A trustworthy by design classification model for building energy retrofit decision support

Improving energy efficiency is a major concern in residential buildings for economic prosperity and environmental stability. Despite growing interest in this area, limited research has been conducted to systematically identify the primary factors that influence residential energy efficiency at scale, leaving a significant research gap. This paper addresses the gap by exploring the key determinant factors of energy efficiency in residential properties using a large-scale energy performance certificate dataset. Dimensionality reduction and feature selection techniques were used to pinpoint the key predictors of energy efficiency. The consistent results emphasise the importance of CO2 emissions per floor area, current energy consumption, heating cost current, and CO2 emissions current as primary determinants, alongside factors such as total floor area, lighting cost, and heated rooms. Further, machine learning models revealed that Random Forest, Gradient Boosting, XGBoost, and LightGBM deliver the lowest mean square error scores of 6.305, 6.023, 7.733, 5.477, and 5.575, respectively, and demonstrated the effectiveness of advanced algorithms in forecasting energy performance. These findings provide valuable data-driven insights for stakeholders seeking to enhance energy efficiency in residential buildings. Additionally, a customised machine learning interface was developed to visualise the multifaceted data analyses and model evaluations, promoting informed decision-making.

The multifamily housing stock in Russia has a high degree of wear and tear, low energy efficiency, and inadequate maintenance, which leads to over-consumption of energy resources. High energy intensity is related to the fact that energy equipment is obsolete, high heat and energy losses during transportation occur, energy is spent by enterprises and population unreasonably, buildings and constructions have high heat losses. The main reasons for the efficient use of energy in the building sector is to reduce the heating and hot water costs of building owners, improve the indoor climate of buildings, save taxpayers' money in order to use the saved funds in other areas, introduce energy efficient and renewable energy technologies, improve the air quality, reduce the negative impact on the environment and climate change Based on the above mentioned, the article examines the aspects of energy efficiency of residential buildings, namely identifying the problems of energy efficiency, the main parameters of the state programme of energy efficiency of residential and public buildings. An example of potential incentives, comprehensive and reasonable use of incentives, which can change the current situation in the shortest possible time, is presented.

The paper contains the analysis of the goals and objectives of improving energy efficiency of residential buildings, as well as the methodology of selecting energy-efficient modernization measures for historic buildings. The priority objective of this study was selected as a residential housing energy efficiency of historic buildings as a tool to improve the quality of the human environment. The development of renewable energy technologies is presented as an alternative for energy saving. If we take the purpose of improving energy efficiency as an improvement of the quality of the human environment (from the living quarters level to global environmental sustainability), the alternative to energy saving of traditional energy resources can be the replacement of them by energy from renewable sources, even lavish spending of which does not harm the environment . All energy saving should be focused primarily in man-made environments (industrial processes, heating systems and etc.), the anthropogenic environment (living environment) should get the maximum energy for the stable provision of quality microclimate.

Achieving the European Union's ambitious goal of transforming buildings into carbon-neutral assets is crucial for combating climate change and improving energy efficiency. The harmonization of Energy Performance Certificates (EPCs) plays a vital role in this effort. It can provide a standardized measure of building energy performance, which is crucial for benchmarking and improving energy efficiency across Member States. In the context of the recast Energy Performance of Buildings Directive (EPBD) of 2024, this study examines the progress made by EU Member States in standardizing EPCs. The research aims to evaluate the implementation of EPC indicators, focusing on data quality, calculation methods, and costs. The methodological procedure followed for the study led to the development of a dedicated Cross Comparative Matrix (CCM) to assess the implementation of EPC indicators across 27 EU Member States and their reading feature allowed the declination into three scenarios for different EPC data analysis: qualitative, quantitative and burden costs. The findings reveal significant disparities, particularly in areas such as thermal comfort and smart readiness, due to the lack of a common methodology and local regulatory differences. While key indicators like energy performance class and primary energy use are widely implemented, compliance with EU standards varies. The study concludes by recommending standardized methodologies and enhanced assessor training to improve EPC quality, harmonization, and effectiveness. This research contributes to policy discussions by offering a comprehensive framework for EPC evaluation and insights into improving data quality, reliability, and accessibility. The originality of this study lies in its cross-national analysis, highlighting the need for a unified EPC scheme to promote building sustainability across the EU.

One of the most important strategic objectives is to create a sustainable model for the development of the Kazakhstan economy, which is closely linked to the solution of energy saving and energy efficiency in the construction industry. The main principles of the energy efficiency strategy include integrated approach (efficient use of energy at all stages of the facility's life cycle for all types of energy resources, with legal support and organizational and technical control by the state) and systematic approach (the study of the process using general rather than partial approach). The article analyzes key problems and barriers that prevent the successful implementation of energy saving and energy efficiency policies. A significant role in the rational use of energy resources belongs to the regulatory framework. It was revealed that currently existing regulatory and technical documents in the country, unlike similar documents in EU countries, have a number of inaccuracies and disadvantages that do not allow for a full accounting of energy consumption. At present, there is practically no effective methodology for determining the energy efficiency of residential buildings and structures in Kazakhstan. In this regard, a methodology for determining the energy efficiency of buildings and structures, harmonized with EU requirements (taking into account energy consumption for heating, ventilation, high-temperature water supply, cooling and lighting) is currently being developed. The energy efficiency improvement of new, renovated and operated buildings will make it possible to reduce greenhouse gas emissions, thus contributing to solving both global environmental problems and energy and environmental security in the Republic of Kazakhstan.

Purpose This study aims to explore the impact of energy efficiency on rental prices with advanced predictive modeling techniques. It aims to analyze the evolution of energy efficiency’s role in Belgian residential rental markets from 2014 to 2023 using machine learning (ML) and explainable AI (XAI). Design/methodology/approach Using a cleaned data set of 636,848 Belgian rental listings from 2014 to 2023 with inflation-adjusted rental prices, this study analyzes XGBoost models and compares their predictive performance with ordinary least squares (OLS). Furthermore, it uses Shapley additive explanations (SHAP) to interpret models and assess the changing influence of energy performance certificate (EPC) scores on rental prices over time. Findings XGBoost consistently outperforms OLS in predictive performance, capturing complex nonlinear relationships. EPC scores exhibit a growing influence on rental prices over time, although their impact remains modest compared to key drivers such as postal code and habitable area. Results indicate that while energy efficiency’s role has increased, it has not yet become a dominant factor in rental pricing. Originality/value This study contributes to both ML-based real estate analytics and policy discussions on sustainable housing. To the best of the authors’ knowledge, it is the first to apply XGBoost and SHAP to assess the temporal evolution of energy efficiency’s impact on rental prices. In addition, the findings suggest that current market conditions or policies may not sufficiently incentivize landlords to invest in energy efficiency improvements, which might underscore the need for stronger regulations to better align with climate goals.

Focus issue: Artificial intelligence in medical physics.

The policy of European Union is focused mostly on green growth, therefore there is an increasing number of regulations concerning sustainable development. The situation is the same in the matter of life cycle of buildings, where pressure is put on eco-friendly solutions. The purpose of the paper is to describe the green residential real estate market in Wroclaw's county and to characterize its main attributes.The paper closely examines put to use, 'green' residential properties. The research studies statistical and real estate data from government agencies, such as District Inspector of Building Control for Wroclaw's county and Cadastre office of Wroclaw's county. The information gathered allows marking out energy efficient residential buildings and determining their value. The research shows that 'green' residential buildings have just a small share in the real estate market of new buildings. The result of the study indicates that just a few of those buildings were sold by the end of the year 2012. Furthermore almost all of 'green' houses were constructed by their owners. An interest in energy efficiency can be the result of the increased willingness to reduce the operating costs of the property. By analyzing the information gathered, the most popular eco-friendly solutions used in buildings (such as: solar panels, heat pumps, biomass boilers) were identified. The research estimates the actual value of 'green' buildings. One of the research limitations was the quality of Energy Performance Certificates (EPC). Even if the EPC was done according to the national approach it may give misleading information about energy efficiency of the building. In the future a further research on the effectiveness and correctness of EPC is necessary. The paper is unique in the field of real estate in Poland as it concentrates on the residential real estate in regards to sustainable development. The results of the research can be useful to valuers and investors since it gives genuine data on the green residential buildings market in Wroclaw's county.

The Minimum Energy Efficiency Standards (MEES) set out minimum energy standards for rented property. Those standards are expressed in the form of an Energy Performance Certificate (EPC) rating, so securing a good EPC rating is critical, as poorly performing buildings will become unlettable. It is vital that the EPC truly reflects the building’s construction and services, rather than relying on inaccurate assumptions, which will often return a poorer EPC. Also, the EPC’s Recommendations Report should steer away from an automatically generated list of energy efficiency measures (which may well be completely irrelevant to the building), towards a listing specific to the building and its location. Key to MEES compliance in a building with a poor EPC rating is taking the time to properly assess a number of energy improvement measures to find cost-effective improvements that deliver the required improvement in EPC rating. Careful analysis of various options can reveal that some ‘easy fixes’, such as upgrading lighting or an air-conditioning system, may not deliver enough of an improvement over existing installations, or they may be prohibitively costly, negating any savings overall. Exactly who pays for such energy efficiency improvements is a complex, multifaceted dilapidations question encompassing issues such as repair, statutory compliance, reinstatement/alterations, damages/supersession and loss of rent. Depending on the circumstances, the costs associated with energy improvements are more likely to rest with the landlord — at least currently. The situation is fluid, however, as future leases may well be drafted to include provisions relating to energy efficiency improvements.

Private landlords and energy efficiency: Evidence for policymakers from a large-scale study in the United Kingdom

Predictive modeling in reproductive medicine: Where will the future of artificial intelligence research take us?

A large amount of personal health data that is highly valuable to the scientific community is still not accessible or requires a lengthy request process due to privacy concerns and legal restrictions. As a solution, synthetic data has been studied and proposed to be a promising alternative to this issue. However, generating realistic and privacy-preserving synthetic personal health data retains challenges such as simulating the characteristics of the patients’ data that are in the minority classes, capturing the relations among variables in imbalanced data and transferring them to the synthetic data, and preserving individual patients’ privacy. In this paper, we propose a differentially private conditional Generative Adversarial Network model (DP-CGANS) consisting of data transformation, sampling, conditioning, and network training to generate realistic and privacy-preserving personal data. Our model distinguishes categorical and continuous variables and transforms them into latent space separately for better training performance. We tackle the unique challenges of generating synthetic patient data due to the special data characteristics of personal health data. For example, patients with a certain disease are typically the minority in the dataset and the relations among variables are crucial to be observed. Our model is structured with a conditional vector as an additional input to present the minority class in the imbalanced data and maximally capture the dependency between variables. Moreover, we inject statistical noise into the gradients in the networking training process of DP-CGANS to provide a differential privacy guarantee. We extensively evaluate our model with state-of-the-art generative models on personal socio-economic datasets and real-world personal health datasets in terms of statistical similarity, machine learning performance, and privacy measurement. We demonstrate that our model outperforms other comparable models, especially in capturing the dependence between variables. Finally, we present the balance between data utility and privacy in synthetic data generation considering the different data structures and characteristics of real-world personal health data such as imbalanced classes, abnormal distributions, and data sparsity.

A building energy performance gap can be illustrated as the difference between the theoretical (methodologically defined) and the actual energy consumption. In EU countries, Energy Performance Certificates are issued when buildings are constructed, sold, or leased. This information is the first step in order to evaluate the energy performance of the building stock. In Serbia, when issuing an energy certificate, the adopted national methodology recognizes only energy consumption for heating. The main purpose of this paper is to evaluate the energy gap and estimate the relevance of an Energy Performance Certificate to meet the national energy efficiency or carbon target. An Energy Performance Certificate determines the theoretical residential and commercial building energy efficiency or its “design intent”. This research stresses the necessity of measuring and achieving reductions in actual energy consumption through system regulation and consumers’ self-awareness in buildings. The research compares the performance of the building stock (135) that is connected to the District Heating System (DHS), with its own integrated heat meter, to Individual Gas Boiler (IGB) systems (18), in the city of Novi Sad, Serbia, built after 2014. For the purpose of comparing energy consumption, 16 buildings were selected that are very similar in terms of design, operation, and location. The data used are derived from metered consumption data, official evidence of city service companies, and Energy Performance Certificates of the considered buildings. We have determined that IGB systems have a much wider specific annual performance gap (11.19–101 kWh/m2a) than the buildings in the DHS (3.16–18.58 kWh/m2a).

Artificial intelligence (AI) hallucinations occur when large language models, such as chatbots or computer vision systems, generate outputs containing non-existent patterns, leading to inaccurate results. Also known as AI confabulations or delusions, these instances challenge expectations of appropriate responses from AI tools due to unrelated or pattern-lacking outputs, similar to human hallucinations. Addressing such issues with generative AI presents significant challenges despite ongoing efforts to resolve them.[1,2] CAUSES OF AI HALLUCINATIONS Various causes of AI hallucinations have been identified and include: Insufficient or biased training data: An AI model designed to assist anaesthesiologists in administering anaesthesia may be trained predominantly on data from patients of a certain demographic, such as adults of average weight. When faced with a paediatric patient or an obese patient, the AI model may possibly hallucinate dosage recommendations that are inaccurate or unsafe, as it lacks sufficient exposure to diverse patient populations.[3] Model complexity: A highly complex AI system tasked with monitoring vital signs during surgery may exhibit hallucinatory responses when encountering unusual physiological patterns. This complexity underscores the need for simpler models to avoid such hallucinations.[4] Lack of explainability (black box): An AI algorithm designed to predict anaesthesia induction times may produce unexpectedly long or short estimates without providing clear explanations for its predictions. In cases where anaesthesiologists cannot understand or verify the AI system’s reasoning, there is a risk of blindly following its recommendations, potentially leading to errors or patient harm. This highlights the urgent need for explainable AI in anaesthesia.[5] MULTIFACETED THREAT OF AI HALLUCINATIONS IN ANAESTHESIA An AI hallucination occurs when an AI system produces demonstrably incorrect or misleading outputs, appearing confident and plausible despite factually flawed. The possible impacts of AI hallucinations on anaesthesia domains are varied[6-9] [Table 1].Table 1: Examples of AI hallucinations’ possible impact on anaesthesia domainsMisdiagnosis and mistreatment: Hallucinations can misinterpret patient data, resulting in unnecessary interventions or delayed treatments. Medication errors: AI-driven systems may recommend incorrect drug dosages, impacting patient safety. Communication and documentation: Misinterpreted verbal commands or procedure details can hinder accurate documentation and patient safety. Research skewing: AI-driven analysis of anaesthesia data for research could be skewed by hallucinations, leading to misleading conclusions. Legal and ethical concerns: Liability: Who is responsible for the errors caused by AI hallucinations? This remains a complex question with no clear answer. Depending on the specific circumstances, potential targets include the AI developer, healthcare provider or hospital. Informed consent: How can patients be adequately informed about the risks of AI hallucinations in anaesthesia, given the technical complexity involved and the dynamic nature of AI outputs? Striking a balance between transparency and patient anxiety is crucial. Bias: AI algorithms can perpetuate societal biases, leading to discriminatory outcomes in health care. Imagine an AI system trained on biased data; it might recommend different treatments based on a patient’s race or socioeconomic background.[10-12] STRATEGIES TO MITIGATE AI HALLUCINATIONS Various mitigation strategies need to be adhered to for the impact of AI hallucination on health care [Figure 1].Figure 1: Impact of AI hallucination on health care and mitigation strategies. AI = artificial intelligenceHigh-quality, diverse training data: Utilising diverse datasets improves AI model accuracy and reduces hallucination risks. For example, research by Jones et al.[13] demonstrated how incorporating various demographic factors and medical histories in training data significantly improved the accuracy of an AI-driven diagnostic tool for skin cancer detection. Explainable AI: Developing transparent AI models aids in identifying and rectifying hallucinations. For instance, the explainable nature of a deep learning model used in financial fraud detection allowed analysts to trace back erroneous predictions to specific data points, enabling targeted adjustments to the model’s training data and architecture.[14] Human oversight and collaboration: Human involvement reduces hallucination risks, especially in sensitive domains like health care. Collaborative efforts between AI systems and human experts have effectively reduced hallucination risks.[15] Continuous monitoring and evaluation: Regular evaluation detects and addresses hallucinations promptly. Continuous monitoring of its AI-powered recommendation system and real-time user feedback analysis allows for swift identification and correction of hallucinated product suggestions, improving user satisfaction and trust.[16] Algorithmic auditing and regulatory frameworks: Establishing robust auditing mechanisms and regulatory frameworks ensures AI system’s accountability and reliability.[17] To conclude, AI hallucinations in anaesthesia pose risks of misdiagnosis, medication errors and skewed research outcomes. Prioritising diverse training data, embracing explainable AI, maintaining human oversight, continuous monitoring and regulatory frameworks are crucial in mitigating these risks and fostering trust in AI technologies in health care. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.

<span lang="EN-US">The burgeoning field of artificial intelligence (AI) necessitates a nuanced approach to governance that integrates technological advancement, ethical considerations, and regulatory oversight. As various AI governance frameworks emerge, a fragmented landscape hinders effective implementation. This article examines the driving forces behind AI regulation and the essential control mechanisms that underpin these frameworks. We analyze market-driven, state-driven, and rights-driven regulatory approaches, focusing on their underlying motivations. Furthermore, critical regulatory controls such as data governance, risk management, and human oversight are highlighted to demonstrate their roles in establishing effective governance structures. Additionally, the importance of international cooperation and stakeholder collaboration in addressing the challenges posed by rapid technological change is emphasized. By providing insights into the strengths, weaknesses, and potential synergies of different governance models, this study contributes to the development of equitable and effective AI regulatory frameworks that encourage innovation while safeguarding societal interests. Ultimately, the findings aim to inform policymakers, industry leaders, and civil society organizations in their efforts to foster a future where AI is utilized responsibly and equitably for the betterment of humanity.</span>