Toward Reliable Compressive Strength Prediction of Marble Powder-Blended Concrete: Comparative Machine Learning and SHAP-Based Interpretability

Authors

• Shamal Abdullah Mohammed, Department of Civil and Environmental Engineering, Faculty of Engineering, Soran University, Soran 44008, Kurdistan Region, Iraq.
• Aryan Far H. Sherwani, Scientific Research Center, Soran University, Soran 44008, Kurdistan Region, IraqFollow
• Nzar Shakr Piro, Mechatronic and Robotics Engineering Department, Soran University, Soran 44008, Kurdistan Region, Iraq

How to Cite This Article

Mohammed, Shamal Abdullah; Sherwani, Aryan Far H.; and Piro, Nzar Shakr (2026) "Toward Reliable Compressive Strength Prediction of Marble Powder-Blended Concrete: Comparative Machine Learning and SHAP-Based Interpretability," Polytechnic Journal: Vol. 16: Iss. 2, Article 2.
DOI: https://doi.org/10.59341/2707-7799.1881

Document Type

Original Article

Abstract

The production of conventional concrete has a significant impact on the environment, primarily due to the cement (C) component. To mitigate this impact, binders with reduced environmental footprints were used in concrete mixtures as a partial replacement for C, including powders of industrial byproducts. Waste marble powder (WMP) is one such alternative; however, its incorporation causes complicated interactions with cementitious materials, making the compressive strength (CS) prediction difficult. This study developed and evaluated four predictive models (multilinear regression (MLR), full quadratic regression (FQ), M5P tree, and artificial neural network (ANN)) using a dataset of 306 WMP-blended concrete mixes from 16 published studies. Statistical evaluation parameters (mean absolute error (MAE), root mean squared error (RMSE), coefficient of determination (R²), scatter index (SI), and objective function (OBJ)) were used to assess the performance of models. The ANN model achieved superior performance on the testing subset (R²=0.96, RMSE=2.95 MPa), while 10-fold cross-validation confirmed its predictive robustness, yielding mean and RMSE values of 0.93 0.04 and 3.54 1.17 MPa, respectively. The SHAP (SHapley Additive exPlanations) method of sensitivity analysis illustrated that C and curing time (t) were the main factors affecting CS, while WMP had a negative context-dependent effect at higher levels of replacement. The results indicated that ANN models can successfully capture the nonlinear behavior of WMP-modified concrete and can be used as a practical tool for optimizing concrete mix designs for sustainability.

Receive Date

15 Apr 2026

Revise Date

15 July 2026

Accept Date

25 July 2026

Publication Date

9-4-2026

References

[1] Ahmad SA, Ahmed HU, Rafiq SK, Jafer FS, Fqi KO. A comparative analysis of simulation approaches for predicting permeability and compressive strength in pervious concrete. Low-Carbon Mater Green Constr 2024;2. https:// doi.org/10.1007/s44242-024-00041-x.

[2] Lin CJ, Wu NJ. An ann model for predicting the compressive strength of concrete. Appl Sci 2021;11. https://doi.org/ 10.3390/app11093798.

[3] Vargas JF, Oviedo AI, Ortega NA, Orozco E, G omez A, Londo~no JM. Machine-learning-based predictive models for compressive strength, flexural strength, and slump of concrete. Appl Sci 2024;14. https://doi.org/10.3390/app14114426.

[4] Kumar A, Arora HC, Kapoor NR, Mohammed MA, Kumar K, Majumdar A, et al. Compressive strength prediction of lightweight concrete: machine learning models. Sustainability 2022;14. https://doi.org/10.3390/su14042404.

[5] Chung KL, Wang L, Ghannam M, Guan M, Luo J. Prediction of concrete compressive strength based on early-age effective conductivity measurement. J Build Eng 2021;35. https:// doi.org/10.1016/j.jobe.2020.101998.

[6] Peng X, Zhuang Z, Yang Q. Predictive modeling of compressive strength for concrete at super early age. Materials 2022;15. https://doi.org/10.3390/ma15144914.

[7] Latif SD. Concrete compressive strength prediction modeling utilizing deep learning long short-term memory algorithm for a sustainable environment. Environ Sci Pollut Control Ser 2021;28:30294—302. https://doi.org/10.1007/ s11356-021-12877-y.

[8] Philip S, Nidhi M, Nakkeeran G. Soft computing techniques for predicting the compressive strength properties of fly ash geopolymer concrete using regression-based machine learning approaches. J Build Pathol Rehabil 2024;9. https:// doi.org/10.1007/s41024-024-00461-z.

[9] Sathiparan N, Jeyananthan P. Soft computing techniques to predict the compressive strength of groundnut shell ash- blended concrete. J Eng Appl Sci 2023;70. https://doi.org/ 10.1186/s44147-023-00302-9.

[10] Gogineni A, Panday IK, Kumar P, Paswan R kr. Predictive modelling of concrete compressive strength incorporating GGBS and alkali using a machine-learning approach. Asian J Civil Eng 2024;25:699—709. https://doi.org/10.1007/s42107- 023-00805-z.

[11] Murad Y, Imam R, Abu Hajar H, Habeh D, Hammad A, Shawash Z. Predictive compressive strength models for green concrete. Int J Struct Integr 2020;11:169—84. https:// doi.org/10.1108/IJSI-05-2019-0044.

[12] Gacu JG, Sim AAM. Effect of marble microparticles as additive on the physical and mechanical properties of concrete mixes. Mater Today Proc 2022;65:1491—7. https://doi.org/ 10.1016/j.matpr.2022.04.469. Elsevier Ltd.

[13] Sharma N, Thakur MS, Upadhya A, Sihag P. Machine learning algorithms for evaluating concrete strength using marble powder. IOP Conf Ser Earth Environ Sci 2023;1110. https:// doi.org/10.1088/1755-1315/1110/1/012058. Institute of Physics.

[14] Gamil Y. Machine learning in concrete technology: A review of current researches, trends, and applications. Front Built Environ 2023;9. https://doi.org/10.3389/fbuil.2023.1145591.

[15] Piro NS, Mohammed AS, Hamad SM, Kurda R, Qader BS. Multifunctional computational models to predict the long- term compressive strength of concrete incorporated with waste steel slag. Struct Concr 2023;24:2093—112. https:// doi.org/10.1002/suco.202200023.

[16] Zrar YJ, Abdulrahman PI, Sherwani AFH, Younis KH, Mohammed AS. Sustainable innovation in self-compacted concrete: Integrating by-products and waste rubber for green construction practices. Structures 2024;62. https:// doi.org/10.1016/j.istruc.2024.106234.

[17] Gupta S, Sihag P. Prediction of the compressive strength of concrete using various predictive modeling techniques. Neural Comput Appl 2022;34:6535—45. https://doi.org/ 10.1007/s00521-021-06820-y.

[18] Wang W, Zhong Y, Liao G, Ding Q, Zhang T, Li X. Prediction of Compressive Strength of Concrete Specimens Based on Interpretable Machine Learning. Materials 2024;17. https://doi.org/10.3390/ma17153661.

[19] Wang Y, Iqtidar A, Amin MN, Nazar S, Hassan AM, Ali M. Predictive modelling of compressive strength of fly ash and ground granulated blast furnace slag based geopolymer concrete using machine learning techniques. Case Stud Constr Mater 2024;20. https://doi.org/10.1016/j.cscm.2024.e03130.

[20] Sherwani AFH, Abdulrahman PI, Younis KH, Mohammed AS, Zrar YJ. Compressive strength modeling of the eco-friendly self-compacting rubberized concrete with fly ash and ground granulated blast furnace slag additives. Innov Infrastruct Solut 2025;10. https://doi.org/10.1007/s41062-024-01811-z.

[21] Kumar M, Biswas R, Kumar DR, Samui P, Kaloop MR, Eldessouki M. Soft computing-based prediction models for compressive strength of concrete. Case Stud Constr Mater 2023;19. https://doi.org/10.1016/j.cscm.2023.e02321.

[22] Alzlfawi A, Sobuz MHR, Kawsarul M, Kabbo I, Khatun M, Abubakar SA, et al. Prediction and parametric modeling of compressive strength of waste marble dust concrete through machine learning and experimental analysis. Sci Rep 2025;15(15):42982. https://doi.org/10.1038/s41598-025- 26978-y. 1 2025.

[23] Onyelowe KC, Kamchoom V, Ebid AM, Hanandeh S, Zurita Polo SM, Zabala Vizuete RF, et al. Evaluating the impact of waste marble on the compressive strength of traditional concrete using machine learning. Sci Rep 2025;15(15):13417. https://doi.org/10.1038/s41598-025-98431-z. 1 2025.

[24] Abbas MM, Muntean R. Predicting mechanical properties of marble powder concrete using artificial neural networks and blockchain-rock for sustainable construction. Front Built Environ 2025;11:1594735. https://doi.org/10.3389/ FBUIL.2025.1594735/TEXT.

[25] Kurniati EO, Zeng H, Latypov MI, Kim HJ. Machine learning for predicting compressive strength of sustainable cement paste incorporating copper mine tailings as supplementary cementitious materials. Case Stud Constr Mater 2024;21:e03373. https://doi.org/10.1016/J.CSCM.2024.E03373.

[26] Tan X, Xing J, Wang Y, Qiu H, Mahjoubi S, Guo P. Explainable machine learning for predicting compressive strength of rubberized concrete: SHAP interpretation, lifecycle assessment, and design recommendations. J Clean Prod 2026;538:147338. https://doi.org/10.1016/J.JCLEPRO. 2025.147338.

[27] Deb S, Mahajan Y, Sharma V, Kapse M, Ghosal A. Machine learning-based optimization of sustainable concrete mix for strength prediction. Discov Sustain 2026;7(7):989. https:// doi.org/10.1007/S43621-026-03227-4. 1 2026.

[28] € Ozkõlõç YO, Zeybek € O, Bahrami A, Çelik A_ I, Othuman Mydin MA, Karalar M, et al. Optimum usage of waste marble powder to reduce use of cement toward eco-friendly concrete. J Mater Res Technol 2023;25:4799—819. https:// doi.org/10.1016/j.jmrt.2023.06.126.

[29] Babouri L, Biskri Y, Khadraoui F, El Mendili Y. Mechanical performance and corrosion resistance of reinforced concrete with marble waste. Eur J Environ Civil Eng 2022;26:4112—29. https://doi.org/10.1080/19648189.2020.1838952.

[30] Singh M, Srivastava A, Bhunia D. Long term strength and durability parameters of hardened concrete on partially replacing cement by dried waste marble powder slurry. Constr Build Mater 2019;198:553—69. https://doi.org/10.1016/ j.conbuildmat.2018.12.005.

[31] Parmar S, Das SK. Effect of marble slurry as a partial substitution of ordinary Portland cement in lean concrete mixes. Struct Concr 2021;22:E514—27. https://doi.org/ 10.1002/suco.202000059.

[32] Majeed M, Khitab A, Anwar W, Khan RBN, Jalil A, Tariq Z. Evaluation of concrete with partial replacement of cement by waste marble powder. Civil Eng J 2021;7:59—70. https:// doi.org/10.28991/cej-2021-03091637.

[33] El-Mandouh MA, Hu JW, Mohamed AS, Abd El-Maula AS. Assessment of Waste Marble Powder on the Mechanical Properties of High-Strength Concrete and Evaluation of Its Shear Strength. MDPI 2022;15. https://doi.org/10.3390/ ma15207125.

[34] Ergün A. Effects of the usage of diatomite and waste marble powder as partial replacement of cement on the mechanical properties of concrete. Constr Build Mater 2011;25:806—12. https://doi.org/10.1016/j.conbuildmat.2010.07.002.

[35] Khodabakhshian A, de Brito J, Ghalehnovi M, Asadi Shamsabadi E. Mechanical, environmental and economic performance of structural concrete containing silica fume and marble industry waste powder. Constr Build Mater 2018;169: 237—51. https://doi.org/10.1016/j.conbuildmat.2018.02.192.

[36] Mashaly AO, El-Kaliouby BA, Shalaby BN, El-Gohary AM, Rashwan MA. Effects of marble sludge incorporation on the properties of cement composites and concrete paving blocks. J Clean Prod 2015;112:731—41. https://doi.org/ 10.1016/j.jclepro.2015.07.023.

[37] Wang Y, Xiao J, Zhang J, Duan Z. Mechanical behavior of concrete prepared with waste marble powder. Sustainability 2022;14. https://doi.org/10.3390/su14074170.

[38] Rana A, Kalla P, Csetenyi LJ. Sustainable use of marble slurry in concrete. J Clean Prod 2015;94:304—11. https:// doi.org/10.1016/j.jclepro.2015.01.053.

[39] Belouadah M, Abidine Rahmouni Z El, Tebbal N. Experimental characterization of ordinary concretes obtained by adding construction waste (glass, marble). Procedia Comput Sci 2019;158:153—62. https://doi.org/10.1016/j.procs.2019.09 .038. Elsevier B.V.

[40] Singh M, Srivastava A, Bhunia D. An investigation on effect of partial replacement of cement by waste marble slurry. Constr Build Mater 2017;134:471—88. https://doi.org/10.1016/ j.conbuildmat.2016.12.155.

[41] Aliabdo AA, Abd Elmoaty AEM, Auda EM. Re-use of waste marble dust in the production of cement and concrete. Constr Build Mater 2014;50:28—41. https://doi.org/10.1016/ j.conbuildmat.2013.09.005.

[42] Afzal Basha S, Shaikh FUA. Suitability of marble powders in production of high strength concrete. Low-Carbon Mater Green Constr 2023;1. https://doi.org/10.1007/s44242-023- 00029-z.

[43] Azougay A, Filali S, Miraoui Z, Nasser A. Utilizing recycled marble powder dust to part of the cement in concrete, 19; 2024. https://doi.org/10.59018/042457.

[44] El-Mandouh MA, Hu JW, Mohamed AS, Abd El-Maula AS. Assessment of Waste Marble Powder on the Mechanical Properties of High-Strength Concrete and Evaluation of Its Shear Strength. Materials 2022;15. https://doi.org/10.3390/ ma15207125.

[45] Mohammed AS, Omer MAI, Haider RS, Abd NT, Mohammed AS. Predicting the Compressive Strength of Fly Ash Composite Foam Concrete Using Artificial Neural Networks and Soft Computing Techniques. Bulletin of Comput Intell 2026;2:83—102. https://doi.org/10.53941/ bci.2026.100005.

[46] Aree Jamal R, Khaleefa T, Ali M, Saeed S, Mohammed AS. Soft computing approaches for tensile strength prediction of sustainable concrete mixtures A R T I C L E I N F O, 12; 2025.

[47] Ali BA, Mohammed AS, Salih MA, Salih SH, Omer M, Kurda R, et al. Artificial neural network and soft computing models to predict the compressive strength in self-compacting green concrete. Innov Infrastruct Solut 2024;9. https://doi.org/10.1007/s41062-024-01723-y.

[48] Ahmed ZQ, Soran Othma S, Salih Mohammed A, Hassan AMT, Asteris PG. Predicting the compressive strength of UHPFRC using machine learning and soft computing models: optimization of fiber content and additives ARTICLEINFO, 13; 2026.

[49] Specification for Portland cement. 2024. https://doi.org/ 10.1520/C0150_C0150M-24.

[50] British Standard Institution. Cement. Composition, specifications and conformity criteria for common cements. Bsi. 2019. 2011.

[51] Bhawan M, Shah B, Marg Z. PM/IS 383/2/may 2020 bureau of INDIAN standards product manual for coarse and fine aggregate for concrete according to IS 383 : 2016. n.d..

[52] Specification for concrete aggregates. 2018. https://doi.org/ 10.1520/C0033_C0033M-18.

[53] Specification for chemical admixtures for concrete. https:// doi.org/10.1520/C0494_C0494M-24; 2024.

[54] Specification for mixing water used in the production of hydraulic cement concrete. https://doi.org/10.1520/C1602_ C1602M-05; 2005.

[55] Sathiparan N. Prediction model for compressive strength of rice husk ash blended sandcrete blocks using a machine learning models. https://doi.org/10.21203/rs.3.rs-4373207/v1; 2024.

[56] Zhang X, Akber MZ, Zheng W. Prediction of seven-day compressive strength of field concrete. Constr Build Mater 2021;305. https://doi.org/10.1016/j.conbuildmat.2021.124604.

[57] Ahmed HU, Abdalla AA, Mohammed AS, Mohammed AA. Mathematical modeling techniques to predict the compressive strength of high-strength concrete incorporated metakaolin with multiple mix proportions. Clean Mater 2022;5. 0132.

[58] Sathiparan N, https://doi.org/10.1016/j.clema.2022.10 Wijekoon SH, Jeyananthan P, Subramaniam DN. Prediction of characteristics of pervious concrete by machine learning technique using mix parameters and non-destructive test measurements. Nondestr Test Eval 2468273. 2025. https://doi.org/10.1080/10589759.2025.

[59] Terlumun S, Onyia ME, Okafor FO. Predicting the compressive strength of self-compacting concrete using artificial intelligence techniques: a review. AI Civil Eng 2024; 3. https://doi.org/10.1007/s43503-024-00029-3.

[60] Sathiparan N, Jeyananthan P, Subramaniam DN. A machine learning approach to predicting pervious concrete properties: a review. Innov Infrastruct Solut 2025;10. https://doi.org/10.1007/s41062-024-01829-3.

[61] Hoang ND, Tran VD, Tran XL. Predicting compressive strength of high-performance concrete using hybridization of nature-inspired metaheuristic and gradient boosting machine. Mathematics 2024;12. https://doi.org/10.3390/ math12081267.

[62] Willmott CJ, Matsuura K. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Clim Res 2005;30: 79—82. https://doi.org/10.3354/CR030079.

[63] Hyndman RJ, Koehler AB. Another look at measures of forecast accuracy. Int J Forecast 2006;22:679—88. https:// doi.org/10.1016/j.ijforecast.2006.03.001.

[64] Amlashi AT, Ghanizadeh AR, Firouzranjbar S, Moghaddam HM, Navazani M, Isleem HF, et al. Predicting workability and mechanical properties of bentonite plastic concrete using hybrid ensemble learning. Sci Rep 2025;15. https://doi.org/10.1038/s41598-025-92253-9.

[65] Kakasor Ismael Jaf D, Ismael Abdulrahman P, Salih Mohammed A, Kurda R, Qaidi SMA, Asteris PG. Machine learning techniques and multi-scale models to evaluate the impact of silicon dioxide (SiO2) and calcium oxide (CaO) in fly ash on the compressive strength of green concrete. Constr Build Mater 2023;400. https://doi.org/10.1016/ j.conbuildmat.2023.132604.

[66] Emarah DA. Multivariate predictive modeling of compressive strength in ground granulated blast furnace slag/fly ash-based alkali-activated concrete. Clean Eng Technol 2025;27. https://doi.org/10.1016/j.clet.2025.101021.

[67] Sah AK, Hong YM. Performance comparison of machine learning models for concrete compressive strength prediction. Materials 2024;17. https://doi.org/10.3390/ma17092075.

[68] Mouawad F, Homsi F, Geara F, Mina R. Predicting compressive strength of sustainable concrete using machine learning and artificial neural networks. Constr Mater 2025;5. https://doi.org/10.3390/CONSTRMATER5030056.

[69] Wan Z, Xu Y,  Savija B. On the use of machine learning models for prediction of compressive strength of concrete: Influence of dimensionality reduction on the model performance. Materials 2021;14:1—23. https://doi.org/10.3390/ ma14040713.

[70] Shah HA, Nehdi ML, Khan MI, Akmal U, Alabduljabbar H, Mohamed A, et al. Predicting Compressive and Splitting Tensile Strengths of Silica Fume Concrete Using M5P Model Tree Algorithm. Materials 2022;15. https://doi.org/ 10.3390/ma15155436.

[71] Fathy IN, Dahish HA, Alkharisi MK, Mahmoud AA, Fouad HEE. Predicting the compressive strength of concrete incorporating waste powders exposed to elevated temperatures utilizing machine learning. Sci Rep 2025;15. https:// doi.org/10.1038/s41598-025-11239-9.

[72] Lundberg S, Lee S-I. A unified approach to interpreting model predictions. 2017.

[73] Mangalathu S, Karthikeyan K, Feng DC, Jeon JS. Machine- learning interpretability techniques for seismic performance assessment of infrastructure systems. Eng Struct 2022;250:112883. https://doi.org/10.1016/J.ENGSTRUCT.2021 .112883.

[74] Abioye SO, Babatunde YO, Abikoye OA, Shaibu AN, Bankole BJ. Optimized machine learning algorithms with SHAP analysis for predicting compressive strength in high- performance concrete. AI Civil Eng 2025;4. https://doi.org/ 10.1007/s43503-025-00061-x.

[75] Neville AM. Properties of concrete-5th edition. Concrete mix design. Qual Control Specif 2011:28—40.

[76] Islam MM, Das P, Rahman MM, Naz F, Kashem A, Nishat MH, et al. Prediction of compressive strength of high-performance concrete using optimization machine learning approaches with SHAP analysis. J Build Pathol Rehabil 2024;9. https://doi.org/10.1007/s41024-024-00445-z.

[77] Feng DC, Liu ZT, Wang XD, Chen Y, Chang JQ, Wei DF, et al. Machine learning-based compressive strength prediction for concrete: An adaptive boosting approach. Constr Build Mater 2020;230. https://doi.org/10.1016/j.conbuildmat .2019.117000.

[78] Shamal A, Rekani M. GUI application for predicting the compressive strength of marble powder-blended concrete. https://doi.org/10.5281/zenodo.21379813; 2026.