Key points are not available for this paper at this time.
This paper presents a computationally efficient two-stage analytical framework for predicting daylight performance in buildings. It is designed to support real-time applications in smart lighting and intelligent building management systems. This approach combines a facade lighting model—driven by solar geometry and atmospheric transmittance—with an interior light distribution module that represents the window as a discretized light source. This formulation provides a lightweight alternative to computationally intensive ray tracing methods. It allows rapid estimation of spatial lighting patterns with minimal input data. The framework is validated using a one-year measurement campaign with class A photometric sensors in three facade orientations. The facade module achieved an average relative error below 15%, while the interior lighting model yielded an RMSE of 83 lx (≈10% error). The integrated system demonstrated an overall average deviation of 18.6% under different sky and season conditions. Owing to its low computational complexity and physically transparent formulation, the proposed method is suitable for deployment in smart building platforms, including daylight-responsive lighting control, embedded energy management systems, and digital twins requiring fast and continuous simulation of daylight availability.
Belány et al. (Sat,) studied this question.