[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121429-en":3,"doc-seo-121429-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},121429,962075114101,"Seraphina","https://ap-avatar.wpscdn.com/avatar/e000253a75eb197efd?x-image-process=image/resize,m_fixed,w_180,h_180&k=1780044092746381165",8,"Research & Report","Enhancing Building Energy Efficiency Estimations Through Graph Machine Learning: A Focus on Heating and Cooling Loads","This paper introduces graph machine learning to improve the estimation of heating and cooling loads, a key driver of building energy efficiency. Traditional approaches often miss how building topology and geometric characteristics interact, reducing prediction accuracy. The study proposes a parametric generative workflow to build a synthetic dataset covering multiple building forms with distinct topological connections and attributes. Heating and cooling loads are simulated across varied shapes and glazing scenarios, using different window sizes and orientations. Deep Graph Learning is trained with Random Forest as a validation baseline, and both models achieve strong predictive performance.","Article  \nEnhancing Building Energy Efficiency Estimations Through Graph Machine Learning: A Focus on Heating and Cooling Loads  \nWassim Jabi 1, Abdulrahman Ahmed Alymani 2, * and Ammar Alammar 3  \nAcademic Editor: Cinzia Buratti  \nReceived: 4 July 2025  \nRevised: 22 August 2025  \nAccepted: 26 August 2025  \nPublished: 9 September 2025  \nCitation: Jabi, W.; Alymani, A.A.; Alammar, A. Enhancing Building Energy Efficiency Estimations Through Graph Machine Learning: A Focus on Heating and Cooling Loads. Buildings 2025, 15, 3256. [https://](https://)[ ](https://)[doi.org/10.3390/buildings15183256](doi.org/10.3390/buildings15183256)  \n[Copyright:](Copyright:) © 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://creativecommons.org/](https://creativecommons.org/)[ ](https://creativecommons.org/)[licenses/by/4.0/](licenses/by/4.0/)) .  \n1 Welsh School of Architecture, Cardiff University, Cardiff CF10 3NB, UK; [jabiw@cardiff.ac.uk](jabiw@cardiff.ac.uk)  \n2 Department of Architectural Engineering, College of Engineering, Alfaisal University, Riyadh 11533, Saudi Arabia  \n3 Department of Architecture and Building Sciences, King Saud University, Riyadh 145111, Saudi Arabia; [aammar@ksu.edu.sa](aammar@ksu.edu.sa)  \n* Correspondence: [abalymani@alfaisal.edu](abalymani@alfaisal.edu)  \nAbstract  \nIn this paper, we introduce graph machine learning to enhance the estimation of heating and cooling loads in buildings, a critical factor in building energy efficiency. Traditional methods often overlook the complex interaction between building topology and geometric characteristics, leading to less accurate predictions. This research bridges this gap by incorporating these elements into a graph-based machine learning framework. This study introduces a parametric generative workflow to create a synthetic dataset, which is central to this research. This dataset encompasses multiple building forms, each with unique topological connections and attributes, ensuring a thorough analysis across varied building scenarios. The research involves simulating diverse building shapes and glazing scenarios with different window sizes and orientations. The study primarily utilizes Deep Graph Learning (DGL) for training, with Random Forest (RF) serving as a baseline for validation. Both DGL and RF algorithms demonstrate high performance in predicting heating and cooling loads.  \nKeywords: machine learning for energy analysis; graph machine learning; deep graph learning (DGL); building energy simulation (BES); heating and cooling loads  \n1. Introduction  \nBuildings are significant contributors to overall energy consumption. With the advancement of machine learning techniques, there is growing interest in exploring their potential to reduce building energy use and improve the accuracy of load estimation. Common machine learning methods applied to building energy prediction include support vector machines (SVMs) and artificial neural networks (ANNs) .  \nSupport vector machines (SVMs) are employed for predicting building energy consumption with limited samples, using structural risk minimization. Zhong et al. enhanced prediction accuracy by optimizing feature space through support vector regression [1] . Dong et al. applied SVMs to predict tropical commercial building energy consumption, yielding small prediction errors due to a modest data pool [2] . However, SVM’s suitability for non-linear high-dimensional patterns is offset by memory and computational demands, posing challenges for large-scale training samples.  \nArtificial neural networks (ANNs) are favored for accurate building energy consumption predictions due to self-learning and robust non-linear function fitting capabilities. Research highlights ANNs’ superiority; Wei et al. utilized a feed-forward neural network for office building occupancy and energy forecasts, ","cbCaigItG87jS6En","https://ap.wps.com/l/cbCaigItG87jS6En","pdf",2461416,1,27,"English","en",105,"# Introduction\n## Machine learning methods for building energy prediction\n## Limitations of traditional ML approaches\n## Motivation for graph machine learning\n# Methodology\n## Parametric generative workflow for synthetic data\n## Simulation scenarios for building geometry and glazing\n## Model training and validation\n# Results\n## Predictive performance of DGL\n## Baseline comparison with Random Forest\n# Conclusion","[{\"question\":\"What problem does the study address in building energy efficiency estimation?\",\"answer\":\"The study targets inaccurate heating and cooling load predictions caused by traditional methods overlooking interactions between building topology and geometric characteristics.\"},{\"question\":\"How is the synthetic dataset created for model training?\",\"answer\":\"A parametric generative workflow generates a synthetic dataset that includes multiple building forms, each with unique topological connections and geometric attributes.\"},{\"question\":\"Which algorithms are used to train and validate the prediction models?\",\"answer\":\"Deep Graph Learning (DGL) is used for training, while Random Forest (RF) serves as a baseline for validation and comparison.\"}]","Enhancing Building Energy Efficiency Estimations Through Graph Machine Learning: A Focus on Heating and Cooling Loads | PDF",1785735624,68,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"enhancing-building-energy-efficiency-estimations-through-graph-machine-learning-a-focus-on-heating-and-cooling-loads","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/enhancing-building-energy-efficiency-estimations-through-graph-machine-learning-a-focus-on-heating-and-cooling-loads/121429/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the study address in building energy efficiency estimation?","Question",{"text":75,"@type":76},"The study targets inaccurate heating and cooling load predictions caused by traditional methods overlooking interactions between building topology and geometric characteristics.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How is the synthetic dataset created for model training?",{"text":80,"@type":76},"A parametric generative workflow generates a synthetic dataset that includes multiple building forms, each with unique topological connections and geometric attributes.",{"name":82,"@type":73,"acceptedAnswer":83},"Which algorithms are used to train and validate the prediction models?",{"text":84,"@type":76},"Deep Graph Learning (DGL) is used for training, while Random Forest (RF) serves as a baseline for validation and comparison.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]