Hybrid machine learning framework for transverse cracking prediction in CRCP with PSO and GBM

Abstract

Transverse cracking is a major distress mechanism in Continuously Reinforced Concrete Pavement (CRCP), affecting ride smoothness, service life, and maintenance strategies. This research introduces a hybrid predictive framework that couples Particle Swarm Optimization (PSO) with Gradient Boosting Machine (GBM) to enhance the accuracy of transverse crack prediction in CRCP. The analysis utilized 395 records from 33 pavement sections obtained from the Long-Term Pavement Performance (LTPP) program, encompassing structural, environmental, traffic, and performance-related parameters. PSO was applied to fine-tune critical GBM hyperparameters, namely the number of iterations, learning rate, and tree depth. The optimized PSO–GBM model demonstrated excellent performance, yielding an average RMSE of 1.62 and an R² of 0.99 under 5-fold cross-validation, surpassing benchmark models such as conventional GBM, Random Forest, Support Vector Regression (SVR), Artificial Neural Networks (ANN), and Linear Regression. Sensitivity analysis revealed that L3 thickness, L4 thickness, and Annual Average Daily Traffic (AADT) were the most significant contributors, consistent with engineering knowledge of crack development. Validation through residual distribution and equality line plots confirmed the robustness and stability of the proposed approach across varying severity levels.

Keywords

Transverse cracking / Concrete pavement / Particle Swarm Optimization / Gradient Boosting Machine / Hybrid machine learning / Pavement Management Systems

References

[1] Choi S, Ha S, Won MC. Mechanism of transverse crack development in continuously reinforced concrete pavement at early ages. Transp Res Rec, 2015, 2524(1): 42-58

[2] Zhang Y, Roesler J, Dahal S. Predicting transverse crack properties in continuously reinforced concrete pavement. Constr Build Mater, 2023, 364: 129842

[3] Rao BNM, Durga CSS, Venkatesh C, Rao TM. Sustainable geopolymer concrete for pavements: performance evaluation of recycled concrete aggregates in fly ash-based mixtures. J Sustain Constr Mater Technol, 2024, 9(3): 211-220

[4] Zeiada W, Alnaqbi AJ, Al-Khateeb GG, Abuzwidah M. Machine learning modeling of transverse cracking in flexible pavement. Disc Civil Eng, 2024, 1(1): 1-26

[5] Tiza MT, Jırgba K, Sanı HA, Sesugh T. Effect of thermal variances on flexible pavements. J Sustain Constr Mater Technol, 2022, 7(3): 220-230

[6] Acıkök F. Effect of fly ash and ground granulated blast furnace slag on the strength of concrete pavement. J Sustain Constr Mater Technol, 2018, 3(3): 278-285

[7] Alnaqbi A, Zeiada W, Al-Khateeb GG, Abuzwidah M. Machine learning modeling of wheel and non-wheel path longitudinal cracking. Buildings, 2024, 14(3): 709

[8] Al-Samahi S, Zeiada W, Al-Khateeb GG, Hamad K, Alnaqbi A. A comparative study of pavement roughness prediction models under different climatic conditions. Infrastructures, 2024

[9] Federal Highway Administration (2003) Distress identification manual for the long-term pavement performance program (Publication No. FHWA-RD-03-031). U.S. Department of Transportation. https://www.fhwa.dot.gov/publications/research/infrastructure/pavements/ltpp/reports/03031/03031.pdf

[10] Tao J, Luo X, Qiu X, Wang F. Data quality assessment of automated pavement cracking measurements in Mississippi. Int J Pavement Res Technol, 2021, 14: 117-128

[11] Abubakar AS, Wahab MMBA, Shafiq N, Danyaro KU, Adebanjo AU (2024) Integrating life cycle cost analysis into pipeline asset integrity management: A comprehensive approach in decision support systems. J Hunan Univ Nat Sci 51(1):91–111. https://doi.org/10.55463/issn.1674-2974.51.1.10

[12] Khichad JS, Vishwakarma RJ. Overview and discussion of pavement performance prediction techniques for maintenance and rehabilitation decision-making. Int J Pavement Res Technol, 2024

[13] Haddad A (2020) Machine learning based models coupled with data assimilation techniques for pavement rutting prediction, American University of Beirut

[14] Ghasemi P (2019) Application of optimization and machine learning techniques in predicting pavement performance and performance-based pavement design, Iowa State University

[15] Kilani A, Adelani AA, Oladejo OS, Ikotun BD, Olubambi A. Analyzing an efficient mix design for the production of quality asphalt concrete: a means of reducing roads’ maintenance cost. J Sustain Constr Mat Technol, 2024, 9(2): 177-198

[16] Luo X, Gu F, Ling M, Lytton RL. Review of mechanistic-empirical modeling of top-down cracking in asphalt pavements. Constr Build Mater, 2018, 191: 1053-1070

[17] Zeiada W, Al-Khateeb G, Fattouh I, Souliman M, Alnaqbi A. Impact of confinement condition of dynamic modulus test on the performance of flexible pavement structures. Innov Infrastr Solut, 2024, 9(8): 290

[18] Alnaqbi A, Zeiada W, Al-Khateeb GG. Machine learning modeling of pavement performance and IRI prediction in flexible pavement. Innov Infrastr Solut, 2024, 910385

[19] Kotb MM (2024) Prediction of distresses in pavement networks: a machine learning approach, The American University in Cairo, Egypt

[20] Alnaqbi A, Al-Khateeb GG, Zeiada W (2025) Machine learning applications for predicting longitudinal cracking in continuously reinforced concrete pavement. Constr Econ Build 25(1)

[21] Kang J (2022) Pavement performance prediction using machine learning and instrumentation in smart pavement, University of Waterloo

[22] Tamagusko T, Gomes Correia M, Ferreira A. Machine learning applications in road pavement management: a review, challenges and future directions. Infrastructures, 2024

[23] Siddiqui AR, Khan R, Akhtar MN. Sustainable concrete solutions for green infrastructure development: a review. J Sustain Constr Mater Technol, 2025, 101108-141

[24] Alnaqbi AJ, Zeiada W, Al-Khateeb GG, Hamad K, Barakat S. Creating rutting prediction models through machine learning techniques utilizing the long-term pavement performance database. Sustainability, 2023, 15(18): 13653

[25] Tamagusko T, Ferreira A. Machine learning for prediction of the international roughness index on flexible pavements: a review, challenges, and future directions. Infrastructures, 2023, 812170

[26] Alnaqbi AJ, Zeiada W, Al-Khateeb G, Abttan A, Abuzwidah M. Predictive models for flexible pavement fatigue cracking based on machine learning. Transport Eng, 2024, 16: 100243

[27] Marcelino P, de Lurdes Antunes M, Fortunato E, Gomes MC. Machine learning approach for pavement performance prediction. Int J Pavement Eng, 2021, 223341-354

[28] Justo-Silva R, Ferreira A, Flintsch G. Review on machine learning techniques for developing pavement performance prediction models. Sustainability, 2021, 13(9): 5248

[29] Kang J, Tavassoti P, Chaudhry MNAR, Baaj H, Ghafurian M. Artificial intelligence techniques for pavement performance prediction: a systematic review. Road Mater Pavement Des, 2025, 263497-522

[30] Shtayat A, Moridpour S, Best B, Rumi S. An overview of pavement degradation prediction models. J Adv Transp, 2022, 2022(1): 7783588

[31] Rane N, Choudhary SP, Rane J. Ensemble deep learning and machine learning: applications, opportunities, challenges, and future directions. Studies in Medical and Health Sciences, 2024, 1218-41

[32] Abbas MA, Al-Mudhafar WJ, Wood DA. Improving permeability prediction in carbonate reservoirs through gradient boosting hyperparameter tuning. Earth Sci Inform, 2023, 16(4): 3417-3432

[33] Mwita M, Mbelwa J, Agbinya J, Sam AE. The effect of hyperparameter optimization on the estimation of performance metrics in network traffic prediction using the gradient boosting machine model. Eng Technol Appl Sci Res, 2023, 13(3): 10714-10720

[34] Saha S, Saha A, Roy B, Sarkar R, Bhardwaj D, Kundu B. Integrating the particle swarm optimization (PSO) with machine learning methods for improving the accuracy of the landslide susceptibility model. Earth Sci Inform, 2022, 15(4): 2637-2662

[35] Benbouras MA. Hybrid meta-heuristic machine learning methods applied to landslide susceptibility mapping in the Sahel-Algiers. Int J Sediment Res, 2022, 37(5): 601-618

[36] Pourpanah F, Wang R, Lim CP, Wang XZ, Yazdani D. A review of artificial fish swarm algorithms: recent advances and applications. Artif Intell Rev, 2023, 56(3): 1867-1903

[37] Su H, Fu Y. Research on dynamic adjustment of investment portfolio and time series prediction technology based on ESG criteria. J Comput Methods Sci Eng, 2024

[38] Alnaqbi A, Al-Khateeb GG, Zeiada W. Predictive modeling of transverse cracking in continuously reinforced concrete pavement: a machine learning approach. Eng Res Express, 2025, 71015106

[39] Rada GR, Elkins GE, Hicks RG (2018) Long-Term Pavement Performance (LTPP) program specific pavement studies (SPS)-development of experiment design: SPS-11 asphalt concrete pavement preservation study (No. FHWA-HIF-18-063). Federal Highway Administration, United States

RIGHTS & PERMISSIONS

The Author(s)