Few-shot meta-learning for concrete strength prediction: a model-agnostic approach with SHAP analysis

Abstract

Predicting concrete compressive strength with limited data remains a critical challenge in civil engineering. This study proposes a novel framework integrating Model-Agnostic Meta-Learning (MAML) with SHAP (Shapley Additive Explanations) to improve predictive accuracy and interpretability in data-scarce scenarios. Unlike conventional machine learning models that require extensive data, the MAML-based approach enables rapid adaptation to new tasks using minimal samples, offering robust generalization in few-shot learning contexts. The proposed pipeline includes structured preprocessing, normalization, a neural network-based meta-learning core, and SHAP-based feature attribution. A curated dataset of 430 samples was used, focusing on 28-day compressive strength, with input features including cement, water, aggregates, admixtures, and age. Compared to standard models like XGBoost and Random Forest, the MAML framework achieved superior performance, with MAE = 3.56 MPa, RMSE = 5.55 MPa, and R² = 0.913. SHAP analysis revealed nonlinear interactions and dominant factors like water-cement ratio, curing age, and aggregate content. Statistical validation via the Wilcoxon Signed-Rank Test confirmed the significance of the model’s improvements (p < 0.05). Furthermore, SHAP insights closely align with domain knowledge and mix design principles, enhancing model transparency for practical application. This work demonstrates the applicability of meta-learning in civil engineering and provides a scalable, interpretable solution for strength prediction in real-world, data-limited conditions.

Keywords

Meta-learning / Few-shot learning / SHAP interpretability / Predictive analytics / Machine learning in civil engineering / Sustainable construction / Data-efficient modeling

References

[1] AhmadA, FarooqF, NiewiadomskiP, OstrowskiK, AkbarA, AslamF, AlyousefR. Prediction of compressive strength of fly ash based concrete using individual and ensemble algorithm. Materials, 2021, 144794.

[2] AhmadA, OstrowskiKA, MaślakM, FarooqF, MehmoodI, NafeesA. Comparative study of supervised machine learning algorithms for predicting the compressive strength of concrete at high temperature. Materials, 2021, 14154222.

[3] AirlanggaG. Comparison of predictive modeling concrete compressive strength with machine learning approaches. U Karst, 2024.

[4] Al-HashemMN, AminMN, RaheelM, KhanK, AlkadhimHA, ImranM, UllahS, IqbalM. Predicting the compressive strength of concrete containing fly ash and rice husk ash using ANN and GEP models. Materials, 2022, 15217713.

[5] AliPJM. Investigating the impact of min-max data normalization on the regression performance of K-nearest neighbor with different similarity measurements. ARO-the Scientific Journal of Koya University, 2022, 10(1): 85-91.

[6] Al-MughanamT, AldhyaniTHH, AlsubariB, Al-YaariM. Modeling of compressive strength of sustainable self-compacting concrete incorporating treated palm oil fuel ash using artificial neural network. Sustainability, 2020, 12229322.

[7] AlyamiMG, KhanM, FawadM, NawazR, HammadAWA, NajehT, GamilY. Predictive modeling for compressive strength of 3D printed fiber-reinforced concrete using machine learning algorithms. Case Studies in Construction Materials, 2023.

[8] AslamF, FarooqF, AminMN, KhanK, WaheedA, AkbarA, JavedMF, AlyousefR, AlabdulijabbarH. Applications of gene expression programming for estimating compressive strength of high-strength concrete. Advances in Civil Engineering, 2020, 202018850535.

[9] AslamF, ShahabMZ. Supplementary cementitious materials in blended cement concrete: Advancements in predicting compressive strength through machine learning. Materials Today Communications, 2023.

[10] AydınY, CakirogluC, BekdaşG, GeemZW. Explainable ensemble learning and multilayer perceptron modeling for compressive strength prediction of ultra-high-performance concrete. Biomimetics, 2024.

[11] BakhoumES, AmirA, OsamaF, AdelM. Prediction model for the compressive strength of green concrete using cement kiln dust and fly ash. Dental Science Reports, 2023.

[12] BendichP, GasparovicE, HarerJ, IzmailovR, NessL. Multi-scale local shape analysis and feature selection in machine learning applications. Proceedings of the International Joint Conference on Neural Networks, 2015.

[13] BorrohouS, FissouneR, BadirH. Data cleaning survey and challenges – improving outlier detection algorithm in machine learning. Journal of Smart Cities and Society, 2023, 2(3): 125-140.

[14] BotchkarevA. Performance metrics (Error Measures) in machine learning regression, forecasting and prognostics: properties and typology. Interdisciplinary Journal of Information, Knowledge, and Management, 2018, 14: 45-76.

[15] BuiDK, NguyenT, ChouJS, Nguyen-XuanH, NgoTD. 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.

[16] Cayce GI, II AC Depoian, Bailey CP, Guturu P. (2023). Ethical data splitting utilizing L1-norm principal component analysis. 12522: 21–26. https://doi.org/10.1117/12.2664127

[17] ChenX, ZhangY, ChenW. Advanced predictive modeling of concrete compressive strength and slump characteristics: A comparative evaluation of BPNN, SVM, and RF models optimized via PSO. Materials, 2024.

[18] CitraZ, WibowoPD, MalindaY, WibisonoA, ApdeniR, HerolH. Testing of concrete structures with non-destructive test method (NDT) using ultrasonic pulse velocity (UPV) at the Building on the Ancol Beach. CIVED, 2024, 11(1): 217-225.

[19] Compressive strength testing of hpc cylinders utilizing confined caps. (1995). https://typeset.io/papers/compressive-strength-testing-of-hpc-cylinders-utilizing-36vrfa4pnx

[20] Concrete strength prediction. (n.d.).https://www.kaggle.com/code/shrikantchine/concrete-strength-prediction/notebook. Accessed 7 Jan 2025

[21] da PaixãoRCF, PenidoRE-K, CuryA, MendesJC. Comparison of machine learning techniques to predict the compressive strength of concrete and considerations on model generalization. Revista IBRACON De Estruturas e Materiais, 2022.

[22] DasP, KashemA. Hybrid machine learning approach to prediction of the compressive and flexural strengths of UHPC and parametric analysis with shapley additive explanations. Case Studies in Construction Materials, 2024, 20. e02723

[23] de Prado-GilJ, Martínez-GarcíaR, FernándezFJ, PalenciaCDesign of mixtures and manufacture of self-compacting concretes with recycled aggregates (Eco-Concretes): Prediction of compressive strength using machine learning models, 2024Springer Nature Switzerland.

[24] DedhiaA. Comparative analysis of ensemble learning techniques and ann for prediction of compressive strength of concrete mix. International Journal for Research in Applied Science and Engineering Technology, 2024.

[25] EltrissiM, YousefO. Predicting system surface parameters using artificial neural network. ADIPEC, 2024.

[26] FengDC, LiuZT, WangXD, ChenY, ChangJQ, WeiDF, JiangZM. Machine learning-based compressive strength prediction for concrete: An adaptive boosting approach. Construction and Building Materials, 2020, 230. 117000

[27] Finn, C., Abbeel, P., & Levine, S. (2017b). Model-agnostic meta-learning for fast adaptation of deep networks. 34th International Conference on Machine Learning. 3: 1856–1868. https://arxiv.org/abs/1703.03400v3

[28] Finn, C., Abbeel, P., & Levine, S. (2017a). Model-agnostic meta-learning for fast adaptation of deep networks. 34th International Conference on Machine Learning. 3, 1856–1868. https://arxiv.org/abs/1703.03400v3

[29] FinnC, XuK, LevineS. Probabilistic model-agnostic meta-learning. Advances in Neural Information Processing Systems., 2018, 31: 9516-9527

[30] Gil-MartínLM, Hdz-GilL, MoleroE, Hernández-MontesE. The relationship between concrete strength and classes of resistance against corrosion induced by carbonation: A proposal for the design of extremely durable structures in accordance with eurocode 2. Sustainability, 2023, 15107976.

[31] GolafshaniEM, BehnoodA, ArashpourM. Predicting the compressive strength of normal and high-performance concretes using ANN and ANFIS hybridized with grey wolf optimizer. Construction and Building Materials, 2020, 232. 117266

[32] HaqueMA, ChenB, KashemA, QureshiT, AhmedAAM. Hybrid intelligence models for compressive strength prediction of MPC composites and parametric analysis with SHAP algorithm. Materials Today Communications, 2023, 35. 105547

[33] HidayatI, RahardhaniAFS, SuhardjoGA. Utilizing hammer tests and ultrasonic pulse velocity to ascertain the compressive strength of concrete. IOP Conference Series: Earth and Environmental Science, 2024, 13241. 012006

[34] Hong, Y. M., & Sah, A. K. (2024). Estimation of Concrete Compressive Strength Using Machine Learning. https://doi.org/10.20944/PREPRINTS202404.0427.V1

[35] HuangJ, SabriMMS, UlrikhDV, AhmadM, AlsaffarKAM. Predicting the compressive strength of the cement-fly ash-slag ternary concrete using the firefly algorithm (FA) and random forest (RF) hybrid machine-learning method. Materials, 2022, 15124193.

[36] IslamN, KashemA, DasP, AliMN, PaulS. Prediction of high-performance concrete compressive strength using deep learning techniques. Asian Journal of Civil Engineering, 2024, 25: 327-341.

[37] KarimR, IslamMH, DattaSD, KashemA. Synergistic effects of supplementary cementitious materials and compressive strength prediction of concrete using machine learning algorithms with SHAP and PDP analyses. Case Studies in Construction Materials, 2024, 20. e02828

[38] KashemA, DasP. Compressive strength prediction of high-strength concrete using hybrid machine learning approaches by incorporating SHAP analysis. Asian Journal of Civil Engineering, 2023, 24(8): 3243-3263.

[39] KashemA, KarimR, DasP, DattaSD, AlharthaiM. Compressive strength prediction of sustainable concrete incorporating rice husk ash (RHA) using hybrid machine learning algorithms and parametric analyses. Case Studies in Construction Materials, 2024, 20. e03030

[40] KatlavM, ErgenF. Improved forecasting of the compressive strength of ultra-high-performance concrete (UHPC) via the CatBoost model optimized with different algorithms. Structural Concrete, 2024.

[41] LiD, TangZ, KangQ, ZhangX, LiY. Machine learning-based method for predicting compressive strength of concrete. Processes, 2023, 112390.

[42] LiT, HuX, LiT, LiaoJ, MeiL, TianH, GuJ. Enhanced prediction and evaluation of hydraulic concrete compressive strength using multiple soft computing and metaheuristic optimization algorithms. Buildings, 2024, 14113461.

[43] LiY, YangX, RenC, WangL, NingX. Predicting the compressive strength of ultra-high-performance concrete based on machine learning optimized by meta-heuristic algorithm. Buildings, 2024, 1451209.

[44] MahmoodMS, ElahiA, ZaidO, AlashkerY, ȘerbănoiuAA, GrădinaruCM, UllahK, AliT. Enhancing compressive strength prediction in self-compacting concrete using machine learning and deep learning techniques with incorporation of rice husk ash and marble powder. Case Studies in Construction Materials, 2023.

[45] MandalA. Predicting compressive strength of concrete using advanced machine learning techniques: A combined dataset approach. Asian Journal of Civil Engineering, 2024.

[46] NalinaMM. Efficacies of suggested strength-based prediction models for estimation of compressive and tensile properties of normal concrete. Challenge Journal of Concrete Research Letters, 2023, 14(2): 47-58.

[47] NaserMZ, AlaviAH. Error metrics and performance fitness indicators for artificial intelligence and machine learning in engineering and sciences. Architecture, Structures and Construction, 2021, 3(4): 499-517.

[48] Nikoopayan TakMS, FengY, MahgoubM. Advanced machine learning techniques for predicting concrete compressive strength. Infrastructures, 2025, 10226.

[49] NoguchiT, NematiKM. Relationship between compressive strength and modulus of elasticity of high strength concrete. Journal of Structural and Construction Engineering (Transactions of Aij), 1995, 3: 1305-1311.

[50] PaulS, DasP, KashemA, IslamN. Sustainable of rice husk ash concrete compressive strength prediction utilizing artificial intelligence techniques. Asian Journal of Civil Engineering, 2024, 25(2): 1349-1364.

[51] PillayAB, PathmanathanD, AbuA, OmarH. RFE-based feature selection to improve classification accuracy for morphometric analysis of craniodental characters of house rats. Sains Malaysiana, 2023, 52(7): 1901-1914.

[52] Prediction of Concrete Strength Based on Random Forest and Gradient Boosting Machine. (2023). https://doi.org/10.1109/icpeca56706.2023.10075839

[53] SahAK, HongYM. Performance comparison of machine learning models for concrete compressive strength prediction. Materials, 2024, 1792075.

[54] SaudAF, AbdelgaderHS, El-BadenAS. Compressive and tensile strength of two-stage concrete. Advanced Materials Research, 2014, 893: 585-592.

[55] ShabanWM, ElbazK, YangJ, ShenSL. A multi-objective optimization algorithm for forecasting the compressive strength of RAC with pozzolanic materials. Journal of Cleaner Production, 2021, 327. 129355

[56] SharkoJ, GrinsteinGG, MarxKA, ZhouJ, ChengCH, OdelbergS, SimonHG. Heat map visualizations allow comparison of multiple clustering results and evaluation of dataset quality: Application to microarray data. Proceedings of the International Conference on Information Visualisation, 2007.

[57] ShiY, CaoY, ChenY, ZhangL. Meta learning based residual network for industrial production quality prediction with limited data. Scientific Reports, 2024, 14(1): 1-13.

[58] Smith, K., & Climer, S. (2023). Mr. Clean: An ensemble of data cleaning algorithms for increased data retention. Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023. https://doi.org/10.1109/BIBM58861.2023.10385522

[59] SongY, ZhaoJ, OstrowskiKA, JavedMF, AhmadA, KhanMI, AslamF, KinaszR. Prediction of compressive strength of Fly-Ash-based concrete using ensemble and non-ensemble supervised machine-learning approaches. Applied Sciences, 2021, 121361.

[60] Sunata, K., Kume, T., Nakazawa, J., & Okoshi, T. (2024). Poster: Feature-adaptive Re-MAML optimised for the input data set. MOBISYS 2024 - Proceedings of the 2024 22nd Annual International Conference on Mobile Systems, Applications and Services. https://doi.org/10.1145/3643832.3661441

[61] TakJH, HongBW. Enhancing model agnostic meta-learning via gradient similarity loss. Electronics, 2024.

[62] TakM, FengY, MahgoubM. Advanced machine learning techniques for predicting concrete compressive strength. Infrastructures, 2025.

[63] TANASHO, MuthusamyK. Concrete industry, environment issue, and green concrete: A review. Construction, 2022, 2(1): 01-09.

[64] TipuRK. Enhancing prediction accuracy of workability and compressive strength of high-performance concrete through extended dataset and improved machine learning models. Asian Journal of Civil Engineering, 2024, 25(1): 197-218.

[65] TizaMT, ImoniS, AkandeEO, MogboO, JiyaVH, OnuzulikeC. Revolutionizing infrastructure development: Exploring cutting-edge advances in civil engineering materials. Recent Progress in Materials., 2024, 6(3): 1-68.

[66] Using Explainable Machine Learning to Predict Compressive Strength of Blended Concrete: A Data-Driven Metaheuristic Approach. (2023). https://doi.org/10.21203/rs.3.rs-2998256/v1

[67] WakjiraTG, KuttyAA, AlamMS. A novel framework for developing environmentally sustainable and cost-effective ultra-high-performance concrete (UHPC) using advanced machine learning and multi-objective optimization techniques. Construction and Building Materials, 2024.

[68] WangM, KangJ, LiuW, SuJ, LiM. Research on prediction of compressive strength of fly ash and slag mixed concrete based on machine learning. PLoS ONE, 2022, 1712. e0279293

[69] YadalamPK, AnegundiRV, Alarcón-SánchezMA, HeboyanA. Classification and detection of dental images using meta-learning. World Journal of Clinical Cases, 2024, 12(32): 6559-6562.

[70] YuY, LiW, LiJ, NguyenTN. A novel optimised self-learning method for compressive strength prediction of high performance concrete. Construction and Building Materials, 2018, 184: 229-247.

[71] ZhangJ, ZhaoY, LiH. Experimental investigation and prediction of compressive strength of ultra-high performance concrete containing supplementary cementitious materials. Advances in Materials Science and Engineering, 2017, 201714563164.

[72] ZhangY, RenW, ChenY, MiY, LeiJ, SunL. Predicting the compressive strength of high-performance concrete using an interpretable machine learning model. Scientific Reports., 2024, 14(1): 1-18.

RIGHTS & PERMISSIONS

The Author(s)