From Traditional Passive Design to Artificial Intelligence-Based Optimization: A Systematic Review of Passive Design in Residential Buildings
Authors:
- Juan Serrano-ArellanoJuan Serrano-Arellano
- Yamila Caridad Rodríguez GómezYamila Caridad Rodríguez Gómez
- María Isabel Hernández-LópezMHMaría Isabel Hernández-López
- Francisco Noé Demesa-LópezFDFrancisco Noé Demesa-López
- Griselda Stephany Abarca-JiménezGAGriselda Stephany Abarca-Jiménez
- Karla María Aguilar CastroKCKarla María Aguilar Castro
- Juan Manuel Belman-FloresJuan Manuel Belman-Flores
Open Access
- 9 October 2026
- journal article
- review article
- Published by MDPI AG in Sustainability
- Vol. 18 (20) , 10280
Abstract
Buildings account for a significant share of global energy consumption and greenhouse gas emissions, highlighting the importance of passive design strategies for improving residential energy efficiency and thermal comfort. This systematic review, conducted and reported in accordance with PRISMA 2020, examines the evolution of passive residential design from bioclimatic principles toward building energy simulation, computational optimization, and artificial intelligence (AI). A total of 191 studies were included in the systematic evidence synthesis. The reviewed evidence identifies orientation, insulation, glazing properties, window-to-wall ratio, shading, thermal mass, and ventilation as recurrent design variables influencing residential energy and thermal performance. The literature also reveals a methodological transition from the evaluation of individual passive strategies toward integrated computational frameworks combining building-performance simulation, predictive modeling, and multi-objective optimization. Validation or verification was explicitly reported in 72 of the 191 studies (37.7%), highlighting an important methodological limitation in the transferability of simulation- and AI-derived results. Methodological heterogeneity, climatic uncertainty, occupant behavior, and restricted transferability across climatic and construction contexts also remain important challenges. Overall, the evidence indicates that AI can complement rather than replace the physical principles of passive design by supporting prediction, design-space exploration, and optimization. From an engineering perspective, the findings support a hierarchical workflow in which passive variables guide early design decisions, simulation evaluates their interactions, multi-objective optimization explores performance trade-offs, and AI or surrogate models provide decision support only after adequate validation for the specific climate and building context.Keywords
This publication has 219 references indexed in Scilit:
- The effect of courtyard and atrium on energy performance of buildings in hot and arid climates: a reviewJournal of Umm Al-Qura University for Engineering and Architecture, 2025
- Thermal performance of energy efficient houses in cold areasInternational Journal of Energy and Water Resources, 2025
- Multi-objective optimal energy-efficient retrofit determination using hybrid urban building energy model: Considering uncertainties between modelsBuilding Simulation, 2024
- Proceedings of the 11th International Symposium on Heating, Ventilation and Air Conditioning (ISHVAC 2019)2020
- A state of the art review of PV‐Trombe wall system: Design and applicationsEnvironmental Progress & Sustainable Energy, 2019
- Impact of passive climate adaptation measures and building orientation on the energy demand of a detached lightweight semi-portable buildingBuilding Simulation, 2018
- Citation-based clustering of publications using CitNetExplorer and VOSviewerScientometrics, 2017
- The journal coverage of Web of Science and Scopus: a comparative analysisScientometrics, 2015
- Software survey: VOSviewer, a computer program for bibliometric mappingScientometrics, 2009
- Thermal simulation of a passive solar house using a trombe-michel wall structureSolar Energy, 1978