SIGraDi 2025: Meta Responsive Approaches
Full paper accepted (2025)
To advance the understanding of airborne disease transmission in indoor environments, researchers have developed models that independently employ well-established Computational Fluid Dynamics (CFD) and Agent-Based Simulation (ABS). CFD models capture the dispersion of airborne particles and the influence of ventilation and airflow patterns, while ABS frameworks can be implemented to simulate individual human behaviors, movement, and interactions within a given scenario. Although typically applied independently, both approaches offer critical insights into the probabilistic assessment of transmission, each representing a unique spread dynamics modality: airborne transmission (smaller particles floating in the air) and direct transmission (particles attached to droplets), respectively. A fully coupled CFD-ABS modeling framework would better address challenges such as a more realistic, high-resolution spatiotemporal models, which are critical for more accurate modeling of airborne and direct disease transmission, which in turn requires a high computational cost and timeframe desynchronization, for real-time feedback in architectural design. Our research provides a two-tiered review of the state-of-the-art at the intersection of CFD and ABS modeling frameworks, with a particular focus on implementation strategies and reported outcomes. The analysis is synthesized in a comparative table that systematically evaluates nineteen studies across a set of parameters, including the CFD-ABS integration platform, software architecture, integration schema, field of application, simulation domain scale, temporal resolution, metrics, and key findings. The conclusion suggests the next steps to advance this area.
Building and Environment
2025
Significant advances in microclimate modeling tools range from no coupling to fully coupled urban physics, requiring a tradeoff between simulation speed and accuracy. This study was conducted to validate the urbanMicroclimateFoam solver for accuracy, as a key step toward improving its reliability for urban microclimate modeling. A detailed 3D model was created, and transient simulations were performed and compared against weather station data from Georgia Tech’s campus. The solver includes heat, air, moisture (HAM), radiation (RAD), and vegetation (VEG) physics, and its performance was evaluated in terms of temperature, humidity, and thermal comfort. Results show strong agreement with measured data, achieving a temperature RMSE as low as 1.03 °C and a humidity RMSE of 4.78 % at the best performing location. Simulations also captured diurnal WBGT trends and vegetation driven cooling up to 2.1 °C. The findings highlight the value of fully coupled CFD models for understanding the urban heat island effect and designing more sustainable, climate resilient cities.
View PublicationSIGraDi 2021: Designing Possibilities
2021
In this paper, we develop a structural pre-evaluation and optimization technique for vault-like shapes. This implementation focuses on exploring design spaces in early design stages. The proposed technique approaches the problem of reducing the flexibility of the design space while advancing to later design stages for vault-like shapes. We start with a custom design space based on design intent. Then, we define a sampling criteria to study a reduced number of candidates. Later, the optimization process focuses on minimizing structural deformation values through shape manipulation. Results show a notorious enhancement for maximum deflection and displacement of the structure. Generally speaking, the shape optimization pattern is consistent with how vault-like shape works. All solutions reduce their span and boundary area while increasing the maximum height. Also, reaching maximum deformation values that are ten times better than the admissible final values on average.
View PublicationSIGraDi 2020: Transformative Design
2020
This study is based on the development of a modeling technique for vault-like structure generation through topological manipulation. Currently, topology-driven form-finding has been implemented in tensile structures, but no further studies have been conducted for compression-only structures. The focus of this study is to approach the problem of highly determined vault shapes by their input topology. The technique operates at the topological level between vertices and edges to create an input 2D topology map. The particle-spring system uses such a map to simulate the resulting 3D mesh geometry. For testing purposes, we explore three generative approaches. The results show the effectiveness of the technique to manipulate the topological relationships that controls the generation of the funicular structures.
View Publication