This dissertation presents a new, optimal window design procedure for an office that uses a combined daylighting and thermal simulation in a hot and dry climate. The purpose of this work is to better inform the design of building windows used for daylighting in the preliminary design stage for improving building performance. This study used a simple office model to develop and test a prototype for the combined daylighting+thermal simulation by comparing the combined simulation methods of DOE-2+Split-Flux, EnergyPlus+Split-Flux, EnergyPlus+Radiosity, and EnergyPlus+Radiance. The results showed that different window size and location designs could have very different annual energy consumption results when using the combined EnergyPlus+Radiance simulation tool for North, South, East, and West orientations. However, the other three combined simulation methods could not simulate the differences between the different window sizes and placement designs (with the same window areas). Therefore, this study proposes guidelines for how to conduct a combined daylighting and thermal simulation to obtain more accurate results. This study demonstrated the use of an improved procedure for using the Radiance simulation for speeding up daylighting optimization. This new method produces accurate annual daylighting results while minimizing run time. This study also proposes a new customized, Radiance rendering parameters (called custom preset) into the DIVA software to simulate the annual daylighting. This custom preset only took 30 seconds to obtain annual daylighting results, while the most accurate preset (high-quality preset) in DIVA takes over one hour to complete the same simulation. The statistical software JMP Pro 14 was used to calculate the correlation between high-quality preset and custom preset. The results show that the high accuracy annual daylighting results can be predicted using the simulation results from the custom preset together with the multi-linear regression method that was developed. This study developed a window design plugin in grasshopper using Python. This new window design plugin was used to generate thousands of different window sizes and placement designs. The window design plugin used a Multi-Objective Optimization (MOO) tool for analyzing different window sizes and placement designs. Finally, four optimization studies were conducted for the case-study office. The results showed that top positioned windows had the best daylighting and thermal performance, whereas lower positioned windows had the worst results. Therefore, national standards, such as ASHRAE Standard 90.1 and the IECC should not give the same credits for all the window location placements on an external wall. The Standard should provide the guidelines for the combined thermal and daylighting simulation. In addition, standardized testing of combined simulation programs that model the daylighting and thermal characteristics of a building, similar to the existing ASHRAE Standard 140 procedures, need to be developed and used by whole-building energy simulation programs.
