Surrogate-Based Multi-Objective Optimization of Passive Building Envelope Design Under Future Climate Scenarios with Robustness Assessment
Supervisor Name
Sameh Monna
Supervisor Email
samehmona@najah.edu
University
An Najah National University
Research field
Architecture
Bio
Dr. Sameh Monna has a PhD degree of architectural engineering from Polytechnic University of Milan in Italy, with a specialization in built environment, science and technology. Dr. Sameh is specialized in the area of sustainable architecture mainly in environmental design, energy efficiency in buildings, renewable energy use in buildings and green buildings. Dr. Sameh is currently working as director of architecture and civil engineering department at An Najah National University. Dr. Sameh worked for one year at Lawrence Berkeley National Laboratory – University of California Berkeley in USA as a visiting researcher and for one year at EPFL - Federal Institute of Technology in Lausanne - Switzerland as a post-doctoral researcher. He is fluent in English and Italian languages.
Description
This research project aims to develop a surrogate-based multi-objective optimization framework for passive building envelope design that considers future climate uncertainty. The study focuses on improving building energy performance, thermal comfort, and daylight performance while ensuring that optimized design solutions remain robust under multiple future climate scenarios. The building sector represents one of the largest contributors to global energy consumption and greenhouse gas emissions. Passive design strategies such as building orientation, envelope insulation, glazing properties, and shading devices play a crucial role in reducing energy demand and improving indoor environmental quality. However, most building optimization studies rely on historical weather data (TMY), which assumes climate stability and does not adequately reflect projected climate changes over a building’s lifespan. To address this limitation, this research integrates machine-learning surrogate models with multi-objective evolutionary optimization algorithms. Artificial Neural Networks (ANNs) will be trained using datasets generated from building performance simulations to approximate the behavior of detailed simulation tools such as EnergyPlus. The surrogate model will then be coupled with the NSGA-II optimization algorithm to efficiently explore a large design space and identify optimal envelope configurations. The framework will evaluate building performance under current and multiple future climate scenarios derived from global climate models and representative concentration pathways (RCPs). The resulting solutions will be assessed using robustness metrics such as standard deviation across scenarios and minimax regret, ensuring that the selected passive envelope configurations maintain stable performance under climate uncertainty. A representative school building in Palestine will be used as a case study to test the proposed framework. The research aims to contribute to climate-resilient building design methodologies and provide decision-support tools for sustainable architecture in hot and semi-arid regions.
