Integrating Post-Occupancy Evaluation And Energy Performance Through GANs For Human-Centered School Design And Enhanced Satisfaction

Subject Areas :

Marjan Ilbeigi ¹ , Mohammad Ghomeishi ^{2 *} , Ali Asgharzadeh ³ , Nima Amani ⁴

1 - Department of Architecture, Cha.C., Islamic Azad University, Chalous, Iran.
2 - Department of Architecture, Dam.C., Islamic Azad University, Damavand, Iran.
3 - Department of Civil Engineering, Cha.C., Islamic Azad University, Chalous, Iran.
4 - Department of Civil Engineering, Cha.C., Islamic Azad University, Chalous, Iran.

Received: 2025-08-20 Accepted : 2025-09-02 Published : 2025-09-16

Keywords: Satisfaction, Energy, Spatial Layout, User-Centered,

Abstract :

This study proposes a hybrid framework for generating user-centered and energy-efficient architectural patterns for primary schools by integrating Generative Adversarial Networks (GANs) with Post-Occupancy Evaluation (POE). Based on empirical data from 384 users in several public schools in Tehran, the POE results identified ventilation (r = 0.586), lighting quality, and furniture/materials (r = 0.701) as the most influential factors on user satisfaction. Factor analysis revealed latent features such as circulation clarity, sensory comfort, and spatial quality. The GAN model was trained over 500 epochs on these data and produced synthetic layouts that prioritized user preferences across six functional zones (e.g., administrative, workshop, and classrooms). In the generated layouts, up to 34% of the area was allocated to classrooms, making them the dominant spaces. A comparison between GAN-generated layouts and baseline school samples showed that EnergyPlus simulations predicted an average 22% reduction in annual energy consumption, particularly in heating, cooling, and lighting. Adjacency with service areas, window-to-wall ratio, and classroom orientation significantly influenced thermal performance and natural daylighting. Additionally, a feedback loop was employed to iteratively refine GAN outputs using Mean Squared Error (MSE) and Structural Similarity Index (SSIM) for improved spatial accuracy. The results confirm the effectiveness of integrating GAN and POE in predictive, performance-driven school design. By facilitating decision-making in the early design stages, the proposed framework offers a scalable approach to flexible and sustainable educational architecture that balances environmental objectives with user experience.

References:

-Bayer, J., Bukhari, S. S., & Dengel, A. (2018). Floor Plan Generation and Auto Completion Based on Recurrent Neural Networks. Proceedings of the International Conference on Document Analysis and Recognition, ICDAR, 2, 49–50. https://doi.org/10.1109/ICDAR.2017.272
-Belpoliti, V., Saleem, A. A., Yahia, M. W., & Nassif, R. (2024). Energy consumption of UAE public schools. Mapping of a diversified sector assessing typology, conditions, and educational systems. Energy and Buildings, 320, 114599. https://doi.org/10.1016/j.enbuild.2024.114599
-Buldur, S., Bursal, M., Yalcin Erik, N., & Yucel, E. (2020). The impact of an outdoor education project on middle school students’ perceptions and awareness of the renewable energy. Renewable and Sustainable Energy Reviews, 134. https://doi.org/10.1016/j.rser.2020.110364
-Caetano, I., Santos, L., & Leitão, A. (2020). Computational design in architecture: Defining parametric, generative, and algorithmic design. Frontiers of Architectural Research, 9(2), 287–300. https://doi.org/10.1016/j.foar.2019.12.008
-Caldas, L. G., & Norford, L. K. (2002). A design optimization tool based on a genetic algorithm. Automation in Construction, 11(2), 173–184. https://doi.org/10.1016/S0926-5805(00)00096-0
-Chung, W., & Yeung, I. M. H. (2020). A study of energy consumption of secondary school buildings in Hong Kong. Energy and Buildings, 226, 110388. https://doi.org/10.1016/j.enbuild.2020.110388
-Dogan, T., Saratsis, E., & Reinhart, C. (2015). The optimization potential of floor-plan typologies in early design energy modeling. 14th International Conference of IBPSA - Building Simulation 2015, BS 2015, Conference Proceedings, 1853–1860.
-Dong, H., Neekhara, P., Wu, C., & Guo, Y. (2017). Unsupervised Image-to-Image Translation with Generative Adversarial Networks. https://arxiv.org/pdf/1701.02676
-Egor, G., Sven, S., Martin, D., & Reinhard, K. (2020). Computer-aided approach to public buildings floor plan generation. Magnetizing Floor Plan Generator. Procedia Manufacturing, 44, 132–139. https://doi.org/10.1016/j.promfg.2020.02.214
-Eisenstadt, V., Langenhan, C., & Althof, K.-D. (2020). Generation of Floor Plan Variations with Convolutional Neural Networks and Case-based Reasoning An approach for transformative adaptation of room configurations within a framework for support of early conceptual design phases (pp. 79–84). https://doi.org/10.5151/proceedings-ecaadesigradi2019_648
-Elsayed, M., Pelsmakers, S., … L. P.-B. and, & 2023, undefined. (2023). Post-occupancy evaluation in residential buildings: A systematic literature review of current practices in the EU. Building and Environment,. https://www.sciencedirect.com/science/article/pii/S0360132323003347
-Ezz, M. S., Mahdy, M. A. F., Baharetha, S., Hassanain, M. A., & Gomaa, M. M. (2025). Post occupancy evaluation of architectural design studio facilities. Frontiers in Built Environment, 11. https://doi.org/10.3389/fbuil.2025.1549313
-Gharaei, B., Mohamad Sadegh Hayeri Zadeh, S., & Ghomeishi, M. (2024). Developing a Neuroarchitecture-based User Centered Design for Elementary Schools in Tehran. Ain Shams Engineering Journal, 15(9), 102898. https://doi.org/10.1016/j.asej.2024.102898
-Lizana, J., Serrano-Jimenez, A., Ortiz, C., Becerra, J. A., & Chacartegui, R. (2018). Energy assessment method towards low-carbon energy schools. Energy, 159, 310–326. https://doi.org/10.1016/j.energy.2018.06.147
-Ma, H., Lai, J., Li, C., Yang, F., & Li, Z. (2019). Analysis of school building energy consumption in Tianjin, China. Energy Procedia, 158, 3476–3481. https://doi.org/10.1016/j.egypro.2019.01.924
-Michalek, J. J., Choudhary, R., & Papalambros, P. Y. (2002). Architectural layout design optimization. In Engineering Optimization (Vol. 34, Issue 5, pp. 461–484). https://doi.org/10.1080/03052150214016
-Preiser, W. F. E. (1995). Post-occupancy evaluation: How to make buildings work better. Facilities, 13(11), 19–28. https://doi.org/10.1108/02632779510097787
-Rahbar, M., Mahdavinejad, M., Markazi, A. H. D., & Bemanian, M. (2022). Architectural layout design through deep learning and agent-based modeling: A hybrid approach. Journal of Building Engineering, 47, 103822. https://doi.org/10.1016/j.jobe.2021.103822
-Rizi, R. A., & Eltaweel, A. (2021). A user detective adaptive facade towards improving visual and thermal comfort. Journal of Building Engineering, 33. https://doi.org/10.1016/j.jobe.2020.101554
-Rizvi, S. K. J., Azad, M. A., & Fraz, M. M. (2021). Spectrum of Advancements and Developments in Multidisciplinary Domains for Generative Adversarial Networks (GANs). Archives of Computational Methods in Engineering, 28(7), 4503–4521. https://doi.org/10.1007/s11831-021-09543-4
-Rodrigues, E. (2014). Automated Floor Plan Design: Generation, Simulation, and Optimization.
-Rodrigues, E., Gaspar, A. R., & Gomes, Á. (2014). Improving thermal performance of automatically generated floor plans using a geometric variable sequential optimization procedure. Applied Energy, 132, 200–215. https://doi.org/10.1016/j.apenergy.2014.06.068
-Sanni-Anibire, M. O., Hassanain, M. A., & Al-Hammad, A.-M. (2016). Post-Occupancy Evaluation of Housing Facilities: Overview and Summary of Methods. Journal of Performance of Constructed Facilities, 30(5), 04016009. https://doi.org/10.1061/(asce)cf.1943-5509.0000868
-Shehata, A. M. A., Rashed, E. F., & Elmansoury, A. (2025). Post-Occupancy Evaluation of School Building Performance. European Journal of Architecture and Urban Planning, 4(2), 1–12. https://doi.org/10.24018/ejarch.2025.4.2.53
-Shieh, J.-C., & Wang, T.-Y. (2021). Post-Occupancy Evaluating Space Utilization of Elementary School Libraries. IASL Annual Conference Proceedings. https://doi.org/10.29173/iasl7705
-Turner, E., & on, A. Z. (2014). Floor plan generation and room labeling of indoor environments from laser range data. Ieeexplore.Ieee.OrgE Turner, A Zakhor2014 International Conference on Computer Graphics Theory and, 2014•ieeexplore.Ieee.Org. https://doi.org/10.5220/0004680300220033
-Wu, W., Fu, X. M., Tang, R., Wang, Y., Qi, Y. H., & Liu, L. (2019). Data-driven interior plan generation for residential buildings. ACM Transactions on Graphics, 38(6). https://doi.org/10.1145/3355089.3356556
-Zhang, D., & Bluyssen, P. M. (2021). Energy consumption, self-reported teachers’ actions and children’s perceived indoor environmental quality of nine primary school buildings in the Netherlands. Energy and Buildings, 235. https://doi.org/10.1016/j.enbuild.2021.110735
-Zhao, J., Abdul Aziz, F., Deng, Y., Ujang, N., & Xiao, Y. (2024). A Review of Comprehensive Post-Occupancy Evaluation Feedback on Occupant-Centric Thermal Comfort and Building Energy Efficiency. Buildings, 14(9), 2892. https://doi.org/10.3390/buildings14092892
-Zhou, T., Li, Q., Lu, H., Cheng, Q., & Zhang, X. (2023). GAN review: Models and medical image fusion applications. Information Fusion, 91, 134–148. https://doi.org/10.1016/j.inffus.2022.10.017