Predicting Small-Scale Building Energy Consumption in Early Design Stages in Mediterranean Climates: A Comparative Analysis of Machine Learning Models

Abstract

Accurately forecasting residential energy demand during conceptual architectural design is vital for building decarbonization, yet early models often struggle with balancing input complexity and predictive speed. In this work, we developed a machine learning-driven decision-support framework evaluating five regression architectures Random Forest (RF), Artificial Neural Networks (ANN), Decision Trees (DT), Support Vector Regression (SVR), and Multiple Linear Regression (MLR) across a dataset tailored to Mediterranean residential structures. Eleven architectural and environmental variables were initially evaluated before applying dimensionality reduction to examine how parameter depth affects overall model precision. The results directly challenge the conventional assumption that richer parameter sets automatically improve performance. Rather, reducing input dimensionality down to four decisive parameters floor area, cooling demand, climate classification, and envelope insulation quality minimized data noise and maximized generalization capability, reaching high accuracy (R²> 0.95). Furthermore, introducing vertical spatial analysis showed that story height substantially alters energy behavior: intermediate floors function as passive thermal buffers with stable, minimal loads, whereas envelope-adjacent ground and roof levels exhibit heightened thermal variance and elevated demand. Consequently, embedding vertical floor orientation alongside condensed key parameters establishes a lean, highly reliable protocol for early-stage energy evaluation. Overall, the proposed framework demonstrates that reliable and high-precision energy predictions can be achieved using a limited set of performance-oriented design parameters. The proposed approach provides architects and designers with an efficient, data-driven tool for evaluating energy performance without relying on computationally intensive simulations during the early design process.

Keywords

• Small-scale building energy
• Machine learning
• Architectural design
• Mediterranean climate
• Vertical mass analysis

Project Number

None

Ethical Statement

Ethics committee approval was not required for this study because it did not involve human participants or animals.

Thanks

The author declares that no specific grant or financial support was received from any funding agency in the public, commercial, or non-profit sectors for the research, authorship, and/or publication of this article.

References

1. Ahmed, B., Ahmed, A., Mujeeb, A., Zhang, H. N., Li, X. B., Qu, K. Y., & Li, F. C. (2026). Dynamic occupancy-aware HVAC control in large office building using enhanced soft actor-critic with PV and thermal energy storage integration. Applied Thermal Engineering, 280, Article 129937. https://doi.org/10.1016/j.applthermaleng.2026.129937
2. Ak, M. F. (2020). A comparative analysis of breast cancer detection and diagnosis using data visualization and machine learning applications. Healthcare, 8(2), Article 111. https://doi.org/10.3390/healthcare8020111
3. Al-Jamal, M. Q., Alsarhan, A., Aljamal, M., Aljamal, Q., Khassawneh, B. S., Salhi, A., & Hayat, H. (2026). Integrating generative design and artificial intelligence for optimized energy-efficient composite facades in next-generation smart buildings. Sustainability, 18(5), Article 2379. https://doi.org/10.3390/su18052379
4. Chen, Y. (2026). Leveraging LSTM networks for vehicle stability prediction: A comparative analysis with traditional models under dynamic load conditions. The Journal of Applied Engineering and Technologies, 1(1), 15–28.
5. Diz-Mellado, E., Ruiz-Pardo, Á., Rivera-Gómez, C., de la Flor, F. J. S., & Galán-Marín, C. (2023). Unravelling the impact of courtyard geometry on cooling energy consumption in buildings. Building and Environment, 237, Article 110349. https://doi.org/10.1016/j.buildenv.2023.110349
6. Doğan, O. (2016). Uyarlamalı Sinirsel Bulanık Çıkarım Sisteminin (ANFIS) talep tahmini için kullanımı ve bir uygulama [Usage of Adaptive Neuro-Fuzzy Inference System (ANFIS) for demand forecasting and an application]. Dokuz Eylül Üniversitesi İktisadi İdari Bilimler Fakültesi Dergisi, 31(1), 257–288. https://doi.org/10.24988/deuiibf.2016311513
7. Hamidy, S. M., Erik, A., Kuvvetli, Y., & Gökşen, G. (2026). Artificial intelligence and machine learning for food safety and traceability. In Y. Özoğul & S. M. Hamidy (Eds.), Artificial intelligence in food science (pp. 589–609). Academic Press. https://doi.org/10.1016/B978-0-443-26468-9.00046-1
8. Hamidy, S. M., Kuvvetli, Y., Sakarya, Y., Özkütük, S. T., & Özoğul, Y. (2025). Deep learning-based prediction of fish freshness and purchasability using multi-angle image data. Foods, 15(1), Article 68. https://doi.org/10.3390/foods15010068

9. Hobor, L., Brčić, M., Polutnik, L., & Kapetanović, A. (2026). Comparative analysis of modern machine learning models for retail sales forecasting. Croatian Operational Research Review, 17(2), 145–158. https://doi.org/10.17535/crorr.2026.0021
10. Kuran, F., Tanırcan, G., & Pashaei, E. (2024). Developing machine learning-based ground motion models to predict peak ground velocity in Turkiye. Journal of Seismology, 28(5), 1183–1204. https://doi.org/10.1007/s10950-024-10234-x
11. Liang, Y., Fang, Z., & Xu, Z. (2026). 3D reconstruction and building energy modeling for existing buildings via image and masked Gaussian Splatting. Journal of Building Engineering, 118, Article 115418. https://doi.org/10.1016/j.jobe.2026.115418
12. Olu-Ajayi, R., Alaka, H., Sulaimon, I., Balogun, H., Wusu, G., Yusuf, W., & Adegoke, M. (2023). Building energy performance prediction: A reliability analysis and evaluation of feature selection methods. Expert Systems with Applications, 225, Article 120109. https://doi.org/10.1016/j.eswa.2023.120109
13. Ortiz, L., Gonzalez, J. E., & Lin, W. (2018). Climate change impacts on peak building cooling energy demand in a coastal megacity. Environmental Research Letters, 13(9), Article 094008. https://doi.org/10.1088/1748-9326/aad8d0
14. Pacheco-Torres, R., López-Alonso, M., Martínez, G., & Ordóñez, J. (2015). Efficient design of residential buildings geometry to optimize photovoltaic energy generation and energy demand in a warm Mediterranean climate. Energy Efficiency, 8(1), 65–84. https://doi.org/10.1007/s12053-014-9275-5
15. Renaud, O., & Victoria-Feser, M. P. (2010). A robust coefficient of determination for regression. Journal of Statistical Planning and Inference, 140(7), 1852–1862. https://doi.org/10.1016/j.jspi.2010.01.008
16. Senel Solmaz, A. (2026). A machine learning-based early design energy prediction framework for school buildings across multiple climatic regions of Türkiye. Buildings, 16(4), Article 779. https://doi.org/10.3390/buildings16040779
17. Siqueira, M. V., Ferreira, V. H., & Colombini, A. C. (2025). Long term wind energy forecasting using machine learning techniques. Global Energy Interconnection, 8(3), 312–324. https://doi.org/10.1016/j.gloei.2025.05.013
18. Song, G., Xiao, S., Li, Z., Zhi, X., Fan, F., & Xu, T. (2026). Experimental and numerical investigation on PCM-integrated roofs for enhancing energy efficiency in large-span buildings. Applied Thermal Engineering, 281, Article 130343. https://doi.org/10.1016/j.applthermaleng.2026.130343
19. Yıldırım, S., Tosun, E., Çalık, A., Uluocak, İ., & Avşar, E. (2019). Artificial intelligence techniques for the vibration, noise, and emission characteristics of a hydrogen-enriched diesel engine. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, 41(18), 2194–2206. https://doi.org/10.1080/15567036.2018.1550540
20. Zhang, G., Patuwo, B. E., & Hu, M. Y. (1998). Forecasting with artificial neural networks: The state of the art. International Journal of Forecasting, 14(1), 35–62. https://doi.org/10.1016/S0169-2070(97)00044-7
21. Zhang, Z., Sun, H., & Huang, T. (2025). Spatiotemporal analysis of building energy performance and urban morphology interactions across Chinese climate zones: A machine learning approach to predicting cooling demand in high-density districts. Case Studies in Thermal Engineering, 71, Article 107592. https://doi.org/10.1016/j.csite.2025.107592