Title:
Machine Learning-Based Optimization of Ultra-High-Performance Concrete for Flexural Strength and Sustainability
Author(s):
Fayez Moutassem
Publication:
Materials Journal
Volume:
123
Issue:
3
Appears on pages(s):
169-178
Keywords:
Bayesian neural networks (BNNs); CO2 emissions; flexural strength prediction; multi-objective optimization; SHapley Additive exPlanations (SHAP); sustainability; ultra-high-performance concrete (UHPC)
DOI:
10.14359/51749415
Date:
5/1/2026
Abstract:
This study presents a machine learning-driven framework for the sustainable design of ultra-high-performance concrete (UHPC) mixtures with a focus on maximizing flexural strength while minimizing material costs and embodied CO2 emissions. A curated data set of 333 UHPC mixtures was developed, incorporating 13 input features, including binder composition, steel fiber dosage, and curing parameters. A Bayesian neural network (BNN) was trained to predict flexural strength with high accuracy (R2 = 0.936, root mean square error [RMSE] = 1.37 MPa, and mean absolute error [MAE] = 1.09 MPa), supported by residual analysis confirming minimal prediction bias and robust generalization. SHapley Additive exPlanations (SHAP) analysis was used to interpret model predictions and identify key drivers of flexural behavior—namely, curing time, steel fiber dosage, and silica fume content. The BNN was coupled with the Non-dominated Sorting Genetic Algorithm III (NSGA-III) to perform multi-objective optimization and generate Pareto-optimal UHPC mixtures. A utility-based scoring method was introduced to select designs aligned with different project priorities, enabling the identification of fiber-rich, high-strength mixtures as well as low-emission, cost-efficient alternatives. The framework supports field-level implementation and is well-suited for integration with sustainability rating systems such as Leadership in Energy and Environmental Design (LEED) or Envision. It provides a transparent, generalizable, and industry-ready tool for intelligent UHPC mixture optimization, advancing data-driven design practices for green infrastructure applications.
Related References:
Abuhussain, M. A.; Ahmad, A.; Amin, M. N.; Althoey, F.; Gamil, Y.; and Najeh, T., 2024, “Data-Driven Approaches for Strength Prediction of Alkali-Activated Composites,” Case Studies in Construction Materials, V. 20, July, Article No. e02920. doi: 10.1016/j.cscm.2024.e02920
Amran, M.; Murali, G.; Makul, N.; Tang, W. C.; and Alluqmani, A. E., 2023, “Sustainable Development of Eco-Friendly Ultra-High Performance Concrete (UHPC): Cost, Carbon Emission, and Structural Ductility,” Construction and Building Materials, V. 398, Sept., Article No. 132477. doi: 10.1016/j.conbuildmat.2023.132477
Doheny, M., ed., 2022, Building Construction Costs with RSMeans Data, The Gordian Group, Inc., Greenville, SC, 948 pp.
Graybeal, B., 2014, “Design and Construction of Field-Cast UHPC Connections,” FHWA Report No. FHWA-HRT-14-084, Federal Highway Administration, Turner-Fairbank Highway Research Center, McLean, VA, 36 pp.
Hammond, G., and Jones, C., 2011, “Embodied Carbon: The Inventory of Carbon and Energy (ICE),” Version 3.0, F. Lowrie and P. Tse, eds., BSRIA Limited, Bracknell, Berkshire, UK, https://greenbuildingencyclopaedia.uk/wp-content/uploads/2014/07/Full-BSRIA-ICE-guide.pdf. (last accessed Apr. 23, 2026)
Ibrahim, S. M.; Ansari, S. S.; and Hasan, S. D., 2023, “Towards White Box Modeling of Compressive Strength of Sustainable Ternary Cement Concrete Using Explainable Artificial Intelligence (XAI),” Applied Soft Computing, V. 149, Part B, Dec., Article No. 110997. doi: 10.1016/j.asoc.2023.110997
Imam, A.; Sharma, K. K.; Kumar, V.; and Singh, N., 2022, “A Review Study on Sustainable Development of Ultra High-Performance Concrete,” AIMS Materials Science, V. 9, No. 1, pp. 9-35. doi: 10.3934/matersci.2022002
Kang, S.-T.; Lee, Y.; Park, Y.-D.; and Kim, J.-K., 2010, “Tensile Fracture Properties of an Ultra High Performance Fiber Reinforced Concrete (UHPFRC) with Steel Fiber,” Composite Structures, V. 92, No. 1, Jan., pp. 61-71. doi: 10.1016/j.compstruct.2009.06.012
Kashem, A.; Karim, R.; Malo, S. C.; Das, P.; Datta, S. D.; and Alharthai, M., 2024, “Hybrid Data-Driven Approaches to Predicting the Compressive Strength of Ultra-High-Performance Concrete Using SHAP and PDP Analyses,” Case Studies in Construction Materials, V. 20, July, Article No. e02991. doi: 10.1016/j.cscm.2024.e02991
Meng, W.; Valipour, M.; and Khayat, K. H., 2017, “Optimization and Performance of Cost-Effective Ultra-High Performance Concrete,” Materials and Structures, V. 50, No. 1, Feb., Article No. 29. doi: 10.1617/s11527-016-0896-3
Moutassem, F., and Kharseh, M., 2024, “Artificial Neural Network Model for Concrete Strength Predictions Based on Ultrasonic Pulse Velocity Measurement,” ACI Materials Journal, V. 121, No. 4, July, pp. 61-68. doi: 10.14359/51740776
Qiu, M.; Ke, L.; and Li, Y.-L., 2025, “Dataset of UHPC Compressive and Flexural Strengths,” Mendeley Data. doi: 10.17632/98583gp5hp.1.
Richard, P., and Cheyrezy, M., 1995, “Composition of Reactive Powder Concretes,” Cement and Concrete Research, V. 25, No. 7, Oct., pp. 1501-1511. doi: 10.1016/0008-8846(95)00144-2
WBCSD, 2011, “The Cement CO2 and Energy Protocol: CO2 and Energy Accounting and Reporting Standard for the Cement Industry,” Version 3.0, World Business Council for Sustainable Development, Geneva, Switzerland, https://docs.wbcsd.org/2011/05/CSI-CO2-Protocol.pdf. (last accessed Apr. 24, 2026)
Zhao, X. A.; Nematollahi, B.; Chougan, M.; and Xiao, J., 2025, “Approaches to Reduce Cost and Environmental Impacts of UHPC Production: A Review,” Case Studies in Construction Materials, V. 22, July, Article No. e04644. doi: 10.1016/j.cscm.2025.e04644