The increasing demand for energy efficiency in buildings has driven the need for innovative design approaches that optimize both architectural and energy performance. Parametric design offers a flexible and efficient method for generating and evaluating multiple design alternatives, allowing for the integration of energy performance criteria from the early design stages.

Current design practices often rely on predefined rules and static models, which limit the ability to make real-time adjustments based on energy performance analysis. Parametric modeling, however, provides a dynamic and iterative approach, enabling designers to explore various configurations of geometry, materials, façade elements, insulation strategies, shading devices, and building orientation to enhance energy efficiency. By linking parametric tools with energy simulation software and incorporating predefined energy-efficiency targets, the impact of design choices on heating, cooling, and daylighting performance can be assessed more effectively.

This research presents a state-of-the-art review of parametric design applications in energy-efficient buildings, critically analyzing its advantages and limitations. The study explores how parametric modeling facilitates a more responsive and adaptable design process, allowing for real-time optimization of building elements to minimize energy use. Through a systematic evaluation of case studies and existing methodologies, this research identifies the key benefits and challenges of parametric workflows, particularly in the context of low-energy and nearly Zero Energy Buildings (nZEBs).

A critical analysis highlights the main advantages of parametric design, such as enhanced flexibility, rapid performance evaluation, and multi-objective optimization. However, the study also addresses several limitations, including the complexity of implementation, the need for advanced computational knowledge, and the risk of excessive reliance on automated processes without considering real-world constraints.

The findings emphasize the role of parametric workflows in improving design accuracy, adaptability, and energy performance optimization, while also recognizing the challenges that must be addressed for broader adoption in the construction industry. This research contributes to the ongoing discourse on digital design methodologies, providing insights into how parametric tools can be effectively integrated into sustainable building practices.

The traditional heating load calculation, as per DIN EN 12831, is widely used for the design of supply and distribution systems in buildings. However, this static, worst-case scenario often results in significant oversizing, leading to inefficiencies and resource waste. In the context of the energy transition, such oversizing is unsustainable. Regenerative energy systems operate dynamically, necessitating planning approaches that also consider dynamic behavior. Both the energy demand for heating and the supply from renewable energy sources are highly variable and must be integrated into a responsive and flexible planning framework.

This research employs advanced simulation techniques to address this need. Building component simulation is used to model heat transport processes within construction elements, enabling the evaluation of temperature distributions and thermal comfort requirements. Complementarily, building-level simulation provides a holistic view, allowing for dynamic system optimization.

Two specific practical challenges were analyzed:

• Optimization of installation distances for surface heating systems, particularly underfloor heating (UFH).

• Evaluation of minimal insulation for UFH supply pipes in corridors to reduce heat losses effectively.

A range of simulation models was developed to assess and refine the boundary conditions defined in DIN EN 12831. This approach facilitated the proposal of more dynamic methodologies for heating load calculations. The results highlight opportunities to improve system design by aligning installation strategies and insulation measures with dynamic operational demands. Additionally, the plausibility of very narrow installation distances and minimal insulation strategies was critically reviewed to ensure both energy efficiency and practicality.

The novelty of this work lies in its dynamic approach to heating load analysis, bridging the gap between static standards and the demands of regenerative energy integration. The methodology demonstrates broad applicability in optimizing both component-level and system-level performance. This research offers actionable insights for designing energy-efficient and sustainable building systems, contributing to a future-proof built environment.

This study explores the application of Model Predictive Control (MPC) in building management systems as a strategy for achieving significant energy savings and reducing greenhouse gas emissions compared to traditional control methods. The primary advantage of MPC lies in its ability to integrate future demand trends and operational constraints into the optimization process. The predictive control approach is based on a simplified gray-box mathematical model that is proposed to effectively capture the thermal dynamics of the building. The thermal behaviour is represented using first-order linear differential equations that account for the building’s thermal capacitance and external disturbances. Thermal model parameters are calibrated using data from DesignBuilder simulations to ensure accuracy.

The control methodology optimizes input sequences by minimizing a predefined objective function at 15-minute intervals, balancing temperature comfort and energy efficiency constraints. The study presents a workflow adaptable to various residential buildings, enabling the implementation of model-based controllers for heating systems. The performance of the proposed controller is validated through co-simulation using EnergyPlus and Matlab, with a case study of a single-zone building in Hungary. The manipulated input is the heat emitted by an electric underfloor heating system, while external disturbances include occupancy, other heat sources and weather conditions. The study assesses the performance of the controller over several-day periods with different weather characteristics—overcast, clear-sky, and extreme cold conditions—as well as the entire heating season. For the latter, results show that MPC control can reduce heating energy consumption by 4.5%, improve temperature reference tracking accuracy by 50%, and significantly decrease peak energy demand.

The proposed method provides a foundation for model-based control of buildings with renewable energy integration and thermal energy storage. Furthermore, it offers potential for optimizing the coordinated operation of energy communities through consumer-side controls, contributing to active solutions for load matching challenges.

Fluctuations in temperature and relative humidity within ancient tombs can lead to significant degradation of mural artworks, including salt efflo-rescence, weathering, delamination, and other forms of deterioration, there-by severely compromising their cultural and historical value. In the case of the Southern Tang Dynasty (Shunling Tomb) in Nanjing, Jiangsu Province, China, which are open to public visitation, rapid and accurate prediction of the indoor temperature and humidity is essential for the preventive conser-vation of these murals. However, conventional models based on heat-moisture coupling physics often suffer from inefficiency and lack of real-time applicability. To address these limitations, this study proposes an in-terpretable, high-performance machine learning-based model for real-time prediction of indoor environmental conditions in tombs. This model aims to meet the preventive conservation requirements of significant cultural heritage sites, as well as optimize visitor access management and environ-mental control. In this study, several machine learning algorithms are em-ployed to predict temperature and relative humidity within the Southern Tang Tombs, using hourly outdoor meteorological data as input. Model per-formance is evaluated based on accuracy and precision, with SHapley Addi-tive Explanations (SHAP) used to interpret the model outputs and quantify the influence of different input parameters. The findings indicate that the XGBoost model exhibits the highest accuracy in predicting indoor tempera-ture, while CatBoost is more effective for real-time humidity prediction. SHAP analysis identifies soil cover properties and ambient thermal parame-ters as the governing factors in tomb microclimate regulation. The devel-oped machine learning framework provides an efficient and accurate analyt-ical solution for cultural heritage microclimate studies. Conservation strat-egies should focus on optimizing overburden configuration and implement-ing real-time thermal environment monitoring systems.

Effective monitoring of architectural surfaces is essential for timely maintenance planning. Accurate and systematic assessments enable early intervention, preventing severe deterioration and costly repairs. Traditional inspections rely on visual examinations by skilled technicians, making them subjective, time-consuming and limited to what is visible. In such perspective, Spectral Imaging (SI) is emerging in the construction sector as a valuable advanced diagnostic tool. By integrating spectral and spatial da-ta, SI provides a reliable, non-destructive method for detecting material al-terations with high precision. SI delivers spatially referenced data, enables automated analysis of decay phenomena, and, by leveraging wavelengths beyond the visible spectrum, identifies changes before they become appar-ent to the human eye. This study presents a real-world case demonstrating the workflow for detecting and classifying alterations and decay on archi-tectural surfaces using SI. Hyperspectral images—meaning comprising hundreds of spectral bands—of a concrete surface on a university building in Ancona, Italy, were acquired using a sensor covering the 350–1000 nm spectral range. After optical and spectral calibration, data were processed using a Machine Learning (ML) algorithm, producing a segmented classi-fication of surface alterations. The results were compared to a manual sur-vey by a technician. The SI-based assessment proved to be accurate, com-prehensive, and significantly more time-efficient than the traditional ap-proach. This study underscores the potential of SI as a powerful tool for au-tomated architectural diagnostics, facilitating proactive maintenance strat-egies and optimizing conservation efforts in the built environment.

Buildings are complex assets with evolving uses, variable occupancies, and long-life cycles that drive high operational costs and substantial energy demand. Globally, they account for roughly 30–40% of greenhouse-gas emissions, making accurate short-term forecasting crucial to reduce both economic and environmental impacts. While Machine Learning (ML) is increasingly adopted, purely data-driven models often require large datasets and may violate physical constraints—especially in data-scarce contexts.

This work proposes a hybrid predictive framework that couples Physics-Informed Neural Networks (PINNs) with dynamic energy simulation and energy-balance constraints to forecast, one hour ahead, indoor operative temperature (To) and cooling electricity (Eel,Cool) in a Multi-Storey Residential Building in Southern Italy. The PINN operates in two phases: Phase I predicts To,t+1 and then estimates Eel,Cool,t+1, embedding thermodynamic consistency into the loss to reduce reliance on computationally expensive simulations and deliver physically plausible predictions.

On the test set, the PINN attains 0.265 °C RMSE for To (CVRMSE = 0.98%, MBE = −0.002 °C, NMBE = −0.01%) and 0.185 kWh RMSE for Eel,Cool (CVRMSE = 13.10%, MBE = 0.066 kW, NMBE = 4.65%). Benchmarked against established ML baselines, MLP, Random Forest, and Linear Regression, trained with identical preprocessing and splits, the PINN delivers sub-0.3 °C temperature errors and reduces energy RMSE by 66–81% relative to the best baselines.

These results demonstrate that PINNs can bridge data scarcity and physical interpretability, enabling robust one-hour-ahead forecasts that support proactive control towards sustainability targets.