Sequential Multi-Objective Optimization and Post-optimization Sensitivity Analysis of Light Shelf Variables in Semi-Arid Climates

Global climate action necessitates the optimization of building envelopes during early design to enhance energy efficiency and occupant comfort. Exterior light shelves are a critical passive strategy for improving thermal and visual comfort while simultaneously reducing energy consumption. This study addresses a research gap by integrating sensitivity analysis (SA) and multi-objective optimization (MOO) for light shelf systems in office buildings within Iran's semi-arid continental climate. We systematically investigate the impact of light shelf and window parameters on three key performance metrics: Predicted Mean Vote (PMV) for thermal comfort, Daylight Glare Probability (DGP) for visual comfort, and Energy Use Intensity (EUI) for overall energy performance. Utilizing a robust methodology that employs a multi-objective genetic algorithm (MOGA), the research identifies optimal design solutions by navigating the trade-offs on the Pareto frontier. The key design variables include shading control strategy (SCS), light shelf angle (LSA), length (LSL), height (LSH), viewpoint (VP), visible transmittance (VT), and window-to-wall ratio (WWR). Our findings reveal significant performance improvements: PMV improves by 22%, DGP by 69%, and EUI by 12.6% compared to the baseline model. SA identifies WWR, SCS, and LSA as the most influential parameters, with WWR having a particularly significant effect on glare and energy consumption. The energy simulation is validated against the ASHRAE 140-2020 standard, ensuring the reliability of our results. This research provides a comprehensive framework for designing high-performance façades that prioritize occupant well-being and environmental sustainability.