{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T09:43:37Z","timestamp":1773913417263,"version":"3.50.1"},"reference-count":61,"publisher":"MDPI AG","issue":"11","license":[{"start":{"date-parts":[[2025,11,7]],"date-time":"2025-11-07T00:00:00Z","timestamp":1762473600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"University of Nebraska\u2013Lincoln"},{"name":"University of Nebraska Foundation"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Information"],"abstract":"<jats:p>Buildings account for approximately one-third of global energy usage and associated carbon emissions, making energy benchmarking a crucial tool for advancing decarbonization. Current benchmarking studies have often been limited to mainly the annual scale, relied heavily on simulation-based approaches, or employed regression methods that fail to capture the complexity of diverse building stock. These limitations hinder the interpretability, generalizability, and actionable value of existing models. This study introduces a hybrid AI framework for building energy benchmarking across two time scales\u2014annual and monthly. The framework integrates supervised learning models, including white- and gray-box models, to predict annual and monthly energy consumption, combined with unsupervised learning through neural network-based Self-Organizing Maps (SOM), to classify heterogeneous building stocks. The supervised models provide interpretable and accurate predictions at both aggregated annual and fine-grained monthly levels. The model is trained using a six-year dataset from Washington, D.C., incorporating multiple building attributes and high-resolution weather data. Additionally, the generalizability and robustness have been validated via the real-world dataset from a different climate zone in Pittsburgh, PA. Followed by unsupervised learning models, the SOM clustering preserves topological relationships in high-dimensional data, enabling more nuanced classification compared to centroid-based methods. Results demonstrate that the hybrid approach significantly improves predictive accuracy compared to conventional regression methods, with the proposed model achieving over 80% R2 at the annual scale and robust performance across seasonal monthly predictions. White-box sensitivity highlights that building type and energy use patterns are the most influential variables, while the gray-box analysis using SHAP values further reveals that Energy Star\u00ae rating, Natural Gas (%), and Electricity Use (%) are the three most influential predictors, contributing mean SHAP values of 8.69, 8.46, and 6.47, respectively. SOM results reveal that categorized buildings within the same cluster often share similar energy-use patterns\u2014underscoring the value of data-driven classification. The proposed hybrid framework provides policymakers, building managers, and designers with a scalable, transparent, and transferable tool for identifying energy-saving opportunities, prioritizing retrofit strategies, and accelerating progress toward net-zero carbon buildings.<\/jats:p>","DOI":"10.3390\/info16110964","type":"journal-article","created":{"date-parts":[[2025,11,7]],"date-time":"2025-11-07T11:41:47Z","timestamp":1762515707000},"page":"964","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["An AI Hybrid Building Energy Benchmarking Framework Across Two Time Scales"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7561-7609","authenticated-orcid":false,"given":"Yi","family":"Lu","sequence":"first","affiliation":[{"name":"Tepper School of Business, Carnegie Mellon University, Pittsburgh, PA 15213, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2123-1679","authenticated-orcid":false,"given":"Tian","family":"Li","sequence":"additional","affiliation":[{"name":"Architecture Program, University of Nebraska\u2013Lincoln, Lincoln, NE 68588, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2025,11,7]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1016\/j.rser.2013.11.040","article-title":"A review on the basics of building energy estimation","volume":"31","author":"Fumo","year":"2014","journal-title":"Renew. Sustain. Energy Rev."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"115389","DOI":"10.1016\/j.enbuild.2025.115389","article-title":"A review of air-side economizers in human-centered commercial buildings: Control logic, energy savings, IAQ potential, faults, and performance enhancement","volume":"331","author":"Bie","year":"2025","journal-title":"Energy Build."},{"key":"ref_3","unstructured":"IEA (2021). Net Zero by 2050, IEA."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"126","DOI":"10.1016\/j.enbuild.2019.01.020","article-title":"A comprehensive review of energy-related data for US commercial buildings","volume":"186","author":"Ye","year":"2019","journal-title":"Energy Build."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"524","DOI":"10.1016\/j.enpol.2015.08.041","article-title":"Characterisation of representative building typologies for social housing projects in Brazil and its energy performance","volume":"87","author":"Triana","year":"2015","journal-title":"Energy Policy"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"110898","DOI":"10.1016\/j.enpol.2019.110898","article-title":"Effect of implementing building energy efficiency labeling in China: A case study in Shanghai","volume":"133","author":"Yu","year":"2019","journal-title":"Energy Policy"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"309","DOI":"10.1038\/s41560-020-0589-6","article-title":"The impact of mandatory energy audits on building energy use","volume":"5","author":"Kontokosta","year":"2020","journal-title":"Nat. Energy"},{"key":"ref_8","first-page":"321","article-title":"Energy benchmarking in commercial office buildings","volume":"7996","author":"Sharp","year":"1996","journal-title":"ACEEE"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Papadopoulos, S., Bonczak, B., and Kontokosta, C.E. (2017, January 26\u201328). Spatial and geographic patterns of building energy performance: A cross-city comparative analysis of large-scale data. Proceedings of the International Conference on Sustainable Infrastructure 2017, New York, NY, USA.","DOI":"10.1061\/9780784481219.030"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"115413","DOI":"10.1016\/j.apenergy.2020.115413","article-title":"EnergyStar++: Towards more accurate and explanatory building energy benchmarking","volume":"276","author":"Arjunan","year":"2020","journal-title":"Appl. Energy"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"122716","DOI":"10.1016\/j.apenergy.2024.122716","article-title":"MEBA: AI-powered precise building monthly energy benchmarking approach","volume":"359","author":"Li","year":"2024","journal-title":"Appl. Energy"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"111869","DOI":"10.1016\/j.enbuild.2022.111869","article-title":"BEEM: Data-driven building energy benchmarking for Singapore","volume":"260","author":"Arjunan","year":"2022","journal-title":"Energy Build."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"100320","DOI":"10.1016\/j.dibe.2024.100320","article-title":"Generalized building energy and carbon emissions benchmarking with post-prediction analysis","volume":"17","author":"Li","year":"2024","journal-title":"Dev. Built Environ."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Baset, A., and Jradi, M. (2024). Data-driven decision support for smart and efficient building energy retrofits: A review. Appl. Syst. Innov., 8.","DOI":"10.3390\/asi8010005"},{"key":"ref_15","unstructured":"Sartor, D., Piette, M.A., Tschudi, W., and Fok, S. (2000). Strategies for Energy Benchmarking in Cleanrooms and Laboratory-Type Facilities, Lawrence Berkeley National Lab (LBNL)."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"3586","DOI":"10.1016\/j.rser.2012.02.049","article-title":"A review on the prediction of building energy consumption","volume":"16","author":"Zhao","year":"2012","journal-title":"Renew. Sustain. Energy Rev."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Mahmoud, R., Kamara, J.M., and Burford, N. (2020). Opportunities and limitations of building energy performance simulation tools in the early stages of building design in the UK. Sustainability, 12.","DOI":"10.3390\/su12229702"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"1192","DOI":"10.1016\/j.rser.2017.04.095","article-title":"A review of data-driven building energy consumption prediction studies","volume":"81","author":"Amasyali","year":"2018","journal-title":"Renew. Sustain. Energy Rev."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"244","DOI":"10.1016\/j.apenergy.2018.10.053","article-title":"Grading buildings on energy performance using city benchmarking data","volume":"233","author":"Papadopoulos","year":"2019","journal-title":"Appl. Energy"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"225","DOI":"10.1016\/j.apenergy.2016.04.035","article-title":"Development of a new energy benchmark for improving the operational rating system of office buildings using various data-mining techniques","volume":"173","author":"Park","year":"2016","journal-title":"Appl. Energy"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"889","DOI":"10.1016\/j.apenergy.2017.09.060","article-title":"Machine learning approaches for estimating commercial building energy consumption","volume":"208","author":"Robinson","year":"2017","journal-title":"Appl. Energy"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"319","DOI":"10.1016\/j.enbuild.2018.03.066","article-title":"A data-driven approach for building energy benchmarking using the Lorenz curve","volume":"169","author":"Chen","year":"2018","journal-title":"Energy Build."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"1825","DOI":"10.1016\/j.enbuild.2008.04.001","article-title":"Development and validation of regression models to predict monthly heating demand for residential buildings","volume":"40","author":"Catalina","year":"2008","journal-title":"Energy Build."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"112689","DOI":"10.1016\/j.enbuild.2022.112689","article-title":"Energy benchmarking for secondary school buildings, applying the Top-Down approach","volume":"279","author":"Vaisi","year":"2023","journal-title":"Energy Build."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"607","DOI":"10.1016\/j.enbuild.2014.08.030","article-title":"A new methodology for building energy performance benchmarking: An approach based on intelligent clustering algorithm","volume":"84","author":"Gao","year":"2014","journal-title":"Energy Build."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"111327","DOI":"10.1016\/j.enpol.2020.111327","article-title":"Examining the feasibility of using open data to benchmark building energy usage in cities: A data science and policy perspective","volume":"139","author":"Roth","year":"2020","journal-title":"Energy Policy"},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Bandam, A., Busari, E., Syranidou, C., Linssen, J., and Stolten, D. (2022). Classification of building types in germany: A data-driven modeling approach. Data, 7.","DOI":"10.3390\/data7040045"},{"key":"ref_28","unstructured":"Li, T., Xie, J., Liu, T., Lu, Y., and Sawyer, A.O. (2023, January 24\u201328). An Innovative Building Energy Use Analysis by Unsupervised Classification and Supervised Regression Models. Proceedings of the ASHRAE Annual Conference, Tampa, FL, USA."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"1464","DOI":"10.1109\/5.58325","article-title":"The self-organizing map","volume":"78","author":"Kohonen","year":"2002","journal-title":"Proc. IEEE"},{"key":"ref_30","unstructured":"Vesanto, J. (2000, January 13\u201314). Neural network tool for data mining: SOM toolbox. Proceedings of the Symposium on Tool Environments and Development Methods for Intelligent Systems (TOOLMET2000), Oulu, Finland."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"51","DOI":"10.1016\/j.rser.2018.03.003","article-title":"The national solar radiation data base (NSRDB)","volume":"89","author":"Sengupta","year":"2018","journal-title":"Renew. Sustain. Energy Rev."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Parsons, H.M., Ludwig, C., G\u00fcnther, U.L., and Viant, M.R. (2007). Improved classification accuracy in 1-and 2-dimensional NMR metabolomics data using the variance stabilising generalised logarithm transformation. BMC Bioinform., 8.","DOI":"10.1186\/1471-2105-8-234"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"965","DOI":"10.3390\/make6020045","article-title":"VOD: Vision-Based Building Energy Data Outlier Detection","volume":"6","author":"Tian","year":"2024","journal-title":"Mach. Learn. Knowl. Extr."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"109711","DOI":"10.1016\/j.enbuild.2019.109711","article-title":"A comparative analysis of data-driven methods in building energy benchmarking","volume":"209","author":"Ding","year":"2020","journal-title":"Energy Build."},{"key":"ref_35","first-page":"3295074","article-title":"Lightgbm: A highly efficient gradient boosting decision tree","volume":"30","author":"Ke","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"117407","DOI":"10.1016\/j.apenergy.2021.117407","article-title":"Predicting residential electricity consumption using aerial and street view images","volume":"301","author":"Rosenfelder","year":"2021","journal-title":"Appl. Energy"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.enbuild.2018.08.038","article-title":"A study of city-level building energy efficiency benchmarking system for China","volume":"179","author":"Wei","year":"2018","journal-title":"Energy Build."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"111174","DOI":"10.1016\/j.rser.2021.111174","article-title":"Grey-box modeling and application for building energy simulations-A critical review","volume":"146","author":"Li","year":"2021","journal-title":"Renew. Sustain. Energy Rev."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"106051","DOI":"10.1016\/j.jobe.2023.106051","article-title":"Ensemble learning-based approach for residential building heating energy prediction and optimization","volume":"67","author":"Zhang","year":"2023","journal-title":"J. Build. Eng."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"4291","DOI":"10.5194\/bg-13-4291-2016","article-title":"Predicting carbon dioxide and energy fluxes across global FLUXNET sites with regression algorithms","volume":"13","author":"Tramontana","year":"2016","journal-title":"Biogeosciences"},{"key":"ref_41","first-page":"1525","article-title":"Root mean square error (RMSE) or mean absolute error (MAE)","volume":"7","author":"Chai","year":"2014","journal-title":"Geosci. Model Dev. Discuss."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"115815","DOI":"10.1016\/j.enbuild.2025.115815","article-title":"Interpretable building energy performance prediction using XGBoost Quantile Regression","volume":"340","author":"Uyar","year":"2025","journal-title":"Energy Build."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3561048","article-title":"Explainable AI (XAI): Core ideas, techniques, and solutions","volume":"55","author":"Dwivedi","year":"2023","journal-title":"ACM Comput. Surv."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"6239","DOI":"10.1109\/OJCOMS.2024.3466225","article-title":"Toward explainable reasoning in 6g: A proof of concept study on radio resource allocation","volume":"5","author":"Rezazadeh","year":"2024","journal-title":"IEEE Open J. Commun. Soc."},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"454","DOI":"10.1016\/j.apenergy.2017.06.023","article-title":"Benchmarking by convex non-parametric least squares with application on the energy performance of office buildings","volume":"203","author":"Chung","year":"2017","journal-title":"Appl. Energy"},{"key":"ref_46","first-page":"062080","article-title":"A Study on the Inter-Pretability of Network Attack Prediction Models Based on Light Gradient Boosting Machine (LGBM) and SHapley Additive exPlanations (SHAP)","volume":"83","author":"Zhang","year":"2025","journal-title":"Comput. Mater. Contin."},{"key":"ref_47","unstructured":"Lundberg, S.M., and Lee, S.I. (2017). A unified approach to interpreting model predictions. arXiv."},{"key":"ref_48","doi-asserted-by":"crossref","unstructured":"Rocha, E.M., Brochado, \u00c2.F., Rato, B., and Meneses, J. (2022, January 6\u20139). Benchmarking and Prediction of Entities Performance on Manufacturing Processes through MEA Robust XGBoost and SHAP Analysis. Proceedings of the 2022 IEEE 27th International Conference on Emerging Technologies and Factory Automation (ETFA), Stuttgart, Germany.","DOI":"10.1109\/ETFA52439.2022.9921593"},{"key":"ref_49","doi-asserted-by":"crossref","unstructured":"Joshi, K., Jana, A., Arjunan, P., and Ramamritham, K. (2022, January 9\u201310). XENIA: Axplainable energy informatics and attributes for building energy benchmarking. Proceedings of the 9th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation, Boston, MA, USA.","DOI":"10.1145\/3563357.3566140"},{"key":"ref_50","doi-asserted-by":"crossref","unstructured":"Li, T., and Lu, Y. (2025). AI-Driven Dual-Scale Building Energy Benchmarking for Decarbonization. SSRN.","DOI":"10.2139\/ssrn.5171950"},{"key":"ref_51","doi-asserted-by":"crossref","first-page":"52","DOI":"10.1016\/j.neunet.2012.09.018","article-title":"Essentials of the self-organizing map","volume":"37","author":"Kohonen","year":"2013","journal-title":"Neural Netw."},{"key":"ref_52","unstructured":"Majidi, F. (2023). A Hybrid SOM and K-means Model for Time Series Energy Consumption Clustering. arXiv."},{"key":"ref_53","doi-asserted-by":"crossref","unstructured":"Zhang, W., Liu, X., Wen, Y., Zhou, S., Ge, R., and Xie, X. (2023, January 27\u201329). User classification based on SOM+ K-means cluster analysis. Proceedings of the 2023 International Conference on Applied Physics and Computing (ICAPC), Ottawa, ON, Canada.","DOI":"10.1109\/ICAPC61546.2023.00079"},{"key":"ref_54","doi-asserted-by":"crossref","first-page":"586","DOI":"10.1109\/72.846731","article-title":"Clustering of the self-organizing map","volume":"11","author":"Vesanto","year":"2000","journal-title":"IEEE Trans. Neural Netw."},{"key":"ref_55","doi-asserted-by":"crossref","unstructured":"Vaisi, S., Mohammadi, S., Nastasi, B., and Javanroodi, K. (2020). A new Generation of Thermal Energy Benchmarks for University Buildings. Energies, 13.","DOI":"10.3390\/en13246606"},{"key":"ref_56","doi-asserted-by":"crossref","first-page":"553","DOI":"10.1016\/j.enbuild.2012.08.010","article-title":"A method for energy classification of hotels: A case-study of Greece","volume":"55","author":"Farrou","year":"2012","journal-title":"Energy Build."},{"key":"ref_57","doi-asserted-by":"crossref","first-page":"58","DOI":"10.1016\/j.enbuild.2017.12.040","article-title":"DUE-B: Data-driven urban energy benchmarking of buildings using recursive partitioning and stochastic frontier analysis","volume":"163","author":"Yang","year":"2018","journal-title":"Energy Build."},{"key":"ref_58","doi-asserted-by":"crossref","unstructured":"Keyhani, M., Abbaspour, A., Bahadori-Jahromi, A., Mylona, A., Janbey, A., Godfrey, P., and Zhang, H. (2023). Whole life carbon assessment of a typical UK residential building using different embodied carbon data sources. Sustainability, 15.","DOI":"10.3390\/su15065115"},{"key":"ref_59","doi-asserted-by":"crossref","first-page":"663","DOI":"10.1016\/j.jclepro.2017.05.156","article-title":"A comparative analysis of embodied carbon in high-rise buildings regarding different design parameters","volume":"161","author":"Gan","year":"2017","journal-title":"J. Clean. Prod."},{"key":"ref_60","unstructured":"Trahair, N.S., Bradford, M., Nethercot, D., and Gardner, L. (2017). The Behaviour and Design of Steel Structures to EC3, CRC Press."},{"key":"ref_61","doi-asserted-by":"crossref","first-page":"1218","DOI":"10.1007\/s11367-016-1087-8","article-title":"The ecoinvent database version 3 (part I): Overview and methodology","volume":"21","author":"Wernet","year":"2016","journal-title":"Int. J. Life Cycle Assess."}],"container-title":["Information"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2078-2489\/16\/11\/964\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,9]],"date-time":"2025-11-09T05:18:35Z","timestamp":1762665515000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2078-2489\/16\/11\/964"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,7]]},"references-count":61,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2025,11]]}},"alternative-id":["info16110964"],"URL":"https:\/\/doi.org\/10.3390\/info16110964","relation":{},"ISSN":["2078-2489"],"issn-type":[{"value":"2078-2489","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,11,7]]}}}