A Computational Framework for the Optimization of
the Environmental Performance of Facades in Early
Design Stages

Mahmoud Islam Abdelhay Gadelhak
Technical University of Munich

In this thesis by Mahmoud Islam Abdelhay Gadelhak a simplified framework and design decision support tool was developed to aid designers and architects in the domain of multidimensional optimization of building facades, where environmental performance measures are enhanced by incrementally improving the façades design using multi-objective optimizations.

The framework facilitates the generation of architectural forms of building envelopes according to different performance aspects such as daylighting, energy efficiency, thermal comfort, visual comfort, energy generation, and life cycle analysis.

It utilizes the flexibility of parametric design and scripting interfaces as front- end input tools (e.g. Python and Grasshopper for Rhinoceros 3D) and simulation software as building performance assessment tools (e.g. Energy Plus, Radiance, and Daysim).

The multi-objective optimization problem can then be solved using evolutionary search techniques such as Pareto Genetic Algorithms. A visualization dashboard is developed to facilitate the visualization and analysis of optimization and simulations results.

Case studies were  used to demonstrate the  framework’s potential and  its usability. The thesis aims to show that this approach is specifically practical for contemporary design practices, as it offers the designers with a decision support tool that is based on a robust method for optimizing performative and constructible free-form architecture.