傳統瀝青混凝土的配比設計方法普遍需經過繁瑣的試驗流程，不僅耗時也耗成本，因此本研究動機為利用機器學習方法訓練出能夠預測成效試驗結果的模型，以縮短人力和物力成本。瀝青混凝土領域中，多數機器學習的相關研究著重於變換不同的模型演算法，而非輸入模型的變數，因此本研究的目標為探討同一種變數以不同形式輸入模型對結果的影響，並將訓練完成的模型應用於平衡配比設計，嘗試將此技術與實務結合。
本研究選擇改變級配曲線和膠泥種類兩種質性變數的輸入樣式，分別以篩分析殘留百分比和Bailey 法代表級配曲線；以獨熱編碼和目標編碼代表膠泥種類，並利用隨機森林、XGBoost、支持向量機和人工神經網路模型進行機器學習，輸出的結果分成試驗結果能否通過漢堡輪跡車轍試驗的標準及直接預測漢堡輪跡車轍試驗結果。研究結果顯示，分類任務中，表現最好的是XGBoost搭配篩號殘留百分比與目標編碼，有95.21%的準確率、96.77%的敏感度和92.45%的特異度；回歸預測中，表現最好的是XGBoost搭配篩號殘留百分比與獨熱編碼，結果為R2=0.823和MSE=0.017。
為嘗試平衡配比設計，本研究首先訓練VFA通過與否分類器來驗證新試驗設計參數的體積特性是否符合標準，得到通過率100%，接著將這些新試驗設計參數放入上述訓練完成的回歸預測模型進行車轍深度值的預測。
研究結果指出，不同的變數表示方式確實會影響模型的預測結果，分類模型對級配表示法的變化較不敏感，但整體傾向於將原始數據換算成更精簡的表示方式（目標編碼），而回歸模型傾向於直接使用原始數據（級配曲線與獨熱編碼）。對於新設計參數的預測，雖預測結果較不理想，但對於將機器學習模型與實務相結合又更近一步，即使預測尚有進步空間，模型仍有潛力成為提供初步意見的輔助工具。

Traditional asphalt concrete mix design methods are often time-consuming and resource-intensive. This study proposes a machine learning-based approach to predict performance test outcomes, aiming to reduce labor and material costs. Unlike previous research focusing primarily on model algorithms, this work explores how different input representations of the same variables affect predictive performance. Two categorical variables—gradation curve and binder type—are represented using sieve analysis percentages vs. Bailey parameters and one-hot encoding vs. target encoding, respectively. Four machine learning algorithms—Random Forest, XGBoost, SVM, and ANN—are applied for both classification (pass/fail of the Hamburg Wheel Tracking Test, HWTT) and regression (rut depth prediction).
The results show that XGBoost with sieve percentages and target encoding achieved the best classification accuracy (95.21%), while XGBoost with sieve percentages and one-hot encoding produced the best regression performance (R² = 0.823, MSE = 0.017). A VFA classifier was also developed to validate the volumetric compliance of new designs, achieving a 100% pass rate. These designs were then used to predict rut depths via the trained regression models.
Findings reveal that input representation significantly impacts model performance. Classification models favor simplified inputs, while regression models benefit from raw data. Although predictions for new designs need refinement, this study demonstrates the potential of machine learning to assist practical mix design decisions.

Aggarwal, P. 2017. “Predicting dynamic modulus for bituminous concrete using support vector machine.” 2017 International Conference on Infocom Technologies and Unmanned Systems (Trends and Future Directions) (ICTUS), 751–755. Dubai: IEEE.
Alexander L. Fradkov. 2020. “Early History of Machine Learning.” IFAC-PapersOnLine, 53 (2): 1385–1390. https://doi.org/10.1016/j.ifacol.2020.12.1888.
Aliferis, C., and G. Simon. 2024. “Overfitting, Underfitting and General Model Overconfidence and Under-Performance Pitfalls and Best Practices in Machine Learning and AI.” Artificial Intelligence and Machine Learning in Health Care and Medical Sciences, Health Informatics, G. J. Simon and C. Aliferis, eds., 477–524. Cham: Springer International Publishing.
Alnaqbi, A. J., W. Zeiada, G. G. Al-Khateeb, K. Hamad, and S. Barakat. 2023. “Creating Rutting Prediction Models through Machine Learning Techniques Utilizing the Long-Term Pavement Performance Database.” Sustainability, 15 (18): 13653. https://doi.org/10.3390/su151813653.
Arulkumaran, K., M. P. Deisenroth, M. Brundage, and A. A. Bharath. 2017. “Deep Reinforcement Learning: A Brief Survey.” IEEE Signal Process. Mag., 34 (6): 26–38. https://doi.org/10.1109/MSP.2017.2743240.
Awed, A. M., A. N. Awaad, M. R. Kaloop, J. W. Hu, S. M. El-Badawy, and R. T. Abd El-Hakim. 2023. “Boosting Hot Mix Asphalt Dynamic Modulus Prediction Using Statistical and Machine Learning Regression Modeling Techniques.” Sustainability, 15 (19): 14464. https://doi.org/10.3390/su151914464.
Balandino Di Donato. 2020. “Designing embodied human-computer interactions in music performance.” Unpublished. https://doi.org/10.13140/RG.2.2.26027.36640.
Baldo, N., M. Miani, F. Rondinella, J. Valentin, P. Vackcová, and E. Manthos. 2022. “Stiffness Data of High-Modulus Asphalt Concretes for Road Pavements: Predictive Modeling by Machine-Learning.” Coatings, 12 (1): 54. https://doi.org/10.3390/coatings12010054.
Behnood, A., and D. Daneshvar. 2020. “A machine learning study of the dynamic modulus of asphalt concretes: An application of M5P model tree algorithm.” Construction and Building Materials, 262: 120544. https://doi.org/10.1016/j.conbuildmat.2020.120544.
Behnood, A., and E. Mohammadi Golafshani. 2021. “Predicting the dynamic modulus of asphalt mixture using machine learning techniques: An application of multi biogeography-based programming.” Construction and Building Materials, 266: 120983. https://doi.org/10.1016/j.conbuildmat.2020.120983.
C A Padmanabha Reddy, Y., P. Viswanath, and B. Eswara Reddy. 2018. “Semi-supervised learning: a brief review.” IJET, 7 (1.8): 81. https://doi.org/10.14419/ijet.v7i1.8.9977.
Cai, W., Y. Du, D. Wu, Z. Weng, and C. Liu. 2025. “Engineering-Adaptive Pavement Maintenance Decision-Making Model: A Reinforcement Learning Approach From Expert Feedback.” IEEE Trans. Intell. Transport. Syst., 26 (7): 10865–10880. Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/tits.2025.3547939.
Çeli̇K, Ö. 2018. “A Research on Machine Learning Methods and Its Applications.” Journal of Educational Technology and Online Learning, 1 (3): 25–40. https://doi.org/10.31681/jetol.457046.
Chandola, V., A. Banerjee, and V. Kumar. 2009. “Anomaly detection: A survey.” ACM Comput. Surv., 41 (3): 1–58. https://doi.org/10.1145/1541880.1541882.
Chawla, N. V., K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer. 2002. “SMOTE: Synthetic Minority Over-sampling Technique.” jair, 16: 321–357. https://doi.org/10.1613/jair.953.
Chen, T., and C. Guestrin. 2016. “XGBoost: A Scalable Tree Boosting System.” Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794. San Francisco California USA: ACM.
Choudhary, R., and H. K. Gianey. 2017. “Comprehensive Review On Supervised Machine Learning Algorithms.” 2017 International Conference on Machine Learning and Data Science (MLDS), 37–43. Noida: IEEE.
Chun, C., and S.-K. Ryu. 2019. “Road Surface Damage Detection Using Fully Convolutional Neural Networks and Semi-Supervised Learning.” Sensors, 19 (24): 5501. MDPI AG. https://doi.org/10.3390/s19245501.
Dai, M., F. Zhang, S. Dai, C. Xing, S. Xiao, H. Lv, and Y. Tan. 2024. “Optimizing asphalt mix design through predicting volumetric properties using interpretable machine learning.” Powder Technology, 444: 119954. https://doi.org/10.1016/j.powtec.2024.119954.
Dao, D. V., Q.-A. T. Bui, D. D. Nguyen, I. Prakash, S. H. Trinh, and B. T. Pham. 2022. “Prediction of interlayer shear strength of double-layer asphalt using novel hybrid artificial intelligence models of ANFIS and metaheuristic optimizations.” Construction and Building Materials, 323: 126595. Elsevier BV. https://doi.org/10.1016/j.conbuildmat.2022.126595.
Dao, D. V., N.-L. Nguyen, H.-B. Ly, B. T. Pham, and T.-T. Le. 2020. “Cost-Effective Approaches Based on Machine Learning to Predict Dynamic Modulus of Warm Mix Asphalt with High Reclaimed Asphalt Pavement.” Materials, 13 (15): 3272. https://doi.org/10.3390/ma13153272.
Drive, K., and F. Park. n.d. “S.GEORGIA DOT RESEARCH PROJECT 15-03 FINAL REPORT.”
Gardezi, H., M. Ikrama, M. Usama, M. Iqbal, F. E. Jalal, A. Hussain, and X. Li. 2024. “Predictive modeling of rutting depth in modified asphalt mixes using gene-expression programming (GEP): A sustainable use of RAP, fly ash, and plastic waste.” Construction and Building Materials, 443: 137809. https://doi.org/10.1016/j.conbuildmat.2024.137809.
Gong, H., Y. Sun, Y. Dong, B. Han, P. Polaczyk, W. Hu, and B. Huang. 2020. “Improved estimation of dynamic modulus for hot mix asphalt using deep learning.” Construction and Building Materials, 263: 119912. https://doi.org/10.1016/j.conbuildmat.2020.119912.
Grebenschikov, S., and J. A. Prozzi. 2011. “M.Enhancing Mechanistic–Empirical Pavement Design Guide Rutting-Performance Predictions with Hamburg Wheel-Tracking Results.” Transportation Research Record: Journal of the Transportation Research Board, 2226 (1): 111–118. https://doi.org/10.3141/2226-12.
Guo, N., L. You, Z. Long, S. Lv, and A. Diab. 2023. “Computationally-Affordable Unsupervised Machine Learning Algorithm to Identify the Level of Distress Severity in Pavement Functional Performance.” IEEE Trans. Intell. Transport. Syst., 24 (7): 7342–7356. Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/tits.2023.3253845.
Habibnejad Korayem, A., H. Ziari, M. Hajiloo, M. Abarghooie, and P. Karimi. 2020. “H.Laboratory evaluation of stone mastic asphalt containing amorphous carbon powder as filler material.” Construction and Building Materials, 243: 118280. https://doi.org/10.1016/j.conbuildmat.2020.118280.
Hamed Majidifard, Behnam Jahangiri, Punyaslok Rath, Amir H. Alavi, and William G. Buttlar. 2020. “A Deep Learning Approach to Predict Hamburg Rutting Curve.” arXiv: Learning.
Hasan, M. M., M. Ahmad, M. A. Hasan, H. M. Faisal, and R. A. Tarefder. 2019. “C.Laboratory Performance Evaluation of Fine and Coarse-Graded Asphalt Concrete Mix.” J. Mater. Civ. Eng., 31 (11): 04019259. https://doi.org/10.1061/(ASCE)MT.1943-5533.0002905.
Hassan Divandari. 2019. “Predict of Asphalt Rutting Potential Based on IDT and Validation with ANN.” Journal of Applied Engineering Sciences.
Ismael, M. Q., H. H. Joni, and M. Y. Fattah. 2023. “Neural network modeling of rutting performance for sustainable asphalt mixtures modified by industrial waste alumina.” Ain Shams Engineering Journal, 14 (5): 101972. Elsevier BV. https://doi.org/10.1016/j.asej.2022.101972.
Jabbar, H. K., and R. Z. Khan. 2014. “Methods to Avoid Over-Fitting and Under-Fitting in Supervised Machine Learning (Comparative Study).” Computer Science, Communication and Instrumentation Devices, 163–172. Research Publishing Services.
Ke-Zhen Yan, Dong-Dong Ge, and Zou Zhang. 2014. “Support Vector Machine Models for Prediction of Flow Number of Asphalt Mixtures.” International Journal of Pavement Research and Technology, 7 (1). https://doi.org/10.6135/ijprt.org.tw/2014.7(1).31.
Khan, M. Y., A. Qayoom, M. S. Nizami, M. S. Siddiqui, S. Wasi, and S. M. K.-R. Raazi. 2021. “Automated Prediction of Good Dictionary EXamples (GDEX): A Comprehensive Experiment with Distant Supervision, Machine Learning, and Word Embedding‐Based Deep Learning Techniques.” Complexity, (S. Sarfraz, ed.), 2021 (1): 2553199. https://doi.org/10.1155/2021/2553199.
Kraljevski, I., Y. C. Ju, D. Ivanov, C. Tschöpe, and M. Wolff. 2023. “How to Do Machine Learning with Small Data? -- A Review from an Industrial Perspective.” arXiv.
LEO BREIMAN. 2001. “Random Forests.”
Li, J., L. Qiu, B. Tang, D. Chen, D. Zhao, and R. Yan. 2019. “Insufficient Data Can Also Rock! Learning to Converse Using Smaller Data with Augmentation.” AAAI, 33 (01): 6698–6705. https://doi.org/10.1609/aaai.v33i01.33016698.
Liu, J., F. Liu, and L. Wang. 2024. “Automated, economical, and environmentally-friendly asphalt mix design based on machine learning and multi-objective grey wolf optimization.” Journal of Traffic and Transportation Engineering (English Edition), 11 (3): 381–405. https://doi.org/10.1016/j.jtte.2023.10.002.
Liu, J., F. Liu, Z. Wang, E. O. Fanijo, and L. Wang. 2023. “Involving prediction of dynamic modulus in asphalt mix design with machine learning and mechanical-empirical analysis.” Construction and Building Materials, 407: 133610. https://doi.org/10.1016/j.conbuildmat.2023.133610.
Liu, J., F. Liu, C. Zheng, D. Zhou, and L. Wang. 2022. “Optimizing asphalt mix design through predicting effective asphalt content and absorbed asphalt content using machine learning.” Construction and Building Materials, 325: 126607. https://doi.org/10.1016/j.conbuildmat.2022.126607.
Lv, Q., W. Huang, M. Zheng, H. Sadek, Y. Zhang, and C. Yan. 2020. “B.Influence of gradation on asphalt mix rutting resistance measured by Hamburg Wheel Tracking test.” Construction and Building Materials, 238: 117674. https://doi.org/10.1016/j.conbuildmat.2019.117674.
Majidifard, H., B. Jahangiri, W. G. Buttlar, and A. H. Alavi. 2019. “New machine learning-based prediction models for fracture energy of asphalt mixtures.” Measurement, 135: 438–451. https://doi.org/10.1016/j.measurement.2018.11.081.
Majidifard, H., B. Jahangiri, P. Rath, L. Urra Contreras, W. G. Buttlar, and A. H. Alavi. 2021. “Developing a prediction model for rutting depth of asphalt mixtures using gene expression programming.” Construction and Building Materials, 267: 120543. https://doi.org/10.1016/j.conbuildmat.2020.120543.
Mogawer, W. S., K. Stuart, A. J. Austerman, and A. A. Soliman. 2018. “E.Investigating the Performances of Plant-Produced High-Reclaimed Asphalt Pavement Content Warm Mix Asphalts.” Transportation Research Record, 2672 (28): 130–142. https://doi.org/10.1177/0361198118786828.
Mohammad, L. N., Z. Wu, S. Obulareddy, S. Cooper, and C. Abadie. 1970. “G.Permanent Deformation Analysis of Hot-Mix Asphalt Mixtures with Simple Performance Tests and 2002 Mechanistic–Empirical Pavement Design Software.” Transportation Research Record.
Moussa, G. S., and M. Owais. 2020. “Pre-trained deep learning for hot-mix asphalt dynamic modulus prediction with laboratory effort reduction.” Construction and Building Materials, 265: 120239. https://doi.org/10.1016/j.conbuildmat.2020.120239.
Naeem, S., A. Ali, S. Anam, and M. M. Ahmed. 2023. “An Unsupervised Machine Learning Algorithms: Comprehensive Review.” IJCDS, 13 (1): 911–921. https://doi.org/10.12785/ijcds/130172.
Nguyen, H. L., and V. Q. Tran. 2023. “Data-driven approach for investigating and predicting rutting depth of asphalt concrete containing reclaimed asphalt pavement.” Construction and Building Materials, 377: 131116. https://doi.org/10.1016/j.conbuildmat.2023.131116.
Nguyen, L. N., T.-H. Le, L. Q. Nguyen, and V. Q. Tran. 2023. “Machine learning approaches for predicting Cracking Tolerance Index (CTIndex) of asphalt concrete containing reclaimed asphalt pavement.” PLoS ONE, (A. Khitab, ed.), 18 (10): e0287255. Public Library of Science (PLoS). https://doi.org/10.1371/journal.pone.0287255.
Ortigosa-Hernández, J., I. Inza, and J. A. Lozano. 2017. “Measuring the class-imbalance extent of multi-class problems.” Pattern Recognition Letters, 98: 32–38. https://doi.org/10.1016/j.patrec.2017.08.002.
Pothuganti, S. 2018. “Review on over-fitting and under-fitting problems in Machine Learning and solutions.” 7 (9).
Radhakrishnan, V., G. S. Chowdari, K. S. Reddy, and R. Chattaraj. 2019. “F.Evaluation of wheel tracking and field rutting susceptibility of dense bituminous mixes.” Road Materials and Pavement Design, 20 (1): 90–109. https://doi.org/10.1080/14680629.2017.1374998.
Rahman, S., A. Bhasin, and A. Smit. 2021. “Exploring the use of machine learning to predict metrics related to asphalt mixture performance.” Construction and Building Materials, 295: 123585. https://doi.org/10.1016/j.conbuildmat.2021.123585.
Ren, R., P. Shi, P. Jia, and X. Xu. 2023. “A Semi-Supervised Learning Approach for Pixel-Level Pavement Anomaly Detection.” IEEE Trans. Intell. Transport. Syst., 24 (9): 10099–10107. Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/tits.2023.3267433.
Schafer, J. L., and J. W. Graham. 2002. “Missing data: Our view of the state of the art.” Psychological Methods, 7 (2): 147–177. https://doi.org/10.1037/1082-989X.7.2.147.
Shaikh, S., and A. Gupta. 2024. “Assessing cracking resistance and threshold limits of bituminous mixtures with IDEAL-CT and predictive modeling techniques.” Construction and Building Materials, 449: 138349. Elsevier BV. https://doi.org/10.1016/j.conbuildmat.2024.138349.
Shinde, P. P., and S. Shah. 2018. “A Review of Machine Learning and Deep Learning Applications.” 2018 Fourth International Conference on Computing Communication Control and Automation (ICCUBEA), 1–6. Pune, India: IEEE.
Singh, A., R. Nowak, and J. Zhu. 2008. “Unlabeled data: Now it helps, now it doesn’t.”
Valizadeh, N., M. Mirzaei, M. F. Allawi, H. A. Afan, N. S. Mohd, A. Hussain, and A. El-Shafie. 2017. “Artificial intelligence and geo-statistical models for stream-flow forecasting in ungauged stations: state of the art.” Nat Hazards, 86 (3): 1377–1392. https://doi.org/10.1007/s11069-017-2740-7.
Vijayvargiya, A., C. Prakash, R. Kumar, S. Bansal, and J. M. R.S. Tavares. 2021. “Human knee abnormality detection from imbalanced sEMG data.” Biomedical Signal Processing and Control, 66: 102406. https://doi.org/10.1016/j.bspc.2021.102406.
Wang, J., R. Zhang, R. Wang, H. Bahia, W. Huang, D. Wang, and W. Cai. 2024. “Prediction of the fundamental viscoelasticity of asphalt mixtures using ML algorithms.” Construction and Building Materials, 442: 137573. https://doi.org/10.1016/j.conbuildmat.2024.137573.
Xu, P., X. Ji, M. Li, and W. Lu. 2023. “Small data machine learning in materials science.” npj Comput Mater, 9 (1): 42. https://doi.org/10.1038/s41524-023-01000-z.
Xu, W., X. Huang, Z. Yang, M. Zhou, and J. Huang. 2022. “Developing Hybrid Machine Learning Models to Determine the Dynamic Modulus (E*) of Asphalt Mixtures Using Parameters in Witczak 1-40D Model: A Comparative Study.” Materials, 15 (5): 1791. https://doi.org/10.3390/ma15051791.
Yang, Y., B. Qian, Q. Xu, and Y. Yang. 2020. “Climate Regionalization of Asphalt Pavement Based on the K‐Means Clustering.pdf.”
Yao, L., Z. Leng, J. Jiang, and F. Ni. 2022. “Large-Scale Maintenance and Rehabilitation Optimization for Multi-Lane Highway Asphalt Pavement: A Reinforcement Learning Approach.” IEEE Trans. Intell. Transport. Syst., 23 (11): 22094–22105. Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/tits.2022.3161689.
Yildirim, Y., and T. Kennedy. n.d. “L.Correlation of Field Performance to Hamburg Wheel Tracking Device Results.”
Yip, K. Y., C. Cheng, and M. Gerstein. 2013. “Machine learning and genome annotation: a match meant to be?” Genome Biol, 14 (5): 205. https://doi.org/10.1186/gb-2013-14-5-205.
Zhang, K., Z. Min, X. Hao, T. F. P. Henning, and W. Huang. 2025. “Enhancing understanding of asphalt mixture dynamic modulus prediction through interpretable machine learning method.” Advanced Engineering Informatics, 65: 103111. https://doi.org/10.1016/j.aei.2025.103111.
Zhao, G., J. Huyan, S. Tighe, and W. Li. 2019. “IMPLEMENTING UNSUPERVISED MACHINE LEARNING TO GAIN A BETTER UNDERSTANDING OF THE ASPHALT PAVEMENT CONDITIONS OF ONTARIO PROVINCIAL HIGHWAYS.”
Zhou, F., S. Hu, and D. Newcomb. 2020. “D.Development of a performance-related framework for production quality control with ideal cracking and rutting tests.” Construction and Building Materials, 261: 120549. https://doi.org/10.1016/j.conbuildmat.2020.120549.