ensemble learning / Model interpretability

References

[1] Alabduljabbar H, Khan K, Awan HH, Alyousef R, Mohamed AM, Eldin SM. Modeling the capacity of engineered cementitious composites for self-healing using AI-based ensemble techniques. Case Studies in Construction Materials. 2023, 18: e01805.

[2] Altayeb M, Wang X, Musa TH. An ensemble method for predicting the mechanical properties of strain hardening cementitious composites. Construction and Building Materials. 2021, 286122807.

[3] Aslani F, Sun J, Bromley D, Ma G. Fiber-reinforced lightweight self-compacting concrete incorporating scoria aggregates at elevated temperatures. Structural Concrete. 2019, 20(3): 1022-1035.

[4] Aslani F, Sun J, Huang G. Mechanical behavior of fiber-reinforced self-compacting rubberized concrete exposed to elevated temperatures. Journal of Materials in Civil Engineering. 2019, 311204019302.

[5] Barkhordari MS, Armaghani DJ, Mohammed AS, Ulrikh DV. Data-driven compressive strength prediction of fly ash concrete using ensemble learner algorithms. Buildings. 2022, 122132.

[6] Breiman L. Bagging predictors. Machine Learning. 1996, 24: 123-140.

[7] Bui D-K, Nguyen T, Chou J-S, Nguyen-Xuan H, Ngo TD. A modified firefly algorithm-artificial neural network expert system for predicting compressive and tensile strength of high-performance concrete. Construction and Building Materials. 2018, 180: 320-333.

[8] Chen, T., & Guestrin, C. (2016). Xgboost: A scalable tree boosting system. Paper presented at the Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining

[9] Chen Y, Yu J, Leung CKY. Use of high strength strain-hardening cementitious composites for flexural repair of concrete structures with significant steel corrosion. Construction and Building Materials. 2018, 167: 325-337.

[10] Choi J-I, Jang SY, Kwon S-J, Lee BY. Tensile behavior and cracking pattern of an ultra-high performance mortar reinforced by polyethylene fiber. Advances in Materials Science and Engineering. 2017, 2017: 1-10.

[11] Curosu I, Liebscher M, Mechtcherine V, Bellmann C, Michel S. Tensile behavior of high-strength strain-hardening cement-based composites (HS-SHCC) made with high-performance polyethylene, aramid and PBO fibers. Cement and Concrete Research. 2017, 98: 71-81.

[12] Ding Y, Liu J-P, Bai Y-L. Linkage of multi-scale performances of nano-CaCO3 modified ultra-high performance engineered cementitious composites (UHP-ECC). Construction and Building Materials. 2020.

[13] Ding Y, Yu K, Mao W, Zhang S. Performance-enhancement of high-strength strain-hardening cementitious composite by nano-particles: Mechanism and property characterization. Structural Concrete. 2021, 2342061-2075.

[14] Ding Y, Yu K-Q, Yu J-T, Xu S-L. Structural behaviors of ultra-high performance engineered cementitious composites (UHP-ECC) beams subjected to bending-experimental study. Construction and Building Materials. 2018, 177: 102-115.

[15] Fan W, Zhuge Y, Ma X, Chow CWK, Gorjian N. Strain hardening behaviour of PE fibre reinforced calcium aluminate cement (CAC) – Ground granulated blast furnace (GGBFS) blended mortar. Construction and Building Materials. 2020.

[16] Fei Z, Liang S, Cai Y, Shen Y. Ensemble machine-learning-based prediction models for the compressive strength of recycled powder mortar. Materials. 2023, 16(2): 583.

[17] Feng W, Wang Y, Sun J, Tang Y, Wu D, Jiang Z, Wang X. Prediction of thermo-mechanical properties of rubber-modified recycled aggregate concrete. Construction and Building Materials. 2022, 318: 125970.

[18] Ghafor K, Ahmed HU, Faraj RH, Mohammed AS, Kurda R, Qadir WS, Abdalla AA. Computing models to predict the compressive strength of engineered cementitious composites (ECC) at various mix proportions. Sustainability. 2022, 141912876.

[19] Golafshani EM, Behnood A, Arashpour M. Predicting the compressive strength of normal and high-performance concretes using ANN and ANFIS hybridized with Grey Wolf Optimizer. Construction and Building Materials. 2020, 232117266.

[20] Hastie T, Rosset S, Zhu J, Zou H. Multi-class adaboost. Statistics and its Interface. 2009, 23349-360.

[21] Huang, D., Hu, D., He, J., & Xiong, Y. (2018). Structure damage detection based on ensemble learning. Paper presented at the 2018 9th International Conference on Mechanical and Aerospace Engineering (ICMAE)

[22] Huang B-T, Weng K-F, Zhu J-X, Xiang Y, Dai J-G, Li VC. Engineered/strain-hardening cementitious composites (ECC/SHCC) with an ultra-high compressive strength over 210 MPa. Composites Communications. 2021.

[23] Huang B-T, Wu J-Q, Yu J, Dai J-G, Leung CKY. High-strength seawater sea-sand Engineered Cementitious Composites (SS-ECC): Mechanical performance and probabilistic modeling. Cement and Concrete Composites. 2020.

[24] Huang B-T, Zhu J-X, Weng K-F, Li VC, Dai J-G. Ultra-high-strength engineered/strain-hardening cementitious composites (ECC/SHCC): Material design and effect of fiber hybridization. Cement and Concrete Composites. 2022.

[25] Kamal A, Kunieda M, Ueda N, Nakamura H. Evaluation of crack opening performance of a repair material with strain hardening behavior. Cement and Concrete Composites. 2008, 3010863-871.

[26] Kim M-J, Choi H-J, Shin W, Oh T, Yoo D-Y. Development of impact resistant high-strength strain-hardening cementitious composites (HS-SHCC) superior to reactive powder concrete (RPC) under flexure. Journal of Building Engineering. 2021.

[27] Kim M-J, Chun B, Choi H-J, Shin W, Yoo D-Y. Effects of supplementary cementitious materials and curing condition on mechanical properties of ultra-high-performance, strain-hardening cementitious composites. Applied Sciences. 2021.

[28] Lee T, Kim J-H, Lee S-J, Ryu S-K, Joo B-C. Improvement of concrete crack segmentation performance using stacking ensemble learning. Applied Sciences. 2023, 13(4): 2367.

[29] Lei D-Y, Guo L-P, Chen B, Curosu I, Mechtcherine V. The connection between microscopic and macroscopic properties of ultra-high strength and ultra-high ductility cementitious composites (UHS-UHDCC). Composites Part b: Engineering. 2019, 164: 144-157.

[30] Li H, Chung H, Li Z, Li W. Compressive strength prediction of fly ash-based concrete using single and hybrid machine learning models. Buildings. 2024, 14103299.

[31] Li L, Cai Z, Yu K, Zhang YX, Ding Y. Performance-based design of all-grade strain hardening cementitious composites with compressive strengths from 40 MPa to 120 MPa. Cement and Concrete Composites. 2019, 97: 202-217.

[32] Li L, Shi J, Kou J, Yan Z. Mechanical properties of high-ductility concrete at different temperatures. Magazine of Concrete Research. 2023, 7510518-528.

[33] Li Q, Song Z. Prediction of compressive strength of rice husk ash concrete based on stacking ensemble learning model. Journal of Cleaner Production. 2023, 382: 135279.

[34] Li Y, Yang X, Ren C, Wang L, Ning X. Predicting the compressive strength of ultra-high-performance concrete based on machine learning optimized by meta-heuristic algorithm. Buildings. 2024, 14(5): 1209.

[35] Luo J, Cai Z, Yu K, Zhu W, Lu Z. Temperature impact on the micro-structures and mechanical properties of high-strength engineered cementitious composites. Construction and Building Materials. 2019, 226686-698.

[36] Nguyễn HH, Choi J-I, Park S-E, Cha SL, Huh J, Lee BY. Autogenous healing of high strength engineered cementitious composites (ECC) using calcium-containing binders. Construction and Building Materials. 2020.

[37] Nguyen MH, Trinh SH, Ly H-B. Toward improved prediction of recycled brick aggregate concrete compressive strength by designing ensemble machine learning models. Construction and Building Materials. 2023, 369: 130613.

[38] Nunez I, Marani A, Nehdi ML. Mixture optimization of recycled aggregate concrete using hybrid machine learning model. Materials. 2020, 13194331.

[39] Oh T, Kim M-J, Banthia N, Yoo D-Y. Influence of curing conditions on mechanical and microstructural properties of ultra-high-performance strain-hardening cementitious composites with strain capacity up to 17.3%. Developments in the Built Environment. 2023.

[40] Oh T, Kim M-J, Kim S, Lee SK, Kang M-C, Yoo D-Y. Electrical and mechanical properties of high-strength strain-hardening cementitious composites containing silvered polyethylene fibers. Journal of Building Engineering. 2022.

[41] Peterson LE. K-nearest neighbor. Scholarpedia. 2009, 421883.

[42] Ranstam J, Cook JA. LASSO regression. British Journal of Surgery. 2018, 105101348.

[43] Ren Z, Sun J, Tang W, Zeng X, Zeng H, Wang Y, Wang X. Mechanical and electrical properties investigation for electrically conductive cementitious composite containing nano-graphite activated magnetite. Journal of Building Engineering. 2022, 57104847.

[44] Ren Z, Sun J, Zeng X, Chen X, Wang Y, Tang W, Wang X. Research on the electrical conductivity and mechanical properties of copper slag multiphase nano-modified electrically conductive cementitious composite. Construction and Building Materials. 2022, 339127650.

[45] Shahmansouri AA, Bengar HA, Ghanbari S. Compressive strength prediction of eco-efficient GGBS-based geopolymer concrete using GEP method. Journal of Building Engineering. 2020, 31101326.

[46] Sheta A, Ma X, Zhuge Y, ElGawady MA, Mills JE, Singh A, Abd-Elaal E-S. Structural performance of novel thin-walled composite cold-formed steel/PE-ECC beams. Thin-Walled Structures. 2021.

[47] Shi Q, Abdel-Aty M, Lee J. A Bayesian ridge regression analysis of congestion's impact on urban expressway safety. Accident Analysis & Prevention. 2016, 88124-137.

[48] Singh A, Wang Y, Zhou Y, Sun J, Xu X, Li Y, Wang X. Utilization of antimony tailings in fiber-reinforced 3D printed concrete: A sustainable approach for construction materials. Construction and Building Materials. 2023, 408: 133689.

[49] Song Y-Y, Ying L. Decision tree methods: applications for classification and prediction. Shanghai Archives of Psychiatry. 2015, 272130

[50] Sui L, Zhong Q, Yu K, Xing F, Li P, Zhou Y. Flexural fatigue properties of ultra-high performance engineered cementitious composites (UHP-ECC) reinforced by polymer fibers. Polymers. 2018.

[51] Sun J, Huang Y, Aslani F, Ma G. Electromagnetic wave absorbing performance of 3D printed wave-shape copper solid cementitious element. Cement and Concrete Composites. 2020, 114103789.

[52] Sun J, Liu S, Ma Z, Qian H, Wang Y, Al-azzani H, Wang X. Mechanical properties prediction of lightweight coal gangue shotcrete. Journal of Building Engineering. 2023, 80108088.

[53] Sun J, Ma Y, Li J, Zhang J, Ren Z, Wang X. Machine learning-aided design and prediction of cementitious composites containing graphite and slag powder. Journal of Building Engineering. 2021, 43102544.

[54] Sun J, Wang X, Zhang J, Xiao F, Sun Y, Ren Z, Wang Y. Multi-objective optimisation of a graphite-slag conductive composite applying a BAS-SVR based model. Journal of Building Engineering. 2021, 44103223.

[55] Sun J, Wang Y, Liu S, Dehghani A, Xiang X, Wei J, Wang X. Mechanical, chemical and hydrothermal activation for waste glass reinforced cement. Construction and Building Materials. 2021, 301124361.

[56] Sun J, Wang Y, Yao X, Ren Z, Zhang G, Zhang C, Wang X. Machine-learning-aided prediction of flexural strength and ASR expansion for waste glass cementitious composite. Applied Sciences. 2021, 11156686.

[57] Sun R, Wang G, Cheng Q, Fu L, Chiang K-W, Hsu L-T, Ochieng WY. Improving GPS code phase positioning accuracy in urban environments using machine learning. IEEE Internet of Things Journal. 2020, 887065-7078.

[58] Sun S, Wang S, Wei Y. A new ensemble deep learning approach for exchange rates forecasting and trading. Advanced Engineering Informatics. 2020, 46: 101160.

[59] Sun Y, Li G, Zhang J, Sun J, Xu J. Development of an ensemble intelligent model for assessing the strength of cemented paste backfill. Advances in Civil Engineering. 2020, 20201-6.

[60] Tang Y, Wang Y, Wu D, Chen M, Pang L, Sun J, Wang X. Exploring temperature-resilient recycled aggregate concrete with waste rubber: An experimental and multi-objective optimization analysis. Reviews on Advanced Materials Science. 2023, 62120230347.

[61] Tang Y, Wang Y, Wu D, Liu Z, Zhang H, Zhu M, Wang X. An experimental investigation and machine learning-based prediction for seismic performance of steel tubular column filled with recycled aggregate concrete. Reviews on Advanced Materials Science. 2022, 611849-872.

[62] Uddin MN, Hossain S. Revolutionizing engineered cementitious composite materials (ECC): The impact of XGBoost-SHAP analysis on polyvinyl alcohol (PVA) based ECC predictions. Low-Carbon Materials and Green Construction. 2024, 2111.

[63] Wang D, Ju Y, Shen H, Xu L. Mechanical properties of high performance concrete reinforced with basalt fiber and polypropylene fiber. Construction and Building Materials. 2019, 197: 464-473.

[64] Wei J, Wu C, Chen Y, Leung CKY. Shear strengthening of reinforced concrete beams with high strength strain-hardening cementitious composites (HS-SHCC). Materials and Structures. 2020.

[65] Wu J-D, Guo L-P, Cao Y-Z, Lyu B-C. Mechanical and fiber/matrix interfacial behavior of ultra-high-strength and high-ductility cementitious composites incorporating waste glass powder. Cement and Concrete Composites. 2022.

[66] Wu J-D, Guo L-P, Qin Y-Y. Preparation and characterization of ultra-high-strength and ultra-high-ductility cementitious composites incorporating waste clay brick powder. Journal of Cleaner Production. 2021.

[67] Xu S-L, Xu H-L, Huang B-T, Li Q-H, Yu K-Q, Yu J-T. Development of ultrahigh-strength ultrahigh-toughness cementitious composites (UHS-UHTCC) using polyethylene and steel fibers. Composites Communications. 2022.

[68] Yao X, Lyu X, Sun J, Wang B, Wang Y, Yang M, Wang X. AI-based performance prediction for 3D-printed concrete considering anisotropy and steam curing condition. Construction and Building Materials. 2023, 375: 130898.

[69] Yoo D-Y, Banthia N. High-performance strain-hardening cementitious composites with tensile strain capacity exceeding 4%: A review. Cement and Concrete Composites. 2022, 125104325.

[70] Yoo D-Y, Oh T, Chun B. Highly ductile ultra-rapid-hardening mortar containing oxidized polyethylene fibers. Construction and Building Materials. 2021.

[71] Yoo D-Y, Oh T, Kang M-C, Kim M-J, Choi H-J. Enhanced tensile ductility and sustainability of high-strength strain-hardening cementitious composites using waste cement kiln dust and oxidized polyethylene fibers. Cement and Concrete Composites. 2021.

[72] Yoo S, Kim S, Kim S, Kang BB. AI-HydRa: Advanced hybrid approach using random forest and deep learning for malware classification. Information Sciences. 2021, 546420-435.

[73] Yu K, Ding Y, Zhang YX. Size effects on tensile properties and compressive strength of engineered cementitious composites. Cement and Concrete Composites. 2020.

[74] Yu K, Guo Y, Zhang YX, Soe K. Magnesium oxychloride cement-based strain-hardening cementitious composite: Mechanical property and water resistance. Construction and Building Materials. 2020.

[75] Yu K-Q, Dai J-G, Lu Z-D, Poon C-S. Rate-dependent tensile properties of ultra-high performance engineered cementitious composites (UHP-ECC). Cement and Concrete Composites. 2018, 93: 218-234.

[76] Yu K-Q, Lu Z-D, Dai J-G, Shah SP. Direct tensile properties and stress–strain model of UHP-ECC. Journal of Materials in Civil Engineering. 2020.

[77] Yu K-Q, Yu J-T, Dai J-G, Lu Z-D, Shah SP. Development of ultra-high performance engineered cementitious composites using polyethylene (PE) fibers. Construction and Building Materials. 2018, 158: 217-227.

[78] Yu K-Q, Zhu W-J, Ding Y, Lu Z-D, Yu J-t, Xiao J-Z. Micro-structural and mechanical properties of ultra-high performance engineered cementitious composites (UHP-ECC) incorporation of recycled fine powder (RFP). Cement and Concrete Research. 2019.

[79] Yuan T-F, Lee J-Y, Yoon Y-S. Enhancing the tensile capacity of no-slump high-strength high-ductility concrete. Cement and Concrete Composites. 2020.

[80] Zhang Z, Yang F, Liu J-C, Wang S. Eco-friendly high strength, high ductility engineered cementitious composites (ECC) with substitution of fly ash by rice husk ash. Cement and Concrete Research. 2020.

[81] Zhang Z, Yuvaraj A, Di J, Qian S. Matrix design of light weight, high strength, high ductility ECC. Construction and Building Materials. 2019, 210: 188-197.

[82] Zhou Y, Xi B, Yu K, Sui L, Xing F. Mechanical properties of hybrid ultra-high performance engineered cementitous composites incorporating steel and polyethylene fibers. Materials (Basel). 2018.

[83] Zhou Y, Zheng S, Huang X, Xi B, Huang Z, Guo M. Performance enhancement of green high-ductility engineered cementitious composites by nano-silica incorporation. Construction and Building Materials. 2021.

[84] Zhou Y, Zhong Q, Xing F, Sui L, Huang Z, Guo M. Influence of cyclic loading on the tensile fracture characteristics of ultra-high performance engineered cementitious composites. Construction and Building Materials. 2020.

[85] Zhu B, Wang Y, Sun J, Wei Y, Ye H, Zhao H, Wang X. An experimental study on the influence of waste rubber particles on the compressive, flexural and impact properties of 3D printable sustainable cementitious composites. Case Studies in Construction Materials. 2023, 19e02607.

[86] Zhu J-X, Xu L-Y, Huang B-T, Weng K-F, Dai J-G. Recent developments in engineered/strain-hardening cementitious composites (ECC/SHCC) with high and ultra-high strength. Construction and Building Materials. 2022, 342127956.

Funding

Enterprise Joint Fund Project of the Hunan Provincial Natural Science Foundation(Grant No. S2023JJQYLH0355)

Science and Technology Program Project of Jiangsu Provincial Market Supervision Administration(Grant No. KJ2025069)

RIGHTS & PERMISSIONS

The Author(s)