rsm-thermal-daylight-optimization

This repository provides supplementary materials to reproduce the results outlined in the research article, Response Surface Methodology using desirability functions for multiobjective optimization to minimize indoor overheating hours and maximize useful daylight illuminance, published in Scientific Reports (DOI: https://doi.org/10.1038/s41598-025-96376-x)

All code used in this study is provided as R Markdown and R script files within the repository.

The repository is organized into the following folders:

• rhino-grasshopper-honeybee-files: Contains the .gh script file for running thermal comfort (indoor overheating hours, IOH) and daylight (useful daylight illuminance, UDI) simulations. It also includes: A .3dm file with the 3D model of the dwelling. .csv files that provide input data for different parts of the script (e.g., occupancy schedules, outdoor temperature). A .csv file with the fractional factorial design used to feed the Colibri component for the Screening subsection.
• simulation-inputs: Includes the first-order (orthogonal first-order 2k factorial design) and second-order (Central Composite Design) experimental designs used for simulating IOH and UDI.
• simulation-outputs: If you prefer to skip running the simulations, this folder contains precomputed results that can be directly used with the R code provided in RSM_thermal_daylight_optimization.R and RSM_thermal_daylight_optimization.md.
• randomization: Stores randomized values (following a normal distribution) for factors with negligible effects on Overall Desirability (D), as identified through Lasso and Stepwise Regression.

If you cite this work or repository

Gamero-Salinas, J., López-Fidalgo, J. Response Surface Methodology using desirability functions for multiobjective optimization to minimize indoor overheating hours and maximize useful daylight illuminance. Scientific Reports 15, 12173 (2025). https://doi.org/10.1038/s41598-025-96376-x