{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,31]],"date-time":"2025-12-31T16:55:31Z","timestamp":1767200131728,"version":"3.48.0"},"reference-count":43,"publisher":"MDPI AG","issue":"1","license":[{"start":{"date-parts":[[2025,12,31]],"date-time":"2025-12-31T00:00:00Z","timestamp":1767139200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"2025 National Highway Pavement Management System"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Systems"],"abstract":"<jats:p>Accurate imputation of missing pavement-condition data is critical for proactive infrastructure management, yet it is complicated by spatial non-stationarity\u2014deterioration patterns and data quality vary markedly across regions. This study proposes a Spatially Gated Mixture-of-Experts (SG-MoE) imputation model that explicitly encodes spatial heterogeneity by (i) clustering road segments using geographic coordinates and (ii) supervising a gating network to route each sample to region-specialized expert regressors. Using a large-scale national pavement management database, we benchmark SG-MoE against a strong baseline under controlled missingness mechanisms (MCAR: missing completely at random; MAR: missing at random; MNAR: missing not at random) and missing rates (10\u201350%). Across scenarios, SG-MoE consistently matches or improves upon the baseline; the largest gains occur under MCAR and the challenging MNAR setting, where spatial specialization reduces systematic underestimation of high crack-rate sections. The results provide practical guidance on when spatially aware ensembling is most beneficial for infrastructure imputation at scale. We additionally report comparative results under three missingness mechanisms. Across five random seeds, SG-MoE is comparable to the single LightGBM baseline under MCAR\/MAR and achieves its largest gains under MNAR (e.g., sMAPE improves by 0.82 points at 10% MNAR missingness).<\/jats:p>","DOI":"10.3390\/systems14010048","type":"journal-article","created":{"date-parts":[[2025,12,31]],"date-time":"2025-12-31T16:08:00Z","timestamp":1767197280000},"page":"48","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Spatially Gated Mixture of Experts for Missing Data Imputation in Pavement Management Systems"],"prefix":"10.3390","volume":"14","author":[{"given":"Bongjun","family":"Ji","sequence":"first","affiliation":[{"name":"Graduate School of Data Science, Pusan National University, Busan 46241, Republic of Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Seungyeon","family":"Han","sequence":"additional","affiliation":[{"name":"Korea Institute of Civil Engineering and Building Technology, Goyang-si 10223, Republic of Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2543-6981","authenticated-orcid":false,"given":"Mun-Sup","family":"Lee","sequence":"additional","affiliation":[{"name":"Korea Institute of Civil Engineering and Building Technology, Goyang-si 10223, Republic of Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2025,12,31]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"e25625","DOI":"10.1016\/j.heliyon.2024.e25625","article-title":"A Bayesian decision support system for optimizing pavement management programs","volume":"10","author":"Philip","year":"2024","journal-title":"Heliyon"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"45","DOI":"10.1016\/j.tra.2018.11.001","article-title":"Optimal pavement management: Resilient roads in support of emergency response of cyclone-affected coastal areas","volume":"119","author":"Amin","year":"2019","journal-title":"Transp. Res. Part A Policy Pract."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"257","DOI":"10.1016\/j.jreng.2024.04.003","article-title":"Intelligent pavement condition survey: Overview of current research and practices","volume":"4","author":"Zhang","year":"2024","journal-title":"J. Road Eng."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1080\/15732479.2020.1730409","article-title":"Aggregation of condition survey data in pavement management: Shortcomings of a homogeneous sections approach and how to avoid them","volume":"17","author":"Donev","year":"2021","journal-title":"Struct. Infrastruct. Eng."},{"key":"ref_5","unstructured":"Chang, G., Gilliland, A., Rada, G.R., Serigos, P.A., Simpson, A.L., and Kouchaki, S. (2020). Successful Practices for Quality Management of Pavement Surface Condition Data Collection and Analysis Phase I: Task 2\u2014Document of Successful Practices, U.S. Federal Highway Administration. FHWA-RC-20-0007."},{"key":"ref_6","first-page":"101312","article-title":"Which impacts matter for pavement management decisions? Quantifying social sustainability based on a capability approach","volume":"29","author":"Boakye","year":"2025","journal-title":"Transp. Res. Interdiscip. Perspect."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"227","DOI":"10.1016\/j.ijtst.2024.06.007","article-title":"Smart quality control analysis of pavement condition data for pavement management applications","volume":"18","author":"Chang","year":"2025","journal-title":"Int. J. Transp. Sci. Technol."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Huang, L.L., Lin, J.D., Huang, W.H., Kuo, C.H., and Huang, M.Y. (2024). Application of automated pavement inspection technology in provincial highway pavement maintenance decision-making. Appl. Sci., 14.","DOI":"10.3390\/app14156549"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"1426","DOI":"10.1038\/s41597-025-05748-5","article-title":"A large-scale image repository for automated pavement distress analysis and degradation trend prediction","volume":"12","author":"Yang","year":"2025","journal-title":"Sci. Data"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"03125003","DOI":"10.1061\/JCCEE5.CPENG-6662","article-title":"LiDAR-driven innovations in pavement distress detection: A review","volume":"40","author":"Si","year":"2026","journal-title":"J. Comput. Civ. Eng."},{"key":"ref_11","first-page":"67","article-title":"Advancement of a pavement management system (PMS) for the efficient management of national highways in Korea","volume":"36","author":"Han","year":"2023","journal-title":"Eng. Proc."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"105797","DOI":"10.1016\/j.autcon.2024.105797","article-title":"Automated pavement detection and artificial intelligence pavement image data processing technology","volume":"168","author":"Shang","year":"2024","journal-title":"Autom. Constr."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"681","DOI":"10.1061\/(ASCE)TE.1943-5436.0000378","article-title":"Evaluation of effectiveness and cost-effectiveness of asphalt pavement rehabilitations utilizing LTPP data","volume":"138","author":"Dong","year":"2012","journal-title":"J. Transp. Eng."},{"key":"ref_14","unstructured":"Lee, H., Yun, Y., Kim, Y., and Yun, S. (2024, January 20\u201323). The optimal method for imputing missing data in the preprocessing phase to enhance the performance of a DNN-based construction period prediction model. Proceedings of the International Conference on Construction Engineering and Project Management, Sydney, Australia."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"153","DOI":"10.1177\/03611981231188373","article-title":"Considering the spatial structure of the road network in pavement deterioration modeling","volume":"2678","author":"Gao","year":"2024","journal-title":"Transp. Res. Rec."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"106134","DOI":"10.1016\/j.autcon.2025.106134","article-title":"Modeling heterogeneous spatiotemporal pavement data for condition prediction and preventive maintenance in digital twin-enabled highway management","volume":"174","author":"Lu","year":"2025","journal-title":"Autom. Constr."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Yu, K., and Gao, L. (2023, January 14\u201317). Pavement missing condition data imputation through collective learning-based graph neural networks. Proceedings of the International Conference on Transportation and Development 2023, Austin, TX, USA.","DOI":"10.1061\/9780784484883.037"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"04022041","DOI":"10.1061\/(ASCE)IS.1943-555X.0000726","article-title":"Estimating the values of missing data related to infrastructure condition states using their spatial correlation","volume":"29","author":"Moghtadernejad","year":"2023","journal-title":"J. Infrastruct. Syst."},{"key":"ref_19","first-page":"1011","article-title":"Research progress on evaluation and prediction of degradation in service performance for asphalt pavement","volume":"12","author":"Zhang","year":"2025","journal-title":"J. Traffic Transp. Eng. Engl. Ed."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Fettahoglu, M., Ahmed, S.S., Benedyk, I., and Anastasopoulos, P.C. (2024). Macroscopic state-level analysis of pavement roughness using time\u2013space econometric modeling methods. Sustainability, 16.","DOI":"10.3390\/su16209071"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"2385013","DOI":"10.1080\/10298436.2024.2385013","article-title":"A spatial-statistical model to analyse historical rutting data","volume":"25","author":"Jourdain","year":"2024","journal-title":"Int. J. Pavement Eng."},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Movilla-Quesada, D., Rojas-Mora, J., and Raposeiras, A.C. (2022). Statistical study based on the kriging method and geographic mapping in rigid pavement defects in southern Chile. Sustainability, 14.","DOI":"10.3390\/su14010585"},{"key":"ref_23","first-page":"04014091","article-title":"Managing missing pavement performance data in pavement management systems","volume":"141","author":"Farhan","year":"2015","journal-title":"J. Transp. Eng."},{"key":"ref_24","first-page":"331","article-title":"A comparison of imputation methods using machine learning models","volume":"30","author":"Suh","year":"2023","journal-title":"Commun. Stat. Appl. Methods"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Khan, M.A. (2024). A comparative study on imputation techniques: Introducing a transformer model for robust and efficient handling of missing EEG amplitude data. Bioengineering, 11.","DOI":"10.3390\/bioengineering11080740"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"913","DOI":"10.1080\/08839514.2019.1637138","article-title":"Comparison of performance of data imputation methods for numeric datasets","volume":"33","author":"Jadhav","year":"2019","journal-title":"Appl. Artif. Intell."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Beretta, L., and Santaniello, A. (2016). Nearest neighbor imputation algorithms: A critical evaluation. BMC Med. Inform. Decis. Mak., 16.","DOI":"10.1186\/s12911-016-0318-z"},{"key":"ref_28","first-page":"48","article-title":"A study of k-nearest neighbour as an imputation method","volume":"87","author":"Batista","year":"2002","journal-title":"His"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"140","DOI":"10.1186\/s40537-021-00516-9","article-title":"A survey on missing data in machine learning","volume":"8","author":"Emmanuel","year":"2021","journal-title":"J. Big Data"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"123654","DOI":"10.1016\/j.eswa.2024.123654","article-title":"Imputation of data missing not at random: Artificial generation and benchmark analysis","volume":"249","author":"Pereira","year":"2024","journal-title":"Expert. Syst. Appl."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"112","DOI":"10.1093\/bioinformatics\/btr597","article-title":"MissForest\u2014Non-parametric missing value imputation for mixed-type data","volume":"28","author":"Stekhoven","year":"2012","journal-title":"Bioinformatics"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"120201","DOI":"10.1016\/j.eswa.2023.120201","article-title":"Deep learning versus conventional methods for missing data imputation: A review and comparative study","volume":"227","author":"Sun","year":"2023","journal-title":"Expert. Syst. Appl."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"6357","DOI":"10.1007\/s41060-025-00825-9","article-title":"A comparative study of imputation techniques for missing values in healthcare diagnostic datasets","volume":"20","author":"Joel","year":"2025","journal-title":"Int. J. Data Sci. Anal."},{"key":"ref_34","first-page":"3367","article-title":"Matrix completion and low-rank SVD via fast alternating least squares","volume":"16","author":"Hastie","year":"2015","journal-title":"J. Mach. Learn. Res."},{"key":"ref_35","first-page":"20977","article-title":"Matrix completion with quantified uncertainty through low-rank Gaussian copula","volume":"33","author":"Zhao","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"137","DOI":"10.1016\/j.neucom.2019.07.065","article-title":"Establishing strong imputation performance of a denoising autoencoder in a wide range of missing data problems","volume":"365","author":"Abiri","year":"2019","journal-title":"Neurocomputing"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"104638","DOI":"10.1016\/j.jfca.2022.104638","article-title":"Missing value imputation in food composition data with denoising autoencoders","volume":"112","author":"Gjorshoska","year":"2022","journal-title":"J. Food Compos. Anal."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Vincent, P., Larochelle, H., Bengio, Y., and Manzagol, P.-A. (2008, January 5\u20139). Extracting and Composing Robust Features with Denoising Autoencoders. Proceedings of the 25th International Conference on Machine Learning (ICML 2008), Helsinki, Finland.","DOI":"10.1145\/1390156.1390294"},{"key":"ref_39","first-page":"3371","article-title":"Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion","volume":"11","author":"Vincent","year":"2010","journal-title":"J. Mach. Learn. Res."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"107501","DOI":"10.1016\/j.patcog.2020.107501","article-title":"Handling Incomplete Heterogeneous Data Using VAEs","volume":"107","author":"Olmos","year":"2020","journal-title":"Pattern Recognit."},{"key":"ref_41","unstructured":"Yoon, J., Jordon, J., and van der Schaar, M. (2018, January 10\u201315). GAIN: Missing data imputation using generative adversarial nets. Proceedings of the International Conference on Machine Learning, Stockholm, Sweden."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"6085","DOI":"10.1038\/s41598-018-24271-9","article-title":"Recurrent neural networks for multivariate time series with missing values","volume":"8","author":"Che","year":"2018","journal-title":"Sci. Rep."},{"key":"ref_43","first-page":"1","article-title":"BRITS: Bidirectional recurrent imputation for time series","volume":"31","author":"Cao","year":"2018","journal-title":"Adv. Neural Inf. Process. Syst."}],"container-title":["Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2079-8954\/14\/1\/48\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,31]],"date-time":"2025-12-31T16:31:49Z","timestamp":1767198709000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2079-8954\/14\/1\/48"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,31]]},"references-count":43,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,1]]}},"alternative-id":["systems14010048"],"URL":"https:\/\/doi.org\/10.3390\/systems14010048","relation":{},"ISSN":["2079-8954"],"issn-type":[{"value":"2079-8954","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,31]]}}}