Applying AI to Generate Accurate Building Models for the new UK’s HEM Energy Assessment Framework
Doctoral Training Grant Funding Information
This funding model includes a 36 month fully funded PhD Studentship, set in-line with UK Research & Innovation values. For 2025/6, this will be £20,780 per year. The tax-free stipend will be paid monthly. This PhD Studentship also includes a Full-Time Fee Scholarship for up to 3 years. The funding is subject to your continued registration on the research degree, making satisfactory progression within your PhD, as well as attendance on and successful completion of the Postgraduate Certificate in Research Practice.
All applicants will receive the same stipend irrespective of fee status.
Application Closing Date:
Midday (UK Time) on Wednesday 17th September 2025 for a start date of 2nd February 2026.
How to Apply
To apply, please follow the below steps:
- Complete the BCU Online Application Form.
- Complete the Doctoral Studentship Proposal Form in full, ensuring that you quote the project ID. You will be required to upload your proposal in place of a personal statement on the BCU online application form.
- Upload two references to your online application form (at least one of which must be an academic reference).
- Upload your qualification(s) for entry onto the research degree programme. This will be Bachelor/Master’s certificate(s) and transcript(s).
- International applicants must also provide a valid English language qualification. Please see the list of English language qualifications accepted here. Please check the individual research degree course page for the required scores.
Frequently Asked Questions
To help support you to complete your application, please consult the frequently asked questions below:
Project title: Applying AI to Generate Accurate Building Models for the new UK’s HEM Energy Assessment Framework
Project Lead: Dr Essa Shahra
Project ID: 11 - 45475970
Project description:
The UK government has identified approximately six million homes constructed with inefficient solid walls that require significant energy performance improvements. Among the most critically affected regions is Greater Birmingham and Solihull, which is recognised as one of the worst areas in the country for energy efficiency within its housing stock. In response to this challenge, the government has historically launched several policy initiatives aimed at improving energy performance. These include the Green Deal Home Improvement Fund (GDHIF) and earlier programmes such as Birmingham Energy Savers (which concluded in October 2015) and the Green Deal (which ended in July 2015). These initiatives collectively targeted the retrofit of 60,000 homes and 1,000 non-residential buildings by 2020.
Typically, retrofit schemes begin with an assessment phase to evaluate a building’s potential for energy savings and determine whether the cost of improvements can be recovered within a reasonable payback period. This stage often relies on government-accredited tools such as the Standard Assessment Procedure (SAP), which is used to generate Energy Performance Certificates (EPCs). SAP provides energy ratings and recommends measures for reducing energy consumption. However, the tool is based on standardised assumptions and annual averages that do not always align with real-world building performance.
To address these limitations, in 2024 and 2025, the UK government introduced a consultation process for a more advanced alternative: the Home Energy Model (HEM). HEM leverages building simulation with half-hourly resolution to provide a dynamic and more accurate assessment of energy use. It factors in real-time variables such as occupancy behaviour, weather conditions, and building orientation. While HEM offers higher accuracy than SAP, it still requires significant manual input, particularly in creating detailed 3D representations of buildings. This becomes especially challenging for non-standard or irregularly shaped structures.
In response, the Computing Research Team at BCU has been developing intelligent tools designed to interface directly with HEM. These tools aim to empower non-architects to generate realistic 3D building models, thereby improving the accuracy and accessibility of energy simulations. Early prototypes have shown promising results, achieving approximately 65% modelling accuracy. This work has attracted interest from industry stakeholders, including AES, a leading energy assessment company with a 10% market share in EPC certifications. Discussions are currently underway to establish a commercial spin-off company based on this technology.
The research builds on the success of previous projects, such as EcRoFit and RetrofitPlus, which have together secured £2.5 million in funding and form the foundation for a major impact case study at BCU. Although the project is progressing toward commercialisation with external support from ERDF and UK Research programmes, further institutional backing from BCU would significantly enhance validation and dissemination of the underpinning research. This work is novel, applied, and aligned with BCU’s 2023 vision, with outcomes already published in prestigious venues including Nature, IBPSA, and IEEE Access.
The project also explores two additional research strands: automated 2D-to-3D modelling using AI and the redefinition of Gross Floor Area (GFA) measurement through digital innovation. A mixed-methods approach will be used, including performance comparisons between AI-generated and human-created models, alongside interviews with architects, engineers, and construction managers to assess usability and adoption.
Anticipated findings and contributions to knowledge:
This research aims to explore the practical adoption and impact of recent advancements in AI-driven 2D-to-3D modelling within the construction industry. It will assess how these technologies enhance design accuracy, project efficiency, and standardisation across sectors such as urban planning, real estate, and construction management.
The study will focus on developing applied deep learning techniques, including a novel framework and evaluation methods, to significantly improve the quality of 3D models generated from 2D inputs—whether architectural drawings or photographs. Alongside enhancing the core AI engine, the research will address the critical need for accessibility, model calibration, and integration with real-world performance data through IoT sensing. These elements are essential for minimising the performance gap between digital models and actual building behaviour.
Qualitative interviews and quantitative surveys will be conducted to assess the benefits and challenges of these tools in practice. Anticipated benefits include improved visualisation, streamlined collaboration, and reduced manual errors. However, the study will also identify technical, economic, and regulatory barriers to widespread adoption. Real-world case studies will provide further insight into practical implementation and measurable project outcomes.
The project aims to bridge the gap between emerging AI capabilities and their integration into construction workflows. It will particularly contribute to standardising complex tasks such as Gross Floor Area (GFA) calculation. Key outcomes will include a deeper understanding of adoption patterns, the limitations and benefits of current tools, and practical recommendations for policy, industry standards, and future innovation in digital construction.
Person Specification:
Essential Criteria
Qualifications & Research Experience:
- First-class Bachelor’s degree in Computer Science, Artificial Intelligence, Building Information Modelling (BIM), Digital Construction, Architectural Technology, or a closely related discipline.
- Master’s degree (completed or near completion) in one of the above fields.
- Demonstrated experience conducting independent or supervised research with an applied and/or interdisciplinary focus.
- Strong written and verbal communication skills, with the ability to document technical findings for both academic and non-specialist audiences.
- Proven ability to work collaboratively in interdisciplinary teams and engage with academic, industry, and public stakeholders.
Programming & Software Development:
- Strong programming skills in Python and/or JavaScript.
- Experience building end-to-end AI applications using frameworks such as FastAPI, Microsoft Power Platform, and cloud services (e.g., AWS, Azure).
- Competent in data validation, performance benchmarking, and error quantification for system testing and optimization.
Machine Learning & Artificial Intelligence:
- Proficient in developing and training scalable models using TensorFlow, PyTorch, and Scikit-learn.
- Practical experience in applying Large Language Models (LLMs) and Natural Language Processing (NLP) to multilingual and domain-specific tasks.
- Familiarity with phonetic similarity models and search engine integration for information retrieval tasks.
Computer Vision & 3D Technologies:
- Experience with computer vision, 2D-to-3D reconstruction techniques, and photogrammetry.
- Developed image processing tools (e.g., Clarity Kit) for forensic analysis and digital transformation workflows.
- Familiar with 3D modelling technologies commonly used in the AEC sector, including Blender, SketchUp, and Autodesk Revit.
Simulation, BIM & Built Environment:
- Exposure to or willingness to develop expertise in energy modelling and simulation platforms such as SAP, HEM, and EnergyPlus.
- Understanding of Building Information Modelling (BIM) standards including IFC, COBie, and ISO 19650.
- Familiarity with UK building regulations, Energy Performance Certificates (EPCs), and retrofit assessment workflows.
- Experience working within the built environment, energy, or construction sectors.
- Previous experience working within the built environment, energy, or construction sectors.
- Knowledge of UK building regulations, Energy Performance Certificates (EPCs), or retrofit assessment workflows.
- Experience working with Building Information Modelling (BIM) standards such as IFC, COBie, or ISO 19650.
- Familiarity with Geographic Information Systems (GIS), LiDAR data processing, or 3D scanning technologies.
- Experience developing or contributing to digital tools or web-based platforms, especially those interfacing with simulation engines or cloud-based systems.
- Experience publishing in high-quality journals or presenting at international conferences.
- Ability to engage with policymakers, local authorities, or public audiences to communicate complex ideas simply.
Overseas applicants:
International applicants must also provide a valid English language qualification, such as International English Language Test System (IELTS) or equivalent with an overall score of 6.5 with no band below 6.0.
Additional information:
This research role at Birmingham City University (BCU) is part of a high-impact project focused on using AI and digital tools to improve energy efficiency in UK housing. The successful candidate will work within a multidisciplinary team developing AI-powered 2D-to-3D building modelling tools integrated with the government’s Home Energy Model (HEM). The project supports national climate targets and aims to modernise retrofit practices in regions like Greater Birmingham and Solihull.
Key responsibilities include technical development of modelling tools, stakeholder engagement (e.g., interviews, surveys), and contributing to academic publications and commercialisation efforts. The role also involves work on automating Gross Floor Area (GFA) measurement and reducing the performance gap using real-world sensor data.
The position offers access to cutting-edge facilities, flexible working options, and professional development support. It is initially fixed-term, with potential for extension. BCU values diversity and encourages applicants from underrepresented groups. This is a unique opportunity to contribute to impactful, interdisciplinary research addressing real-world sustainability challenges.
Contact:
If you have any questions or need further information, please use the contact details below:
- For enquiries about the funding or project proposal, please contact: Essa.Shahra@becu.ac.uk
- For enquiries about the application process, please contact: research.admissions@bcu.ac.uk