{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T06:34:51Z","timestamp":1772692491995,"version":"3.50.1"},"reference-count":63,"publisher":"MDPI AG","issue":"3","license":[{"start":{"date-parts":[[2026,3,3]],"date-time":"2026-03-03T00:00:00Z","timestamp":1772496000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Machines"],"abstract":"<jats:p>Artificial-intelligence-driven wayside monitoring has become a promising solution for early identification of railway wheel flats, enabling safer operations and more efficient maintenance planning. This study introduces a comparative investigation of supervised and unsupervised machine learning strategies for wheel flat identification, with particular emphasis on real-time applicability and sensor cost reduction. Support Vector Machines (SVMs) and k-means clustering are evaluated as representative supervised and unsupervised approaches using vibration data obtained from numerically simulated train\u2013track interactions under realistic operating conditions, including train speeds of 120 km\/h and 200 km\/h and multiple wheel flat severities. A key contribution of this work is the proposal of a simplified supervised classification framework that directly exploits Auto-Regressive features extracted from rail-mounted accelerometers, eliminating the need for feature normalization and multi-sensor data fusion. This simplification significantly reduces computational effort, making the approach suitable for real-time deployment in operational railway environments. In parallel, a systematic sensitivity analysis is conducted to assess the influence of sensor placement and to identify the minimum sensor configuration required to achieve reliable damage classification. The outputs from the current study show that an SVM emerges with more accurate defect classification than the k-means clustering, allowing a wayside system with fewer sensors.<\/jats:p>","DOI":"10.3390\/machines14030286","type":"journal-article","created":{"date-parts":[[2026,3,3]],"date-time":"2026-03-03T15:15:29Z","timestamp":1772550929000},"page":"286","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Support Vector Machine and k-Means Clustering for Advanced Wheel Flat Identification: A Comparison of Supervised and Unsupervised Methods"],"prefix":"10.3390","volume":"14","author":[{"given":"Alireza","family":"Chegini","sequence":"first","affiliation":[{"name":"Department of Civil Engineering, Faculty of Engineering, Islamic Azad University, Qazvin 1477893855, Iran"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1079-4687","authenticated-orcid":false,"given":"Mohammadreza","family":"Mohammadi","sequence":"additional","affiliation":[{"name":"CONSTRUCT, Faculty of Engineering, University of Porto, 4200-465 Porto, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6751-0388","authenticated-orcid":false,"given":"Araliya","family":"Mosleh","sequence":"additional","affiliation":[{"name":"CONSTRUCT, Faculty of Engineering, University of Porto, 4200-465 Porto, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2470-9834","authenticated-orcid":false,"given":"Cecilia","family":"Vale","sequence":"additional","affiliation":[{"name":"CONSTRUCT, Faculty of Engineering, University of Porto, 4200-465 Porto, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6132-4920","authenticated-orcid":false,"given":"Ramin","family":"Ghiasi","sequence":"additional","affiliation":[{"name":"Structural Dynamics and Assessment Laboratory, School of Civil Engineering, University College Dublin, D04 V1W8 Dublin, Ireland"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5809-7971","authenticated-orcid":false,"given":"Ruben","family":"Silva","sequence":"additional","affiliation":[{"name":"CONSTRUCT, Faculty of Engineering, University of Porto, 4200-465 Porto, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-7377-1716","authenticated-orcid":false,"given":"Antonio","family":"Guedes","sequence":"additional","affiliation":[{"name":"CONSTRUCT, Faculty of Engineering, University of Porto, 4200-465 Porto, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8327-1452","authenticated-orcid":false,"given":"Andreia","family":"Meixedo","sequence":"additional","affiliation":[{"name":"CONSTRUCT, Faculty of Engineering, University of Porto, 4200-465 Porto, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1358-1943","authenticated-orcid":false,"given":"Abdollah","family":"Malekjafarian","sequence":"additional","affiliation":[{"name":"Structural Dynamics and Assessment Laboratory, School of Civil Engineering, University College Dublin, D04 V1W8 Dublin, Ireland"}]}],"member":"1968","published-online":{"date-parts":[[2026,3,3]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"13257","DOI":"10.1109\/ACCESS.2023.3240167","article-title":"State-of-the-art wayside condition monitoring systems for railway wheels: A comprehensive review","volume":"11","author":"Shaikh","year":"2023","journal-title":"IEEE Access"},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Bragan\u00e7a, C., Souza, E.F., Ribeiro, D., Meixedo, A., Bittencourt, T.N., and Carvalho, H. (2023). Drive-by Methodologies Applied to Railway Infrastructure Subsystems: A Literature Review\u2014Part II: Track and Vehicle. Appl. Sci., 13.","DOI":"10.3390\/app13126982"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Fu, W., He, Q., Feng, Q., Li, J., Zheng, F., and Zhang, B. (2023). Recent advances in wayside railway wheel flat detection techniques: A review. Sensors, 23.","DOI":"10.3390\/s23083916"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"639","DOI":"10.1080\/23248378.2022.2096132","article-title":"Automatic clustering-based approach for train wheels condition monitoring","volume":"11","author":"Mosleh","year":"2023","journal-title":"Int. J. Rail Transp."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Mohammadi, M., Mosleh, A., Vale, C., Ribeiro, D., Montenegro, P., and Meixedo, A. (2023). An Unsupervised Learning Approach for Wayside Train Wheel Flat Detection. Sensors, 23.","DOI":"10.3390\/s23041910"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1953","DOI":"10.1016\/j.ymssp.2005.12.012","article-title":"Wheel-flat diagnostic tool via wavelet transform","volume":"20","author":"Belotti","year":"2006","journal-title":"Mech. Syst. Signal Process."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"1536","DOI":"10.1177\/1475921720921772","article-title":"A Bayesian machine learning approach for online detection of railway wheel defects using track-side monitoring","volume":"20","author":"Ni","year":"2021","journal-title":"Struct. Health Monit."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"642","DOI":"10.1016\/j.ymssp.2016.07.009","article-title":"Fault detection method for railway wheel flat using an adaptive multiscale morphological filter","volume":"84","author":"Li","year":"2017","journal-title":"Mech. Syst. Signal Process."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"1176","DOI":"10.1109\/TITS.2017.2720721","article-title":"Wheel defect detection with machine learning","volume":"19","author":"Krummenacher","year":"2017","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Vitola, J., Pozo, F., Tibaduiza, D., and Anaya, M. (2017). Distributed Piezoelectric Sensor System for Damage Identification in Structures Subjected to Temperature Changes. Sensors, 17.","DOI":"10.3390\/s17061252"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Shafique, R., Siddiqui, H., Rustam, F., Ullah, S., Siddique, M., Lee, E., Ashraf, I., and Dudley, S. (2021). A Novel Approach to Railway Track Faults Detection Using Acoustic Analysis. Sensors, 21.","DOI":"10.3390\/s21186221"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"2379","DOI":"10.1016\/j.matdes.2006.07.018","article-title":"A Na\u00efve-Bayes classifier for damage detection in engineering materials","volume":"28","author":"Addin","year":"2007","journal-title":"Mater. Des."},{"key":"ref_13","unstructured":"Arthur, D., and Vassilvitskii, S. (2006). k-Means++: The Advantages of Careful Seeding, Stanford University."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"177","DOI":"10.1080\/00423114.2024.2425415","article-title":"M-CLUSTER: Multistage clustering for unsupervised train wheel condition monitoring","volume":"64","author":"Ghiasi","year":"2024","journal-title":"Veh. Syst. Dyn."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"6999","DOI":"10.1109\/TNNLS.2021.3084827","article-title":"A survey of convolutional neural networks: Analysis, applications, and prospects","volume":"33","author":"Li","year":"2021","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"185","DOI":"10.1016\/j.trc.2018.07.007","article-title":"A decision support approach for condition-based maintenance of rails based on big data analysis","volume":"95","author":"Jamshidi","year":"2018","journal-title":"Transp. Res. Part C Emerg. Technol."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Zhang, Q., Peng, J., Tian, K., Wang, A., Li, J., and Gao, X. (2024). Advancing Ultrasonic Defect Detection in High-Speed Wheels via UT-YOLO. Sensors, 24.","DOI":"10.3390\/s24051555"},{"key":"ref_18","first-page":"2906","article-title":"Integrated Multiple-Defect Detection and Evaluation of Rail Wheel Tread Images using Convolutional Neural Networks","volume":"12","author":"Trilla","year":"2021","journal-title":"Int. J. Progn. Health Manag."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"111959","DOI":"10.1016\/j.measurement.2022.111959","article-title":"Rail wheel tread defect detection using improved YOLOv3","volume":"203","author":"Xing","year":"2022","journal-title":"Measurement"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"103918","DOI":"10.1016\/j.aei.2025.103918","article-title":"Multi-objective optimisation of surveillance camera placement for bridge\u2013ship collision early-warning using an improved non-dominated sorting genetic algorithm","volume":"69","author":"Liao","year":"2026","journal-title":"Adv. Eng. Inform."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"1644","DOI":"10.1080\/00423114.2022.2103436","article-title":"Early wheel flat detection: An automatic data-driven wavelet-based approach for railways","volume":"61","author":"Mosleh","year":"2023","journal-title":"Veh. Syst. Dyn."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"232","DOI":"10.1080\/00423114.2024.2333771","article-title":"Early identification of out-of-roundness damage wheels in railway freight vehicles using a wayside system and a stacked sparse autoencoder","volume":"63","author":"Jorge","year":"2025","journal-title":"Veh. Syst. Dyn."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"272","DOI":"10.1177\/09544097251313570","article-title":"Smart railways: AI-based track-side monitoring for wheel flat identification","volume":"239","author":"Mohammadi","year":"2025","journal-title":"Proc. Inst. Mech. Eng. Part F J. Rail Rapid Transit"},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Guo, G., Peng, J., Yang, K., Xie, L., and Song, W. (2017, January 22\u201324). Wheel Tread Defects Inspection Based on SVM. Proceedings of the 2017 Far East NDT New Technology & Application Forum (FENDT), Xi\u2019an, China.","DOI":"10.1109\/FENDT.2017.8584565"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"112","DOI":"10.1177\/09544097251378295","article-title":"Wayside monitoring for railway wheel flat identification: A multiclass supervised learning approach","volume":"240","author":"Mosleh","year":"2026","journal-title":"Proc. Inst. Mech. Eng. Part F J. Rail Rapid Transit"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"e148","DOI":"10.1590\/1679-78255010","article-title":"Wheel-flat detection on trams using envelope analysis with Hilbert transform","volume":"16","author":"Nowakowski","year":"2019","journal-title":"Lat. Am. J. Solids Struct."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"8902","DOI":"10.1109\/TIM.2020.2998888","article-title":"Linear and quadratic time\u2013frequency analysis of vibration for fault detection and identification of NFR trains","volume":"69","author":"Barman","year":"2020","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"3286","DOI":"10.1177\/14759217221149614","article-title":"Experimental detection of train wheel defects using wayside vibration signal processing","volume":"22","author":"Salehi","year":"2023","journal-title":"Struct. Health Monit."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"107433","DOI":"10.1016\/j.engfailanal.2023.107433","article-title":"Adaptive time series representation for out-of-round railway wheels fault diagnosis in wayside monitoring","volume":"152","author":"Ferraz","year":"2023","journal-title":"Eng. Fail. Anal."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Gao, R., He, Q., and Feng, Q. (2019). Railway wheel flat detection system based on a parallelogram mechanism. Sensors, 19.","DOI":"10.3390\/s19163614"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Zhou, C., Gao, L., Xiao, H., and Hou, B. (2020). Railway wheel flat recognition and precise positioning method based on multisensor arrays. Appl. Sci., 10.","DOI":"10.3390\/app10041297"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"85","DOI":"10.1109\/JSEN.2011.2135848","article-title":"Real-time monitoring of railway traffic using fiber Bragg grating sensors","volume":"12","author":"Filograno","year":"2011","journal-title":"IEEE Sens. J."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"106376","DOI":"10.1016\/j.engfailanal.2022.106376","article-title":"Real time implementation of fiber Bragg grating sensor in monitoring flat wheel detection for railways","volume":"138","author":"Mishra","year":"2022","journal-title":"Eng. Fail. Anal."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"e338","DOI":"10.1590\/1679-78256086","article-title":"Advanced acoustic signal analysis used for wheel-flat detection","volume":"18","author":"Komorski","year":"2021","journal-title":"Lat. Am. J. Solids Struct."},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Deilamsalehy, H., Havens, T.C., and Lautala, P. (2017, January 4\u20137). Sensor fusion of wayside visible and thermal imagery for rail car wheel and bearing damage detection. Proceedings of the ASME\/IEEE Joint Rail Conference, Philadelphia, PA, USA.","DOI":"10.1115\/JRC2017-2284"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Mosleh, A., Montenegro, P.A., Costa, P.A., and Cal\u00e7ada, R. (2021). Railway vehicle wheel flat detection with multiple records using spectral kurtosis analysis. Appl. Sci., 11.","DOI":"10.3390\/app11094002"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"1785","DOI":"10.1016\/S0045-7825(03)00206-8","article-title":"On the exploitation of chaos to build reduced-order models","volume":"192","author":"Kerschen","year":"2003","journal-title":"Comput. Methods Appl. Mech. Eng."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"1047","DOI":"10.1002\/stc.1825","article-title":"On-line unsupervised detection of early damage","volume":"23","author":"Santos","year":"2016","journal-title":"Struct. Control Health Monit."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"338","DOI":"10.1002\/suco.201500092","article-title":"Real-time damage detection based on pattern recognition","volume":"17","year":"2016","journal-title":"Struct. Concr."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"741","DOI":"10.1006\/jsvi.2002.5148","article-title":"Novelty Detection in a Changing Environment: Regression and Interpolation Approaches","volume":"258","author":"Worden","year":"2002","journal-title":"J. Sound Vib."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"99","DOI":"10.1016\/j.ress.2017.10.017","article-title":"On the use of AR models for SHM: A global sensitivity and uncertainty analysis framework","volume":"170","author":"Datteo","year":"2018","journal-title":"Reliab. Eng. Syst. Saf."},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Nielsen, J.C.O., Pieringer, A., Thompson, D.J., and Torstensson, P.T. (2021). Wheel\u2013Rail Impact Loads, Noise and Vibration: A Review of Excitation Mechanisms, Prediction Methods and Mitigation Measures. Noise and Vibration Mitigation for Rail Transportation Systems, Springer International Publishing.","DOI":"10.1007\/978-3-030-70289-2_1"},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"120191","DOI":"10.1016\/j.measurement.2025.120191","article-title":"Data-driven diagnosis of railway wheel flat based on multi-channel vibration data fusion","volume":"264","author":"Zhang","year":"2026","journal-title":"Measurement"},{"key":"ref_44","unstructured":"Vladimir, V. (1998). Statistical Learning Theory, Wiley."},{"key":"ref_45","first-page":"779","article-title":"Support vector regression machines","volume":"28","author":"Harris","year":"1997","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_46","first-page":"11","article-title":"Machine Learning: A Review on Binary Classification","volume":"160","author":"Roshan","year":"2017","journal-title":"Int. J. Comput. Appl."},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"181","DOI":"10.1007\/s40534-023-00306-4","article-title":"Wheel\u2013rail contact model for railway vehicle\u2013structure interaction applications: Development and validation","volume":"31","author":"Montenegro","year":"2023","journal-title":"Railw. Eng. Sci."},{"key":"ref_48","unstructured":"(MATLAB\u00ae, 2018). MATLAB\u00ae, version R2018a."},{"key":"ref_49","doi-asserted-by":"crossref","first-page":"156","DOI":"10.1515\/crll.1882.92.156","article-title":"Ueber die Ber\u00fchrung fester elastischer K\u00f6rper","volume":"1882","author":"Hertz","year":"1882","journal-title":"J. F\u00fcr Die Reine Und Angew. Math."},{"key":"ref_50","unstructured":"Kalker, J.J. (1996). Book of Tables for the Herzian Creep-Force Law, Delft University of Technology, Faculty of Technical Mathematics and Informatics."},{"key":"ref_51","unstructured":"(ANSYS\u00ae, 2018). ANSYS\u00ae, release 19.2; Academic Research."},{"key":"ref_52","unstructured":"(2002). Railway Applications Railway Applications\u2014Track-Rail-Part1, Final Draft (Standard No. EN 13674\u20131). European Standard."},{"key":"ref_53","doi-asserted-by":"crossref","first-page":"1349","DOI":"10.1080\/00423110802621561","article-title":"Fundamentals of vehicle\u2013track coupled dynamics","volume":"47","author":"Zhai","year":"2009","journal-title":"Veh. Syst. Dyn."},{"key":"ref_54","unstructured":"(1999). Rail Bridges for Speeds > 200 km\/h: NUMERICAL Investigation of the Effect of Track Irregularities at Bridge Resonance (Standard No. ERRI D 214\/RP 5)."},{"key":"ref_55","unstructured":"(2001). Track\/Bridge Interaction\u2014Recommendations for Calculation, 2nd Edition (Standard No. UIC 774-3-R)."},{"key":"ref_56","doi-asserted-by":"crossref","first-page":"251","DOI":"10.1016\/S0141-0296(02)00147-5","article-title":"Steady-state response and riding comfort of trains moving over a series of simply supported bridges","volume":"25","author":"Wu","year":"2003","journal-title":"Eng. Struct."},{"key":"ref_57","unstructured":"(1999). Improved Knowledge of Forces in CWR Track (Including Switches): Parametric Study and Sensivity Analysis of CWERRI (Standard No. ERRI D 202\/RP 11)."},{"key":"ref_58","doi-asserted-by":"crossref","first-page":"933","DOI":"10.1016\/j.euromechsol.2008.02.001","article-title":"Dynamic interaction of various beams with the underlying soil\u2014Finite and infinite, half-space and Winkler models","volume":"27","author":"Auersch","year":"2008","journal-title":"Eur. J. Mech.\u2014A\/Solids"},{"key":"ref_59","doi-asserted-by":"crossref","first-page":"20","DOI":"10.1007\/s40534-023-00316-2","article-title":"Influence of railway wheel tread damage on wheel\u2013rail impact loads and the durability of wheelsets","volume":"32","author":"Maglio","year":"2024","journal-title":"Railw. Eng. Sci."},{"key":"ref_60","doi-asserted-by":"crossref","first-page":"1005","DOI":"10.1177\/0954409719882828","article-title":"Evaluation of the influential parameters contributing to the reconstruction of railway wheel defect signals","volume":"234","author":"Alemi","year":"2019","journal-title":"Proc. Inst. Mech. Eng. Part F J. Rail Rapid Transit"},{"key":"ref_61","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1177\/0954409718784362","article-title":"Reconstruction of an informative railway wheel defect signal from wheel\u2013rail contact signals measured by multiple wayside sensors","volume":"233","author":"Alemi","year":"2018","journal-title":"Proc. Inst. Mech. Eng. Part F J. Rail Rapid Transit"},{"key":"ref_62","doi-asserted-by":"crossref","first-page":"183","DOI":"10.1177\/0954409719838115","article-title":"A new strategy to estimate static loads for the dynamic weighing in motion of railway vehicles","volume":"234","author":"Mosleh","year":"2020","journal-title":"Proc. Inst. Mech. Eng. Part F"},{"key":"ref_63","unstructured":"(2003). Railway Applications\/Track-Track Geometry Quality (Standard No. BS EN 13848-1)."}],"container-title":["Machines"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-1702\/14\/3\/286\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T05:14:19Z","timestamp":1772687659000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-1702\/14\/3\/286"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,3]]},"references-count":63,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3]]}},"alternative-id":["machines14030286"],"URL":"https:\/\/doi.org\/10.3390\/machines14030286","relation":{},"ISSN":["2075-1702"],"issn-type":[{"value":"2075-1702","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,3]]}}}