Machine Learning Spatial Optimization | Modular Lighting Analysis

An experimental architectural visualization exploring how machine learning can assist modular spatial optimization through environmental evaluation. The project analyzes different modular configurations and identifies optimal spatial arrangements based on factors such as daylight access, spatial area, and programmatic potential. Green areas represent spaces with limited natural light, suitable for functions such as exhibition halls, circulation, infrastructure, or public gathering spaces. Red modules indicate optimal positions under the evaluated conditions, balancing lighting performance and spatial efficiency. Rather than producing a single static solution, the system continuously evaluates and filters spatial possibilities through iterative computational processes, allowing architectural organization to emerge from data-driven environmental feedback. This video visualizes the decision-making and selection process within the generative system. Tools: Rhino / Grasshopper / Machine Learning Workflow / Environmental Analysis Project by Baoying Liu