Comparison Between Popular Genetic Algorithm (GA)-Based Tool and Covariance Matrix Adaptation - Evolutionary Strategy (CMA-ES) for Optimizing Indoor Daylight

To maximize indoor daylight, design projects commonly use commercial optimization tools to determine optimum window configurations. However, experiments show that such tools either grossly suboptimal or are very slow to compute in certain conditions. This paper presents an empirical comparison between a gradient-free optimization technique, Covariance Matrix Adaptation Evolution Strategy (CMA-ES), and the widely used Genetic Algorithm (GA)-based tool, Galapagos, for optimizing window parameters to improve indoor daylight. Results are reported for six locations across different latitudes. A novel combination of daylight metrics, sDA, and ASE, is proposed for single-objective optimization comparison. Results indicate that GA in Galapagos takes progressively more time to converge, from 11 minutes in southernmost to 11 hours in northernmost latitudes, while runtime for CMA-ES is consistently around 2 hours. On average, CMA-ES is 1.5 times faster than Galapagos, while consistently finding optimal solutions. The conclusions from this paper can help researchers in selecting appropriate optimization algorithms for daylight simulation based on latitudes, desired runtime, and desired solution quality.